Umeå University's logo

Magnetic resonance imaging (MRI) utilizes the magnetic properties of tissues to generate image-forming signals. MRI has exquisite soft-tissue contrast and since tumors are mainly soft-tissues, it offers improved delineation of the target volume and nearby organs at risk. The proposed Magnetic Resonance-only Radiotherapy (MR-only RT) work flow allows for the use of MRI as the sole imaging modality in the radiotherapy (RT) treatment planning of cancer. There are, however, issues with geometric distortions inherent with MR image acquisition processes. These distortions result from imperfections in the main magnetic field, nonlinear gradients, as well as field disturbances introduced by the imaged object. In this thesis, we quantified the effect of system related and patient-induced susceptibility geometric distortions on dose distributions for prostate as well as head and neck cancers. Methods to mitigate these distortions were also studied.

In Study I, mean worst system related residual distortions of 3.19, 2.52 and 2.08 mm at bandwidths (BW) of 122, 244 and 488 Hz/pixel up to a radial distance of 25 cm from a 3T PET/MR scanner was measured with a large field of view (FoV) phantom. Subsequently, we estimated maximum shifts of 5.8, 2.9 and 1.5 mm due to patient-induced susceptibility distortions. VMAT-optimized treatment plans initially performed on distorted CT (dCT) images and recalculated on real CT datasets resulted in a dose difference of less than 0.5%.

 The magnetic susceptibility differences at tissue-metallic,-air and -bone interfaces result in local B₀ magnetic field inhomogeneities. The distortion shifts caused by these field inhomogeneities can be reduced by shimming.  Study II aimed to investigate the use of shimming to improve the homogeneity of local  B₀ magnetic field which will be beneficial for radiotherapy applications. A shimming simulation based on spherical harmonics modeling was developed. The spinal cord, an organ at risk is surrounded by bone and in close proximity to the lungs may have high susceptibility differences. In this region, mean pixel shifts caused by local B₀ field inhomogeneities were reduced from 3.47±1.22 mm to 1.35±0.44 mm and 0.99±0.30 mm using first and second order shimming respectively. This was for a bandwidth of 122 Hz/pixel and an in-plane voxel size of 1×1 mm².  Also examined in Study II as in Study I was the dosimetric effect of geometric distortions on 21 Head and Neck cancer treatment plans. The dose difference in D₅₀ at the PTV between distorted CT and real CT plans was less than 1.0%.

In conclusion, the effect of MR geometric distortions on dose plans was small. Generally, we found patient-induced susceptibility distortions were larger compared with residual system distortions at all delineated structures except the external contour. This information will be relevant when setting margins for treatment volumes and organs at risk.  

The current practice of characterizing MR geometric distortions utilizing spatial accuracy phantoms alone may not be enough for an MR-only radiotherapy workflow. Therefore, measures to mitigate patient-induced susceptibility effects in clinical practice such as patient-specific correction algorithms are needed to complement existing distortion reduction methods such as high acquisition bandwidth and shimming.

This book is a compilation of reviews about the pathogenesis of Type 1 Diabetes. T1D is a classic autoimmune disease. Genetic factors are clearly determinant but cannot explain the rapid, even overwhelming expanse of this disease. Understanding etiology and pathogenesis of this disease is essential. A number of experts in the field have covered a range of topics for consideration that are applicable to researcher and clinician alike. This book provides apt descriptions of cutting edge technologies and applications in the ever going search for treatments and cure for diabetes. Areas including T cell development, innate immune responses, imaging of pancreata, potential viral initiators, etc. are considered.

Artificial Intelligence (AI)-based models can help in diagnosing COVID-19 from lung CT scans and X-ray images; however, these models require large amounts of data for training and validation. Many researchers studied Generative Adversarial Networks (GANs) for producing synthetic lung CT scans and X-Ray images to improve the performance of AI-based models. It is not well explored how good GAN-based methods performed to generate reliable synthetic data. This work analyzes 43 published studies that reported GANs for synthetic data generation. Many of these studies suffered data bias, lack of reproducibility, and lack of feedback from the radiologists or other domain experts. A common issue in these studies is the unavailability of the source code, hindering reproducibility. The included studies reported rescaling of the input images to train the existing GANs architecture without providing clinical insights on how the rescaling was motivated. Finally, even though GAN-based methods have the potential for data augmentation and improving the training of AI-based models, these methods fall short in terms of their use in clinical practice. This paper highlights research hotspots in countering the data scarcity problem, identifies various issues as well as potentials, and provides recommendations to guide future research. These recommendations might be useful to improve acceptability for the GAN-based approaches for data augmentation as GANs for data augmentation are increasingly becoming popular in the AI and medical imaging research community.

Objective: In this work, we evaluated a model for the translation of atherosclerotic disease features onto healthy carotid ultrasound images.

Methods: An un-paired domain-to-domain translation model – the cycle Generative Adversarial Network (cycleGAN) – was trained to translate between carotid ultrasound images of healthy arteries and images of pronounced disease. Translation performance was evaluated using the measurement of wall thickness in original and generated images. In addition, we explored disease translation in different tissue segments (subcutaneous tissue, muscle, lumen, far wall, and deep tissues), using structural similarity index measure (SSIM) maps.

Results: Features of pronounced disease were successfully translated to the healthy images (1.2 (0.33) mm vs 0.43 (0.07) mm, p < 0.001), while overall anatomy was retained as SSIM value was equal to 0.78 (0.02). Exploration of translated features showed that both arterial wall and subcutaneous tissues were modified in the translation, but that the subcutaneous tissue was subject to distortion of the anatomy in some cases. The image quality influenced the disease translation performance.

Conclusion: The results show that the model can learn a mapping between healthy and diseased images while retaining the overall anatomical contents. This is the first study on atherosclerosis disease translation in medical images.

Significance: The concept of translating disease onto existing healthy images may serve purposes such as education, cardiovascular risk communication in health conversations, or personalized modelling in precision medicine.

Skeletal muscles are functionally regulated by populations of so-called motor units (MUs). An MU comprises a bundle of muscle fibers controlled by a neuron from the spinal cord. Current methods to diagnose neuromuscular diseases and monitor rehabilitation, and study sports sciences rely on recording and analyzing the bio-electric activity of the MUs. However, these methods provide information from a limited part of a muscle. Ultrasound imaging provides information from a large part of the muscle. It has recently been shown that ultrafast ultrasound imaging can be used to record and analyze the mechanical response of individual MUs using blind source separation. In this work, we present an alternative method - a deep learning pipeline - to identify active MUs in ultrasound image sequences, including segmentation of their territories and signal estimation of their mechanical responses (twitch train). We train and evaluate the model using simulated data mimicking the complex activation pattern of tens of activated MUs with overlapping territories and partially synchronized activation patterns. Using a slow fusion approach (based on 3D CNNs), we transform the spatiotemporal image sequence data to 2D representations and apply a deep neural network architecture for segmentation. Next, we employ a second deep neural network architecture for signal estimation. The results show that the proposed pipeline can effectively identify individual MUs, estimate their territories, and estimate their twitch train signal at low contraction forces. The framework can retain spatio-temporal consistencies and information of the mechanical response of MU activity even when the ultrasound image sequences are transformed into a 2D representation for compatibility with more traditional computer vision and image processing techniques. The proposed pipeline is potentially useful to identify simultaneously active MUs in whole muscles in ultrasound image sequences of voluntary skeletal muscle contractions at low force levels.

Background: Advances in sports medicine, rehabilitation applications and diagnostics of neuromuscular disorders are based on the analysis of skeletal muscle contractions. Recently, medical imaging techniques have transformed the study of muscle contractions, by allowing identifcation of individual motor units’ activity, within the whole studied muscle. However, appropriate image-based simulation models, which would assist the continued development of these new imaging methods are missing. This is mainly due to a lack of models that describe the complex interaction between tissues within a muscle and its surroundings, e.g., muscle fbres, fascia, vasculature, bone, skin, and subcutaneous fat. Herein, we propose a new approach to overcome this limitation.

Methods: In this work, we propose to use deep learning to model the authentic intramuscular skeletal muscle contraction pattern using domain-to-domain translation between in silico (simulated) and in vivo (experimental) image sequences of skeletal muscle contraction dynamics. For this purpose, the 3D cycle generative adversarial network (cycleGAN) models were evaluated on several hyperparameter settings and modifcations. The results show that there were large diferences between the spatial features of in silico and in vivo data, and that a model could be trained to generate authentic spatio-temporal features similar to those obtained from in vivo experimental data. In addition, we used diference maps between input and output of the trained model generator to study the translated characteristics of in vivo data.

