Umeå University's logo

Objectives: Gastroenterology clinics often assess hepatic steatosis using CAP-FibroScan, while radiology departments increasingly apply UGAP instead of subjective B-mode ultrasound. This study compares CAP and UGAP feasibility and diagnostic performance across steatosis stages, using PDFF as reference.

Materials and methods: Healthy controls and a cohort with known steatosis and fibrosis were examined between September 2022 and October 2024. Presence of steatosis (≥S1) defined as ≥5% PDFF, and presence of fibrosis was evaluated with MRE. Participants with even sex distribution were examined in supine and 30° left decubitus position; for UGAP, with normal (4 N) and (30 N) probe force. Diagnostic performance was evaluated by the area under the receiver operating characteristic curve (AUROC).

Results: In the group of N = 97 CAP demonstrated 91% feasibility in supine and 80% in lateral position. UGAP showed 100% feasibility for all examination techniques. The whole group was divided according to steatosis stages of PDFF. When differentiating ≥ S1, CAP supine accuracy was AUC 0.81 (95%CI: 0.71–0.92), and UGAP supine/30N accuracy was 0.88 (95%CI: 0.88–0.95). Differentiating S0 and S1 vs. S2 and S3, the CAP AUC was 0.81 (95% CI: 0.72–0.90), and the UGAP supine/30 N AUC was 0.93 (95%CI: 0.88–0.99). When differentiating S0, S1, and S2 vs. S3, the CAP AUC was 0.90 (95%CI: 0.83–0.97), and the UGAP supine/4N AUC was 0.97 (95%CI: 0.94–1.00). UGAP increased performance in both sexes using increased probe force.

Conclusions: UGAP provides absolute feasibility and higher diagnostic performance. CAP should not be performed in left position.

Aim: Hitherto, breathing phases impact on hepatic steatosis measurements with quantitative ultrasound method in patients with liver fibrosis is not well known. The aim of the study was to evaluate different breathing phases for the ultrasound guided attenuation parameter (UGAP) technology for hepatic steatosis in patients with liver fibrosis.

Material and methods: Healthy controls and a cohort with steatosis and fibrosis was prospectively enrolled between September 2022 and October 2024. Presence of hepatic steatosis (≥ S1) with UGAP was defined as magnetic resonance imaging ≥ 5 % proton density fat fraction (PDFF), and presence of fibrosis was evaluated with magnetic resonance elastography. In a group with N = 55 measurements were sampled during normal breath hold, peak-inspiration and end-expiration phases. In a group with n = 37 free breathing phase was added during recorded volume sampling for measurements after examination. The diagnostic performance of UGAP for all four breathing methods were evaluated based on area under the receiver operating characteristic curve (AUROC) with PDFF.

Results: In group N = 55 no difference in diagnostic performance was seen between AUC for normal breath hold 0.79 (95 %CI: 0.66–0.92), inspiration 0.78 (95 %CI: 0.64–0.91) and expiration 0.77 (95 %CI: 0.64–0.91). In group n = 37 no difference was seen between AUC for normal breath hold 0.71 (95 %CI: 0.53–0.89), inspiration 0.66 (0.47–0.85), expiration 0.67 (95 %CI: 0.49–0.86) and free breathing 0.72 (95 %CI: 0.55–0.90). No difference between normal breath-hold UGAP mean values dB/cm/MHz and all tested breathing phases mean values (n = 37, NS).

Conclusion: Patients with liver fibrosis and inability to hold their breath during measurements for hepatic steatosis can be measured using UGAP technology with sustained diagnostic accuracy.

Deep learning has been used extensively for medical image analysis applications, assuming the training and test data adhere to the same probability distributions. However, a common