Results: This work provides a model to generate authentic intra-muscular skeletal muscle contraction dynamics that could be used to gain further and much needed physiological and pathological insights and assess and overcome limitations within the newly developed research feld of neuromuscular imaging.

In recent studies, it has been found that fat tissue can be used as a microwave communication channel. In this article, the effect of thickness inhomogeneities in fat tissues on the performance of in-body microwave communication at 2.45 GHz is investigated using phantom models. We considered two models namely concave and convex geometrical fat distribution to account for the thickness inhomogeneities. The thickness of the fat tissue is varied from 5 mm to 45 mm and the Gap between the transmitter/receiver and the starting and ending of concavity/convexity is varied from 0 mm to 25 mm for a length of 100 mm to study the behavior in the microwave propagation. The phantoms of different geometries, concave and convex, are used in this work to validate the numerical studies. It was noticed that the convex model exhibited higher signal coupling by an amount of 1 dB (simulation) and 2 dB (measurement) compared to the concave model. From the study, it was observed that the signal transmission improves up to 30 mm thick fat and reaches a plateau when the thickness is increased further.

Imlook4d (http://www.dicom-port.com) is a free Matlab based graphical user interface (GUI) tool useful for static, dynamic and gated PET studies.  It supports reading and writing DICOM, Nifti, Analyze, ECAT.  The DICOM reader is orders of magnitude faster than the Matlab imaging toolbox.  Imlook4d requires no additional Matlab toolboxes.

The main benefit with imlook4d is that it is easily extendable with scripts, accessing exported variables such as the image matrix (4D) and a region-of-interest (ROI) matrix.  Scripts are available via a menu in the imlook4d GUI, and can be used to manipulate the image-matrix and ROI data.  There is also a menu option to export and import these variables to the Matlab workspace for interactive manipulation, useful for one-off fixes or for script development.  There are presently about 30 scripts in categories such as ROI, Matrix, Header info etc.  There is also direct export to ImageJ [1] and import back from ImageJ, thus giving access to all tools available within ImageJ.

Imlook4d has a built in volume-of-interest editor, with a brush tool for quick interactive ROI delineation, and via scripts, different ways of thresholding ROIs from parts of the image.  Time activity data is saved to a tab-delimited text file.

The principal-component (PC) based Hotelling filter is an integrated part of the program, which allows for interactive noise reduction without loss of quantitation [2].  A typical work flow for a dynamic data set is to turn on the filter for ROI delineation, and then there is the choice of turning it off for export of time-activity data.  Also the PC images can be used to draw ROIs on, which under some circumstances gives enhanced contrast.

Calculation of parametric pharmacokinetic modelling images can be performed interactively, calculated slice by slice as the user scrolls through the volume.  Reference models for Patlak, Logan and Averaged Simple Flow Model [3]  applied on 15O-water are implemented, and it is relatively easy to implement other kinetic models.  Similarly, scripts have been developed for regional Patlak and Logan models on ROI data.

[1] Rasband, WS, ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, http://imagej.nih.gov/ij/, 1997-2014

[2] Axelsson J, Sörensen J, The 2D Hotelling filter - a quantitative noise-reducing principal-component filter for dynamic PET data, with applications in patient dose reduction. BMC Med Phys. 2013 Apr 10;13:1. doi: 10.1186/1756-6649-13-1.

[3] Yoshida, K, Mullani, N and Gould KL, Coronary Flow and Flow Reserve by PET Simplified for Clinical Applications Using Rubidium-82 or Nitrogen-13-Ammonia, J Nucl Med 1996; 37:1701-1712

Figure 1.  The imlook4d GUI with the user SCRIPTS menu selected.  The group of ROI scripts was further selected.  In the underlying image, a rough ROI is created.  

Background: In this paper we apply the principal-component analysis filter (Hotelling filter) to reduce noise fromdynamic positron-emission tomography (PET) patient data, for a number of different radio-tracer molecules. Wefurthermore show how preprocessing images with this filter improves parametric images created from suchdynamic sequence.We use zero-mean unit variance normalization, prior to performing a Hotelling filter on the slices of a dynamictime-series. The Scree-plot technique was used to determine which principal components to be rejected in thefilter process. This filter was applied to [11C]-acetate on heart and head-neck tumors, [18F]-FDG on liver tumors andbrain, and [11C]-Raclopride on brain. Simulations of blood and tissue regions with noise properties matched to realPET data, was used to analyze how quantitation and resolution is affected by the Hotelling filter. Summing varyingparts of a 90-frame [18F]-FDG brain scan, we created 9-frame dynamic scans with image statistics comparable to 20MBq, 60 MBq and 200 MBq injected activity. Hotelling filter performed on slices (2D) and on volumes (3D) werecompared.Results: The 2D Hotelling filter reduces noise in the tissue uptake drastically, so that it becomes simple to manuallypick out regions-of-interest from noisy data. 2D Hotelling filter introduces less bias than 3D Hotelling filter in focalRaclopride uptake. Simulations show that the Hotelling filter is sensitive to typical blood peak in PET prior to tissueuptake have commenced, introducing a negative bias in early tissue uptake. Quantitation on real dynamic data isreliable. Two examples clearly show that pre-filtering the dynamic sequence with the Hotelling filter prior toPatlak-slope calculations gives clearly improved parametric image quality. We also show that a dramatic dosereduction can be achieved for Patlak slope images without changing image quality or quantitation.Conclusions: The 2D Hotelling-filtering of dynamic PET data is a computer-efficient method that gives visuallyimproved differentiation of different tissues, which we have observed improve manual or automated regionof-interest delineation of dynamic data. Parametric Patlak images on Hotelling-filtered data display improved clarity,compared to non-filtered Patlak slope images without measurable loss of quantitation, and allow a dramaticdecrease in patient injected dose.

Purpose: There is increasing interest in computed tomography (CT) image estimations from magneticresonance (MR) images. The estimated CT images can be utilized for attenuation correction, patientpositioning, and dose planning in diagnostic and radiotherapy workflows. This study aims to introducea novel statistical learning approach for improving CT estimation from MR images and to compare theperformance of our method with the existing model-based CT image estimation methods.

Methods: The statistical learning approach proposed here consists of two stages. At the trainingstage, prior knowledge about tissue types from CT images was used together with a Gaussian mixturemodel (GMM) to explore CT image estimations from MR images. Since the prior knowledge is notavailable at the prediction stage, a classifier based on RUSBoost algorithm was trained to estimatethe tissue types from MR images. For a new patient, the trained classifier and GMMs were used topredict CT image from MR images. The classifier and GMMs were validated by using voxel-leveltenfold cross-validation and patient-level leave-one-out cross-validation, respectively.

Results: The proposed approach has outperformance in CT estimation quality in comparison withthe existing model-based methods, especially on bone tissues. Our method improved CT image estimationby 5% and 23% on the whole brain and bone tissues, respectively.

Conclusions: Evaluation of our method shows that it is a promising method to generate CTimage substitutes for the implementation of fully MR-based radiotherapy and PET/MRI applications

There is a growing interest to get a fully MR based radiotherapy. The most important development needed is to obtain improved bone tissue estimation. Existing model-based methods have performed poorly on bone tissues. This paper aims to obtainimproved estimation of bone tissues. Skew-Gaussian mixture model (SGMM) isproposed to further investigate CT image estimation from MR images. The estimation quality of the proposed model is evaluated using leave-one-out cross-validation method on real data. In comparison with the existing model-based approaches, the approach utilized in this paper outperforms in estimation of bone tissues, especiallyon dense bone tissues.

Statistical methods are required to estimate computed tomography (CT) images from magnetic resonance (MR) images. The main purpose of estimating CT images was to get a fully MR based radiotherapy. Specifically, bone tissues and air are indistinguishable on MR images. But, there is a good contrast between soft tissue and other tissues on MR images. On CT images, there is eyecatching contrast between bone and non-bone tissues. Therefore, the main reason for CT estimation is to get improved bone tissues estimation and to use the estimated CT in fully MR based radiotherapy. The estimated CT images (also called substitute CT or Pseudo-CT images) are used for attenuation correction and dose planning in MR based radiotherapy. Gaussian mixture model (GMM) is used to investigate CT image estimation from MR images without taking spatial information into account. Markov random field (MRF) and hidden Markov model (HMM) are used to extend the approach by taking spatial dependence into account. Leave-one-dataset-out cross-validation method on five datasets (obtained from head of five patients) is used to evaluate the performance of the models. In terms of MAE, the use of spatial information improves the overall quality of CT image estimation. In this application, HMM is computationally faster and has superior performance on MRF. However, it has poor performance on bone tissues. On the other hand, MRF is computationally expensive and intractable for log-likelihood based model diagnostic. These two behaviour of HMM and MRF motivated this work to further probe the estimation of CT images from MR images by partitioning the data into bone and non-bone tissues. The partitioning of the data was based on CT value threshold. Skew-Gaussian mixture model (SGMM) and GMM applied on each partition. In terms of MAE, SGMM and GMM* (GMM applied to each partition) performed better than HMM and MRF on the bone tissues.

The development of compressive sensing in recent years has given much attention to sparse signal recovery. In sparse signal recovery, spike and slab priors are playing a key role in inducing sparsity. The use of such priors, however, results in non-convex and mixed integer programming problems. Most of the existing algorithms to solve non-convex and mixed integer programming problems involve either simplifying assumptions, relaxations or high computational expenses. In this paper, we propose a new adaptive alternating direction method of multipliers (AADMM) algorithm to directly solve the suggested non-convex and mixed integer programming problem. The algorithm is based on the one-to-one mapping property of the support and non-zero element of the signal. At each step of the algorithm, we update the support by either adding an index to it or removing an index from it and use the alternating direction method of multipliers to recover the signal corresponding to the updated support. Moreover, as opposed to the competing “adaptive sparsity matching pursuit” and “alternating direction method of multipliers” methods our algorithm can solve non-convex problems directly. Experiments on synthetic data and real-world images demonstrated that the proposed AADMM algorithm provides superior performance and is computationally cheaper than the recently developed iterative convex refinement (ICR) and adaptive matching pursuit (AMP) algorithms.

Rationale: Electrical impedance tomography (EIT) allows instantaneous and continuous visualization of regional ventilation and changes in end-expiratory lung volume at the bedside. There is particular interest in using EIT for monitoring in critically ill neonates and young children with respiratory failure. Previous studies have focused only on short-term monitoring in small populations. The feasibility and safety of prolonged monitoring with EIT in neonates and young children have not been demonstrated yet. Objectives: To evaluate the feasibility and safety of long-term EIT monitoring in a routine clinical setting and to describe changes in ventilation distribution and homogeneity over time and with positioning in a multicenter cohort of neonates and young children with respiratory failure. Methods: At four European University hospitals, we conducted an observational study (NCT02962505) on 200 patients with postmenstrual ages (PMA) between 25 weeks and 36 months, at risk for or suffering from respiratory failure. Continuous EIT data were obtained using a novel textile 32-electrode interface and recorded at 48 images/s for up to 72 hours. Clinicians were blinded to EIT images during the recording. EIT parameters and the effects of body position on ventilation distribution were analyzed offline. Results: The average duration of EIT measurements was 53 ± 20 hours. Skin contact impedance was sufficient to allow image reconstruction for valid ventilation analysis during a median of 92% (interquartile range, 77-98%) of examination time. EIT examinations were well tolerated, with minor skin irritations (temporary redness or imprint) occurring in 10% of patients and no moderate or severe adverse events. Higher ventilation amplitude was found in the dorsal and right lung areas when compared with the ventral and left regions, respectively. Prone positioning resulted in an increase in the ventilation-related EIT signal in the dorsal hemithorax, indicating increased ventilation of the dorsal lung areas. Lateral positioning led to a redistribution of ventilation toward the dependent lung in preterm infants and to the nondependent lung in patients with PMA > 37 weeks. Conclusions: EIT allows continuous long-term monitoring of regional lung function in neonates and young children for up to 72 hours with minimal adverse effects. Our study confirmed the presence of posture-dependent changes in ventilation distribution and their dependency on PMA in a large patient cohort. Clinical trial registered with www.clinicaltrials.gov (NCT02962505).

Skeletal muscle fascia is a connective tissue which provides structure and aids with force transfer in a muscle. Currently there are no good ways of detecting and analyzing micrometer thick structures of this tissue in-vivo. In this thesis, we created a model to detect skeletal muscle fascia, and tested its performance using simulated data. Utilizing the ultrasound simulation software Vantage, which operates through MATLAB, we created a simulation model which replicates the properties and behaviour of skeletal muscle fascia. To detect the tissue, we changed and adapted a previously implemented model of ultrasound localization microscopy (ULM), previously only used to create super resolutionimages of blood vessels. Finally we evaluated the models ability to locate and determine the thickness of the simulated fascia. Additionally we tested the models ability to separate adjacent objects.

We found that our model was successful at detecting and localizing the simulated fascia, with a sub wavelength accuracy. The precision of the located fascia appears more accurate for horizontally aligned objects compared to the vertically aligned ones. The results from determining the thickness of the fascia proved relatively successful as well. However the results showed a high variance. This could be improved through an inclusion of stocasticity in the simulation model we developed. Finally the ability to distinguish two objects close to eachother showed successful results as well. The method was able to clearly detect a fascia circle with a 0.5mm diameter. It was unable to detect the sides a fascia circle with a 0.25mm diameter.

The main limitation with the model we have developed lies in the simulations performed. The simulation model we used was very basic, meaning that it did not perfectly represent the skeletal muscle fascia we sought to examine. Further development of the simulation model is required to provide a result which is more representative of real skeletal muscle fascia.

The analysis of this first model shows promise in detecting the simplified fascia provided by our simulation model. At this stage, the method will require more extensive testing, together with a more thorough statistical analysis, before we can state the usefulness of the method.

During resting state the brain exhibits synchronized activity within all major brain networks. Using blood oxygen level dependent (BOLD) resting state functional magnetic resonance imaging (fMRI) based detection it is possible to quantify the degree of correlation, connectivity, between regions of interest and assess information regarding the integrity of the inter-regional functional integration. A newly available multiband echo planar imaging (EPI) fMRI sequence allows for faster scan times which possibly allows us to better examine large-scale networks and increase the understanding of brain function/dysfunction. This thesis will assess how the newly developed sequence compares to a conventional EPI sequence for detecting resting state connectivity of canonical brain networks. The data acquisitions were made on a 3 Tesla scanner using a 32 channel head coil. The hypothesis was that the multiband sequence would produce a better result since it has faster sampling rate, thus more data points in its time-series to support the statistical analyses.

Using Pearson’s linear correlation between the average time-series (approximately 12 minutes long) within a seed-region and all voxels contained in the image volume, correlation maps where created for each of the eight participants using data normalized to Montreal Neurological Institute (MNI) space. The resting state networks (RSN) were then found by performing a one sample T-test on group level. Six seed-coordinates, based on literature, where used revealing the the homotopic connections in anterior Hippocampus, Motor cortex, Dorsal attention, Visual and the Default mode network (DMN) as well for an anterior-posterior connection in the DMN.

By comparing the maximum T-values within the regions for the RSN no systematic difference could be found between the multiband and conventional fMRI data. Further tests were conducted to evaluate if the sequences would differentiate in their results if the acquisition time was shortened, i.e shortening the time-series in the voxels. However no such difference could be established.Importantly, the results are specific to the 32 channel head coil used in the current study. Presumably recently available and improved coil designs could better exploit the multiband technique.

Assessment of image analysis methods and computer software used in Tc-99m-MAG3 dynamic renography is important to ensure reliable study results and ultimately the best possible care for patients. In this work, we present a national multicentre study of the quantification accuracy in Tc-99m-MAG3 renography, utilizing virtual dynamic scintigraphic data obtained by Monte Carlo-simulated scintillation camera imaging of digital phantoms with time-varying activity distributions. Three digital phantom studies were distributed to the participating departments, and quantitative evaluation was performed with standard clinical software according to local routines. The differential renal function (DRF) and time to maximum renal activity (T-max) were reported by 21 of the 28 Swedish departments performing Tc-99m-MAG3 studies as of 2012. The reported DRF estimates showed a significantly lower precision for the phantom with impaired renal uptake than for the phantom with normal uptake. The T-max estimates showed a similar trend, but the difference was only significant for the right kidney. There was a significant bias in the measured DRF for all phantoms caused by different positions of the left and right kidney in the anterior-posterior direction. In conclusion, this study shows that virtual scintigraphic studies are applicable for quality assurance and that there is a considerable uncertainty associated with standard quantitative parameters in dynamic Tc-99m-MAG3 renography, especially for patients with impaired renal function.

Purpose: The purpose of this study was to investigate, using simulations, a method for improved contrast agent (CA) quantification in DCE-MRI.

Methods: We developed a maximum likelihood estimator that combines the phase signal in the DCE-MRI image series with an additional CA estimate, e.g. the estimate obtained from magnitude data. A number of simulations were performed to investigate the ability of the estimator to reduce bias and noise in CA estimates. Noise levels ranging from that of a body coil to that of a dedicated head coil were investigated at both 1.5T and 3T.

Results: Using the proposed method, the root mean squared error in the bolus peak was reduced from 2.24 to 0.11 mM in the vessels and 0.16 to 0.08 mM in the tumor rim for a noise level equivalent of a 12-channel head coil at 3T. No improvements were seen for tissues with small CA uptake, such as white matter.

Conclusion: Phase information reduces errors in the estimated CA concentrations. A larger phase response from higher field strengths or higher CA concentrations yielded better results. Issues such as background phase drift need to be addressed before this method can be applied in vivo.

Purpose: The weighted CT dose index (CTDIw) has been extended for a nominal total collimation width (nT) greater than 40 mm and relies on measurements of CTDfree air. The purpose of this work was to compare three methods of measuring CTDIfree air and subsequent calculations of CTDIw to investigate their clinical appropriateness.

Methods: The CTDIfree air, for multiple nTs up to 160 mm, was calculated from (1) high-resolution air kerma profiles from a step-and-shoot translation of a liquid ionization chamber (LIC) (considered to be a dosimetric reference), (2) pencil ionization chamber (PIC) measurements at multiple contiguous positions, and (3) air kerma profiles obtained through the continuous translation of a solid-state detector. The resulting CTDIfree air was used to calculate the CTDIw, per the extended formalism, and compared.

Results: The LIC indicated that a 40 mm nT should not be excluded from the extension of the CTDIw formalism. The solid-state detector differed by as much as 8% compared to the LIC. The PIC was the most straightforward method and gave equivalent results to the LIC.

Conclusions: The CTDIw calculated with the latest CTDI formalism will differ most for 160 mm nTs (e.g., whole-organ perfusion or coronary CT angiography) compared to the previous CTDI formalism. Inaccuracies in the measurement of CTDIfree air will subsequently manifest themselves as erroneous calculations of the CTDIw, for nTs greater than 40 mm, with the latest CTDI formalism. The PIC was found to be the most clinically feasible method and was validated against the LIC.

The precision of digital vs. manual radiostereometric measurements in total hip arthroplasty was evaluated using repeated stereoradiographic exposures with an interval of 10–15 min. Ten Lubinus SP2 stems cemented into bone specimens and 12 patients with the same stem design were used to evaluate the precision of stem translations and rotations. The precision of translations and rotations of the cup and femoral head penetration was studied in 12 patients with whole polyethylene cups.

The use of a measurement method based on digitised radiographs improved the precision for some of the motion parameters, whereas many of them did not change. A corresponding pattern was observed for both the intra- and interobserver error. Of the wear parameters, the most pronounced improvements were the 3D wear and in the proximal-distal direction, although the anterior-posterior precision was also improved. The mean errors of rigid body and elliptic fitting decreased in all evaluations but one, consistent with a more reproducible identification of the markers centres and the edge of the femoral head.

Increased precision of radiostereometric measurements may be used to increase the statistical power of future randomised studies and to study new fields in orthopaedics requiring higher precision than has been available with RSA based on manual measurements.

Objective: The objective of this study was to compare the accuracy of calculating the primary tumor volumes using a gradient-based method and fixed threshold methods on the standardized uptake value (SUV) maps and the net influx of FDG (Ki) maps from positron emission tomography-computed tomography (PET-CT) images. Materials and Methods: Newly diagnosed patients with head and neck cancer were recruited, and dynamic PET-CT scan and T2-weighted magnetic resonance imaging were performed. The maps of Ki and SUV were calculated from PET-CT images. The tumor volumes were calculated using a gradient-based method and a fixed threshold method at 40% of maximal SUV or maximal Ki. Four kinds of volumes, VOLKi-Gra (from the Ki maps using the gradient-based method), VOLKi-40% (from the Ki maps using the threshold of 40% maximal Ki), VOLSUV-Gra (from the SUV maps using the gradient-based method), and VOLSUV-40% (from the SUV maps using the threshold of 40% maximal SUV), were acquired and compared with VOLMRI (the volumes acquired on T2-weighted images) using the Pearson correlation, paired t test, and similarity analysis. Results: Eighteen patients were studied, of which 4 had poorly defined tumors (PDT). The positron emission tomography-derived volumes were as follows: VOLSUV-40%, 2.1 to 41.2 cm(3) (mean [SD], 12.3 [10.6]); VOLSUV-Gra, 2.2 to 28.1 cm(3) (mean [SD], 13.2 [8.4]); VOLKi-Gra, 2.4 to 17.0 cm(3) (mean [SD], 9.5 [4.6]); and VOLKi-40%, 2.7 to 20.3 cm(3) (mean [SD], 12.0 [6.0]). The VOLMRI ranged from 2.9 to 18.1 cm(3) (mean [SD], 9.1 [3.9]). The VOLKi-Gra significantly correlated with VOLMRI with the highest correlation coefficient (PDT included, R = 0.673, P = 0.002; PDT excluded, R = 0.841, P < 0.001) and presented no difference from VOLMRI (P = 0.672 or 0.561, respectively, PDT included and excluded). The difference between VOLKi-Gra and VOLMRI was also the smallest. Conclusions: The tumor volumes delineated on the Ki maps using the gradient-based method are more accurate than those on the SUV maps and using the fixed threshold methods.

Dental radiography is one of the most frequent types of diagnostic radiological investigations performed. The equipment and techniques used are constantly evolving. However, dental healthcare has long been an area neglected by radiation safety legislation and the medical physicist community, and thus, the quality assurance (QA) regime needs an update. This project aimed to implement and evaluate objective tests of key image quality parameters for intra-oral (IO) X-ray images.

The image quality parameters assessed were sensitivity, noise, uniformity, low-contrast resolution, and spatial resolution. These parameters were evaluated for repeatability at typical tube current, voltage, and exposure time settings by computing the coefficient of variation (CV) of the mean value of each parameter from multiple images. A further aim was to develop a semi-quantitative test for the correct alignment of the position indicating device (PID) with the primary collimator. The overall purpose of this thesis was to look at ways to improve the QA of IO X-rays systems by digitizing and automating part of the process. A single image receptor and an X-ray tube were used in this study. Incident doses at the receptor were measured using a radiation meter. The relationship between incident dose at the receptor and the output signal was used to determine the signal transfer curve for the receptor. The principal sources of noise in the practical exposure range of the system were investigated using a separation of noise sources based upon variance.

The transfer curve of the receptor was found to be linear. Noise separation showed that quantum noise was the dominant noise. Repeatability of the image quality parameters assessed was found to be acceptable. The CV for sensitivity was less than 3%, while that for noise was less than 1%. For the uniformity measured at the center, the CV was less than 10%, while the CV was less than 5% for the uniformity measured at the edge. The low-contrast resolution varied the most at all exposure settings investigated with CV between 6 - 13%. Finally, the CV for the spatial resolution parameters was less than 5%. The method described to test for the correct alignment of the PID with the primary collimator was found to be practical and easy to interpret manually. The tests described here were implemented for a specific sensor and X-ray tube combination, but the methods could easily be adapted for different systems by simply adjusting certain parameters. 

Background: Accelerated 4D flow MRI allows for high‐resolution velocity measurements with whole‐brain coverage. Such scans are increasingly used to calculate flow rates of individual arteries in the vascular tree, but detailed information about the accuracy and precision in relation to different postprocessing options is lacking.

Purpose: To evaluate and optimize three proposed segmentation methods and determine the accuracy of in vivo 4D flow MRI blood flow rate assessments in major cerebral arteries, with high‐resolution 2D PCMRI as a reference.

Study Type: Prospective.

Subjects: Thirty‐five subjects (20 women, 79 ± 5 years, range 70–91 years).

Field Strength/Sequence: 4D flow MRI with PC‐VIPR and 2D PCMRI acquired with a 3 T scanner.

Assessment: We compared blood flow rates measured with 4D flow MRI, to the reference, in nine main cerebral arteries. Lumen segmentation in the 4D flow MRI was performed with k‐means clustering using four different input datasets, and with two types of thresholding methods. The threshold was defined as a percentage of the maximum intensity value in the complex difference image. Local and global thresholding approaches were used, with evaluated thresholds from 6–26%.

Statistical Tests: Paired t‐test, F‐test, linear correlation (P < 0.05 was considered significant) along with intraclass correlation (ICC).

Results: With the thresholding methods, the lowest average flow difference was obtained for 20% local (0.02 ± 15.0 ml/min, ICC = 0.97, n = 310) or 10% global (0.08 ± 17.3 ml/min, ICC = 0.97, n = 310) thresholding with a significant lower standard deviation for local (F‐test, P = 0.01). For all clustering methods, we found a large systematic underestimation of flow compared with 2D PCMRI (16.1–22.3 ml/min).

Data Conclusion: A locally adapted threshold value gives a more stable result compared with a globally fixed threshold. 4D flow with the proposed segmentation method has the potential to become a useful reliable clinical tool for assessment of blood flow in the major cerebral arteries.

Level of Evidence: 2

Technical Efficacy: Stage 2

Objective: With a 10-minute 4D flow MRI scan, the distribution of blood flow to individual arteries throughout the brain can be analyzed. This technique has potential to become a biomarker for treatment decisions, and to predict prognosis after stroke. To efficiently analyze and model the large dataset in clinical practice, automatization is needed. We hypothesized that identification of selected arterial regions using an atlas with a priori probability information on their spatial distribution can provide standardized measurements of blood flow in the main cerebral arteries.

Approach: A new method for automatic placement of measurement locations in 4D flow MRI was developed based on an existing atlas-based method for arterial labeling, by defining specific regions of interest within the corresponding arterial atlas. The suggested method was evaluated on 38 subjects with carotid artery stenosis, by comparing measurements of blood flow rate at automatically selected locations to reference measurements at manually selected locations.

Main results: Automatic and reference measurement ranged from 10 to 580 ml min⁻¹ and were highly correlated (r = 0.99) with a mean flow difference of 0.61 ± 10.7 ml min⁻¹ (p = 0.21). Out of the 559 arterial segments in the manual reference, 489 were correctly labeled, yielding a sensitivity of 88%, a specificity of 85%, and a labeling accuracy of 87%.

Significance: This study confirms that atlas-based labeling of 4D flow MRI data is suitable for efficient flow quantification in the major cerebral arteries. The suggested method improves the feasibility of analyzing cerebral 4D flow data, and fills a gap necessary for implementation in clinical use.

Gliomas are a common type of brain tumour and for the treatment of a patient it is important to determine the tumour’s grade of malignancy. This is done today by a biopsy, a histopathological analysis of the tumourous tissue, that is classified by the World Health Organization on a malignancy scale from I to IV. Recent studies have shown that the local image variance (LIV) and the intratumoural susceptibility signal (ITSS) in susceptibility-weighted MR images correlate to the tumour grade. This thesis project aims to develop a software program as aid for the radiologists when grading a glioma. The software should by image analysis be able to separate the gliomas into low grade (I-II) and high grade (III-IV). The result is a graphical user interface written in Python 3.4.3. The user chooses an image, draws a region of interest and starts the analysis. The analyses implemented in the program are LIV and ITSS mentioned above, and the code can be extended to contain other types of analyses as research progresses. To validate the image analysis, 16 patients with glioma grades confirmed by biopsy are included in the study. Their susceptibility-weighted MR images were analysed with respect to LIV and ITSS, and the outcome of those image analyses was tested versus the known grades of the patients. No statistically significant difference could be seen between the high and the low grade group, in the case of LIV. This was probably due to hemorrhage and calcification, characteristic for some tumours and interpreted as blood vessels. Concerning ITSS a statistically significant difference could be seen between the high and the low grade group (p < 0.02). The sensitivity and specificity was 80% and 100% respec- tively. Among these 16 gliomas, 11 were astrocytic tumours and between low and high grade astrocytomas a statistically significant difference was shown. The degree of LIV was significantly different between the two groups (p < 0.03) and the sensitivity and specificity were 86% and 100% respectively. The degree of ITSS was significantly different between the two groups (p < 0.04) and the sensitivity and specificity were 86% and 100% respectively. Spearman correlation showed a correlation between LIV and tumour grade (for all gliomas r = 0.53 and p < 0.04, for astrocytomas r = 0.84 and p < 0.01). A correlation was also found between ITSS and tumour grade (for all gliomas r = 0.69 and p < 0.01, for astrocytomas r = 0.63 and p < 0.04). The results indicate that SWI is useful for distinguishing between high and low grade astrocytoma with 1.5T imaging within this cohort. It also seems possible to distinguish between high and low grade glioma with ITSS.

Light microscopy combined with well-established protocols of two-dimensional cell culture facilitates high-throughput quantitative imaging to study biological phenomena. Accurate segmentation of individual cells in images enables exploration of complex biological questions, but can require sophisticated imaging processing pipelines in cases of low contrast and high object density. Deep learning-based methods are considered state-of-the-art for image segmentation but typically require vast amounts of annotated data, for which there is no suitable resource available in the field of label-free cellular imaging. Here, we present LIVECell, a large, high-quality, manually annotated and expert-validated dataset of phase-contrast images, consisting of over 1.6 million cells from a diverse set of cell morphologies and culture densities. To further demonstrate its use, we train convolutional neural network-based models using LIVECell and evaluate model segmentation accuracy with a proposed a suite of benchmarks.

Background: Dynamic contrast-enhanced MRI (DCE-MRI) has the potential to produce images of physiological quantities such as blood flow, blood vessel volume fraction, and blood vessel permeability. Such information is highly valuable, e.g., in oncology. The focus of this work was to improve the quantitative aspects of DCE-MRI in terms of better understanding of error sources and their effect on estimated physiological quantities.

Methods: Firstly, a novel parameter estimation algorithm was developed to overcome a problem with sensitivity to the initial guess in parameter estimation with a specific pharmacokinetic model. Secondly, the accuracy of the arterial input function (AIF), i.e., the estimated arterial blood contrast agent concentration, was evaluated in a phantom environment for a standard magnitude-based AIF method commonly used in vivo. The accuracy was also evaluated in vivo for a phase-based method that has previously shown very promising results in phantoms and in animal studies. Finally, a method was developed for estimation of uncertainties in the estimated physiological quantities.

Results: The new parameter estimation algorithm enabled significantly faster parameter estimation, thus making it more feasible to obtain blood flow and permeability maps from a DCE-MRI study. The evaluation of the AIF measurements revealed that inflow effects and non-ideal radiofrequency spoiling seriously degrade magnitude-based AIFs and that proper slice placement and improved signal models can reduce this effect. It was also shown that phase-based AIFs can be a feasible alternative provided that the observed difficulties in quantifying low concentrations can be resolved. The uncertainty estimation method was able to accurately quantify how a variety of different errors propagate to uncertainty in the estimated physiological quantities.

Conclusion: This work contributes to a better understanding of parameter estimation and AIF quantification in DCE-MRI. The proposed uncertainty estimation method can be used to efficiently calculate uncertainties in the parametric maps obtained in DCE-MRI.

Using dynamic contrast-enhanced MRI (DCE-MRI), it is possible to estimate pharmacokinetic (PK) parameters that convey information about physiological properties, e.g., in tumors. In DCE-MRI, errors propagate in a nontrivial way to the PK parameters. We propose a method based on multivariate linear error propagation to calculate uncertainty maps for the PK parameters. Uncertainties in the PK parameters were investigated for the modified Kety model. The method was evaluated with Monte Carlo simulations and exemplified with in vivo brain tumor data. PK parameter uncertainties due to noise in dynamic data were accurately estimated. Noise with standard deviation up to 15% in the baseline signal and the baseline T1 map gave estimated uncertainties in good agreement with the Monte Carlo simulations. Good agreement was also found for up to 15% errors in the arterial input function amplitude. The method was less accurate for errors in the bolus arrival time with disagreements of 23%, 32%, and 29% for Ktrans, ve, and vp, respectively, when the standard deviation of the bolus arrival time error was 5.3 s. In conclusion, the proposed method provides efficient means for calculation of uncertainty maps, and it was applicable to a wide range of sources of uncertainty.

Dynamic contrast-enhanced magnetic resonance imaging (MRI) is a promising tool in the evaluation of tumor physiology. From rapidly acquired images and a model for contrast agent pharmacokinetics, physiological parameters are derived. One pharmacokinetic model, the tissue homogeneity model, enables estimation of both blood flow and vessel permeability together with parameters that describe blood volume and extracellular extravascular volume fraction. However, studies have shown that parameter estimation with this model is unstable. Therefore, several initial guesses are needed for accurate estimates, which makes the estimation slow. In this study a new estimation algorithm for the tissue homogeneity model, based on Fourier domain calculations, was derived and implemented as a Matlab program. The algorithm was tested with Monte-Carlo simulations and the results were compared to an existing method that uses the adiabatic approximation. The algorithm was also tested on data from a metastasis in the brain. The comparison showed that the new algorithm gave more accurate results on the 2.5th and 97.5th percentile levels, for instance the error in blood volume was reduced by 21%. In addition, the time needed for the computations was reduced with a factor 25. It was concluded that the new algorithm can be used to speed up parameter estimation while accuracy can be gained at the same time.

Introduction. The current age structure in Sweden is changing. At the same time, life expectancy is increasing in many parts of the world. This will lead to greater demands of accessible care resources and its application. Well implemented communication is paramount when the healthcare system is as information-intensive as it is. For maintenance of good healthcare, demands will be set on new kinds of services. With eHealth, information and healthpromotion through communication-based technology, healthcare can be performed efficiently with secure output within the healthcare sector. Aim. The aim of this study was to examine challanges for health professionals in the process of patient care through care at a distance. Method. An empirical study of six (6) people was conducted by semistructured interviews. Result. Indicated result show a result of the need of eHealth within the current use of healthcare, and that the technology can be of comparable quality to physical meetings. However, technical solutions should be seen as a complement within healthcare, and not as a substitute of physical interaction. Conclusion. eHealth services is not something that will become relevant within a timeperiod of ten to fifteen years. It is a currently usable service that can be established to be of comparable quality to the person-centered care the nursing profession is striving for.

Neutrophil extracellular trap (NET) formation is an alternate immunologic weapon used mainly by neutrophils. Chromatin backbones fused with proteins derived from granules are shot like projectiles onto foreign invaders. It is thought that this mechanism is highly anti-microbial, aids in preventing bacterial dissemination, is used to break down structures several sizes larger than neutrophils themselves, and may have several more uses yet unknown. NETs have been implied to be involved in a wide array of systemic host immune defenses, including sepsis, autoimmune diseases, and cancer. Existing methods used to visually quantify NETotic versus non-NETotic shapes are extremely time-consuming and subject to user bias. These limitations are obstacles to developing NETs as prognostic biomarkers and therapeutic targets. We propose an automated pipeline for quantitatively detecting neutrophil and NET shapes captured using a flow cytometry-imaging system. Our method uses contrast limited adaptive histogram equalization to improve signal intensity in dimly illuminated NETs. From the contrast improved image, fixed value thresholding is applied to convert the image to binary. Feature extraction is performed on the resulting binary image, by calculating region properties of the resulting foreground structures. Classification of the resulting features is performed using Support Vector Machine. Our method classifies NETs from neutrophils without traps at 0.97/0.96 sensitivity/specificity on n = 387 images, and is 1500X faster than manual classification, per sample. Our method can be extended to rapidly analyze whole-slide immunofluorescence tissue images for NET classification, and has potential to streamline the quantification of NETs for patients with diseases associated with cancer and autoimmunity.

The E/e’ ratio is commonly used in Doppler echocardiographic examinations to estimate the pulmonary capillary wedge pressure. The rationale of using this ratio is to combine left ventricular (LV) filling (E) and relaxation (e’) velocities to indirectly assess left atrial pressure. However, the accuracy of this index has recently been questioned, particularly in patients with controlled heart failure. Likewise, the potential beat-to-beat variability of such measurements remains undetermined. The cardiovascular system is subject to several oscillations with the potential of influencing LV function and its intra-cavitary pressures, hence measurements of its filling and relaxation velocities. The aim of this pilot study was to assess the beat-to-beat variability of the E/e’ ratio in one minute long examination in healthy subjects, and patients with various severity of amyloid heart disease. The results show that despite critical application of the standard echocardiographic recording recommendations, E/e’ beat-to-beat variability was 36 % (22 to 50%) in healthy subjects and 17 % (11-26%) in patients, and where the most severe amyloid heart disease had the least variability. Thus, clinical use of a single or few cardiac beats might not necessarily reflect an accurate ratio between the two velocities, and hence casts doubt over their diagnostic value.

Skeletal muscle contraction is related to rapid mechanical shortening and thickening. Recently, specialized ultrasound systems have been applied to demonstrate and quantify transient tissue velocities and one-dimensional (1-D) propagation of mechanical waves during muscle contraction. Such waves could potentially provide novel information on musculoskeletal characteristics, function and disorders. In this work, we demonstrate two-dimensional (2-D) mechanical wave imaging following the skeletal muscle contraction. B-mode image acquisition during multiple consecutive electrostimulations, speckle-tracking and a time-stamp sorting protocol were used to obtain 1.4 kHz frame rate 2-D tissue velocity imaging of the biceps brachii muscle contraction. The results present novel information on tissue velocity profiles and mechanical wave propagation. In particular, counter-propagating compressional and shear waves in the longitudinal direction were observed in the contracting tissue (speed 2.8-4.4 m/s) and a compressional wave in the transverse direction of the non-contracting muscle tissue (1.2-1.9 m/s). In conclusion, analysing transient 2-D tissue velocity allows simultaneous assessment of both active and passive muscle tissue properties. (E-mail: christer.gronlund@vll.se) (C) 2013 World Federation for Ultrasound in Medicine & Biology.

In SPECT, diagnoses based on quantitative measurements may be uncertain due to high noise levels and low spatial resolution. 123I DAT SPECT has been shown to have a relatively high sensitivity and specificity, but improving image quality could potentially increase these values even further, especially for early cases with parkinsonian syndromes. The aim of the study was to optimise the reconstruction protocol for 123I DAT SPECT with a LEGP collimator, using a resolution recovery algorithm included in the iterative reconstruction, and compare to images reconstructed without resolution recovery. The optimization concentrated on critical frequency of the post-reconstruction Butterworth filter and the number of reconstruction iterations. Monte Carlo simulations of a morphological brain phantom with typical DAT SPECT uptake were used for this part of the study. From contrast-to-noise diagrams, it was found that a critical frequency of 0.50 cm-1 (power factor 8) was the most optimal of the studied filters. The optimal number of OSEM iterations was evaluated by a radiologist, specialized in nuclear medicine, and 8 iterations with 6 subsets were chosen. A group of 20 subjects diagnosed with Parkinson’s disease (PD) were then be compared to a group of 20 healthy controls, with respect to uptake ratios for caudate nucleus, putamen and the whole striatum (background region: whole cortex or the occipital lobe). Uptake ratios were calculated using the software Exini DAT for images reconstructed both with and without resolution recovery. It was found that the group differences were highly significant both with and without resolution recovery. However, in putamen, where early stages of PD first manifests, the group significance of uptake ratios improved from 7.2E-14 to 8.2E-15 (background: occipital lobe) or 2.4E-14 to 8.4E-16 (background: whole cortex) when using resolution recovery. A higher spatial resolution seems to be an advantage for quantitative evaluation of 123I DAT SPECT.

Computed Tomography (CT) high resolution imaging have been used to investigate if there is a significant change in the crystal-to-window distance, i.e. the air gap thickness, in a small n-type detector cooled to 77 K, and in a medium sized p-type HPGe detector when cooled to 100 K. The findings were compared to detector dimension data made available by the manufacturer. The air gap thickness increased by (0.38 +/- 0.07) mm for the n-type detector and by (0.40 +/- 0.15) mm for the p-type detector when the detectors were cooled to 77 resp. 100 K compared to at room temperature. Monte Carlo calculations indicate that these differences have a significant impact on the efficiency in close geometries (< 5 cm). In the energy range of 40-700 keV with a source placed directly on endcap, the change in detector efficiency with temperature is 1.9-2.9% for the n-type detector and 0.3-2.1% for the p-type detector. The measured air gap thickness when cooling the detector was 1.1 mm thicker than manufacturer data for the n-type detector and 0.2 mm thicker for the p-type detector. In the energy range of 40-700 keV and with a source on endcap, this result in a change in detector efficiency of 5.2-7.1% for the n-type detector and 0.2-1.0% for the p-type detector, Le the detector efficiency is overestimated using data available by the manufacturer. (C) 2015 Elsevier B.V. All rights reserved.

Background: Glioblastoma (GBM) is the most common and malignant primary brain tumor. It is a rapidly progressing tumor that infiltrates the adjacent healthy brain tissue and is difficult to treat. Despite modern treatment including surgical resection followed by radiochemotherapy and adjuvant chemotherapy, the outcome remains poor. The median overall survival is 10-12 months. Neuroimaging is the most important diagnostic tool in the assessment of GBMs and the current imaging standard is contrast-enhanced magnetic resonance imaging (MRI). Positron emission tomography (PET) has been recommended as a complementary imaging modality. PET provides additional information to MRI, in biological behavior and aggressiveness of the tumor. This study aims to investigate if the combination of PET and MRI can improve the diagnostic assessment of these tumors.

Patients and methods: In this study, 22 patients fulfilled the inclusion criteria, diagnosed with GBM, and participated in all four 18F-fluorothymidine (FLT)-PET/MR examinations. FLT-PET/MR examinations were performed preoperative (baseline), before the start of the oncological therapy, at two and six weeks into therapy. Optimization of an adaptive thresholding algorithm, a batch processing pipeline, and image feature extraction algorithms were developed and implemented in MATLAB and the analyzing tool imlook4d.

Results: There was a significant difference in radiochemotherapy treatment response between long-term and short-term survivors’ tumor volume in MRI (p<0.05), and marginally significant (p<0.10) for maximum standard uptake value (SUVmax), PET tumor volume, and total lesion activity (TLA).

Preoperative short-term survivors had on average larger tumor volume, higher SUV, and total lesion activity (TLA). The overall trend seen was that long-term survivors had a better treatment response in both MRI and PET than short-term survivors. 

During radiochemotherapy, long-term survivors displayed shrinking MR tumor volume after two weeks, and almost no remaining tumor volume was left after six weeks; the short-term survivors display marginal tumor volume reduction during radiochemotherapy.

In PET, long-term survivors mean tumor volumes start to decrease two weeks into radiochemotherapy. Short-term survivors do not show any PET volume reduction two and six weeks into radiochemotherapy. For patients with more or less than 200 days progression-free survival, PET volume and TLA were significantly different, and MR volume only marginally significant, suggesting that PET possibly could have added value.

Conclusion: The combination of PET and MRI can be used to predict radiochemotherapy response between two and six weeks, predicting overall survival and progression-free survival using MR and PET volume, SUVmax, and TLA. This study is limited by small sample size and further research with greater number of participants is recommended.

Purpose: To mitigate the problem of noisy parameter maps with high uncertainties by casting parameter mapping as a denoising task based on Deep Image Priors.

Methods: We extend the concept of denoising with Deep Image Prior (DIP) into parameter mapping by treating the output of an image-generating network as a parametrization of tissue parameter maps. The method implicitly denoises the parameter mapping process by filtering low-level image features with an untrained convolutional neural network (CNN). Our implementation includes uncertainty estimation from Bernoulli approximate variational inference, implemented with MC dropout, which provides model uncertainty in each voxel of the denoised parameter maps. The method is modular, so the specifics of different applications (e.g., T1 mapping) separate into application-specific signal equation blocks. We evaluate the method on variable flip angle T1 mapping, multi-echo T2 mapping, and apparent diffusion coefficient mapping.

Results: We found that deep image prior adapts successfully to several applications in parameter mapping. In all evaluations, the method produces noise-reduced parameter maps with decreased uncertainty compared to conventional methods. The downsides of the proposed method are the long computational time and the introduction of some bias from the denoising prior.

Conclusion: DIP successfully denoise the parameter mapping process and applies to several applications with limited hyperparameter tuning. Further, it is easy to implement since DIP methods do not use network training data. Although time-consuming, uncertainty information from MC dropout makes the method more robust and provides useful information when properly calibrated.

Computed tomography (CT) equivalent information is needed for attenuation correction in PET imaging and for dose planning in radiotherapy. Prior work has shown that Gaussian mixture models can be used to generate a substitute CT (s-CT) image from a specific set of MRI modalities. This work introduces a more flexible class of mixture models for s-CT generation, that incorporates spatial dependency in the data through a Markov random field prior on the latent field of class memberships associated with a mixture model. Furthermore, the mixture distributions are extended from Gaussian to normal inverse Gaussian (NIG), allowing heavier tails and skewness. The amount of data needed to train a model for s-CT generation is of the order of 10^8 voxels. The computational efficiency of the parameter estimationand prediction methods are hence paramount, especially when spatial dependency is included in the models. A stochastic Expectation Maximization (EM) gradient algorithm is proposed in order to tackle this challenge. The advantages of the spatial model and NIG distributions are evaluated with a cross-validation study based ondata from 14 patients. The study show that the proposed model enhances the predictive quality of the s-CT images by reducing the mean absolute error with 17.9%. Also, the distribution of CT values conditioned on the MR images are better explainedby the proposed model as evaluated using continuous ranked probability scores.

Prostate cancer is one of the most common cancer diagnosis in men. This project aimed to help in characterization and treatment planning of prostate cancer by producing a Gleason grading probability based on apparent diffusion coefficient (ADC).

In a study, from which this project received the patient data, the patients were first imaged using magnetic resonance imaging (MRI) in a 3T positron emission tomography MRI (PET/MRI) scanner. The prostates were surgically removed and placed in a patient specific mold. While inside the mold, the prostates were imaged using the same scanner, producing ex-vivo images of the prostates. Lastly the prostates were cut in histopathology slices and Gleason graded by a pathologist. To get correlation between ADC and Gleason grade all images needed to be correctly related to each other. This was done by three image registrations, which was the main part of this project. The histopathology slices were first registered to the ex-vivo images of the prostate, and then to the in-vivo T2-weighted images. The in-vivo T2w images were matched to images depicting the diffusion of water in the prostates, known as ADC-maps. The ADC-values were collected and matched to their possible Gleason grade.

Information from 149 images were used, which came from 22 different patients. 3D pixels, known as voxels, with a corresponding Gleason grade annotation measured a lower average ADC-value. These voxels also showed more variation with a larger standard deviation. Furthermore, these voxels measured a larger range of ADC-values compared to voxels without a corresponding Gleason grade, but the probability of a Gleason grade was mainly seen for ADC-values below 1200 mm²/s. Filtering the ADC-map before collecting the information showed less spread in measurements, and larger total probability of Gleason grade annotation for lower ADC-values. To test the validity of the result a movement of the Gleason grade map was used to simulate registration errors. No large impact was observed for small movements but more obvious change for large.

The results indicate this method as promising in predicting regions with a probability for Gleason grade of 3 or 4, however it was less accurate in separating the two. Gleason 5 showed very low probability, mainly as a result of the low sample size since only two patients had such tumors. Further research with better optimized filtering is recommended in the future.

The aim of this study is to optimise the acquisition and reconstruction for SPECT with ¹¹¹In- pentetrotide with the iterative reconstruction software OSEMS. For ¹¹¹In-pentetrotide SPECT, the uptake in the tumour is usually high compared to uptake in normal tissue. It may, however, be difficult to detect small tumours with the SPECT method because of high noise levels and the low spatial resolution. The liver is a common region for somatostatin-positive metastases, and to visually detect small tumours in the liver, as early as possible, is important for an effective treatment of the cancer disease.

The study concentrates on the acquired number of projections, the subset size in the OSEM reconstruction and evaluates contrast as a function of noise for a range of OSEM iterations. The raw-data projections are produced using Monte Carlo simulations of an anthropomorphic phantom, including tumours in the liver. Two General Electric (GE) collimators are evaluated, the extended low-energy general-purpose (ELEGP) and the medium energy general-purpose (MEGP) collimator. Three main areas of reconstruction are investigated. First the reconstructions are performed for so called full time scans with the acquisition time used clinically. Also the effect of performing the examination in half-time or with half the injected activity is evaluated with the most optimal settings gained from the full time scans for both collimators. In addition images reconstructed without model-based compensation are also included for comparison.

This study is a continuation of a previous study on ¹¹¹In-pentetrotide SPECT where collimator choice and model-based compensation were evaluated for a cylindrical phantom representing small tumours in liver background. As in the previous study, ELEGP proved to be the better collimator. For ELEGP, the most optimal setting was 30 projection angles and a subset size of 6 projections in the OSEM reconstruction, and for MEGP optimal setting was 60 projections and 4 subsets. The difference between the different collimator settings were, however, rather small but proven significant. For both collimators the half-time scan including model-based compensation was better compared to the full-time reconstructions without model-based compensation.

Purpose: Many methods are available to segment structural magnetic resonance (MR) images of the brain into different tissue types. These have generally been developed for research purposes but there is some clinical use in the diagnosis of neurodegenerative diseases such as dementia. The potential exists for computed tomography (CT) segmentation to be used in place of MRI segmentation, but this will require a method to verify the accuracy of CT processing, particularly if algorithms developed for MR are used, as MR has notably greater tissue contrast.

Methods: To investigate these issues we have created a three-dimensional (3D) printed brain with realistic Hounsfield unit (HU) values based on tissue maps segmented directly from an individual T1 MRI scan of a normal subject. Several T1 MRI scans of normal subjects from the ADNI database were segmented using SPM12 and used to create stereolithography files of different tissues for 3D printing. The attenuation properties of several material blends were investigated, and three suitable formulations were used to print an object expected to have realistic geometry and attenuation properties. A skull was simulated by coating the object with plaster of Paris impregnated bandages. Using two CT scanners, the realism of the phantom was assessed by the measurement of HU values, SPM12 segmentation and comparison with the source data used to create the phantom.

Results: Realistic relative HU values were measured although a subtraction of 60 was required to obtain equivalence with the expected values (gray matter 32.9-35.8 phantom, 29.9-34.2 literature). Segmentation of images acquired at different kVps/mAs showed excellent agreement with the source data (Dice Similarity Coefficient 0.79 for gray matter). The performance of two scanners with two segmentation methods was compared, with the scanners found to have similar performance and with one segmentation method clearly superior to the other.

Conclusion: The ability to use 3D printing to create a realistic (in terms of geometry and attenuation properties) head phantom has been demonstrated and used in an initial assessment of CT segmentation accuracy using freely available software developed for MRI.

Blood flow measurements are important for understanding the development of cerebrovascular diseases. With 4D flow magnetic resonance imaging (4D flow MRI), simultaneous velocity measurements are obtained in all cerebral arteries in a scan of about ten minutes. However, 4D flow MRI is a relatively new technique. For usefulness in both clinics and research, detailed knowledge is needed about its accuracy and precision for flow quantification. In patients with stroke or transient ischemic attack (TIA) from a symptomatic carotid stenosis, the stenosis may generate a difference in blood pressure and flow between the left and right cerebral hemispheres. Such a hemispheric pressure difference could be an early marker of to what extent a stenosis is affecting cerebral hemodynamics, which could be useful in the planning of carotid surgery. 

The overall aim of the thesis was to determine the accuracy of 4D flow MRI to measure cerebral arterial blood flow, and to develop and evaluate an approach combining 4D flow MRI and computational fluid dynamics (CFD) to characterize the cerebral arterial hemodynamics, with implementation in patients with symptomatic carotid stenosis. The thesis is based on four papers, investigating two cohorts.

The first cohort consisted of 35 elderly volunteers (mean age 79 years) and was studied in paper I-II. Blood flow rates were measured in nine cerebral arteries with 4D flow MRI and 2D phase-contrast MRI as reference. Three different flow quantification methods for 4D flow MRI were evaluated and optimized: one clustering approach and two threshold-based methods. The proposed new method, based on a locally adapted threshold, outperformed the previously suggested methods in flow rate quantification. For the clustering method, flow rates were systematically underestimated. 4D flow MRI was also evaluated to assess different arterial pulsatility measures, and a Windkessel model was used to estimate reference values for cerebrovascular resistance and cerebral arterial compliance in elderly.

The second cohort consisted of 28 stroke and TIA patients (mean age 73 years) with symptomatic carotid stenosis and was studied in paper III-IV. With 4D flow MRI and CFD, the preoperative hemispheric pressure laterality was quantified in the patients. The pressure laterality was compared to hemispheric flow lateralities. Estimating the hemispheric pressure laterality was a promising physiological biomarker for grading the cerebral arterial hemodynamic disturbances in patients with symptomatic carotid stenosis. A CFD model was also developed to predict carotid stump pressure, i.e., the important pressure measured in the clamped carotid artery during surgical removal of the stenosis. The predicted stump pressures were correlated with the pressures measured during surgery. Stump pressure prediction was promising and could be a potential tool in the preoperative planning in order to avoid hypoperfusion during surgery. 

In summary, post-processing methods were successfully developed and evaluated for accurate assessment of mean and pulsatile cerebral blood flow rates with 4D flow MRI. Thereby, this thesis provided knowledge about possibilities and limitations of how 4D flow MRI can be used with respect to cerebral arterial blood flow rate assessment. By contributing with models combining 4D flow MRI and CFD, specifically developed for analysis of pressure distributions in cerebral arteries, novel methods were proposed for assessing patients with symptomatic carotid stenosis in the planning of carotid surgery.

An internal carotid artery (ICA) stenosis can potentially decrease the perfusion pressure to the brain. In this study, computational fluid dynamics (CFD) was used to study if there was a hemispheric pressure laterality between the contra- and ipsilateral middle cerebral artery (MCA) in patients with a symptomatic ICA stenosis. We further investigated if this MCA pressure laterality (ΔP_MCA) was related to the hemispheric flow laterality (ΔQ) in the anterior circulation, i.e., ICA, proximal MCA and the proximal anterior cerebral artery (ACA). Twenty-eight patients (73±6 years, range 59–80 years, 21 men) with symptomatic ICA stenosis were included. Flow rates were measured using 4D flow MRI data (PC-VIPR) and vessel geometries were obtained from computed tomography angiography. The ΔP_MCA was calculated from CFD, where patient-specific flow rates were applied at all input- and output boundaries. The ΔP_MCA between the contra- and ipsilateral side was 6.4±8.3 mmHg (p<0.001) (median 3.9 mmHg, range -1.3 to 31.9 mmHg). There was a linear correlation between the ΔP_MCA and ΔQ_ICA (r = 0.85, p<0.001) and ΔQ_ACA (r = 0.71, p<0.001), respectively. The correlation to ΔQ_MCA was weaker (r = 0.47, p = 0.011). In conclusion, the MCA pressure laterality obtained with CFD, is a promising physiological biomarker that can grade the hemodynamic disturbance in patients with a symptomatic ICA stenosis.

There is always a need and drive to improve modern cancer care. Dynamic positron emission tomography (PET) offers the advantage of in vivo functional imaging, combined with the ability to follow the physiological processes over time. In addition, by applying tracer kinetic modeling to the dynamic PET data, thus estimating pharmacokinetic parameters associated to e.g. glucose metabolism, cell proliferation etc., more information about the tissue's underlying biology and physiology can be determined. This supplementary information can potentially be a considerable aid when it comes to the segmentation, diagnosis, staging, treatment planning, early treatment response monitoring and follow-up of cancerous tumors.

We have found it feasible to use kinetic parameters for semi-automatic tumor segmentation, and found parametric images to have higher contrast compared to static PET uptake images. There are however many possible sources of errors and uncertainties in kinetic parameters obtained through compartment modeling of dynamic PET data. The variation in the number of detected photons caused by the random nature of radioactive decay, is of course always a major source. Other sources may include: the choice of an appropriate model that is suitable for the radiotracer in question, camera detectors and electronics, image acquisition protocol, image reconstruction algorithm with corrections (attenuation, random and scattered coincidences, detector uniformity, decay) and so on. We have found the early frame sampling scheme in dynamic PET to affect the bias and uncertainty in calculated kinetic parameters, and that scatter corrections are necessary for most but not all parameter estimates. Furthermore, analytical image reconstruction algorithms seem more suited for compartment modeling applications compared to iterative algorithms.

This thesis and included papers show potential applications and tools for quantitative pharmacokinetic parameters in oncology, and help understand errors and uncertainties associated with them. The aim is to contribute to the long-term goal of enabling the use of dynamic PET and pharmacokinetic parameters for improvements of today's cancer care.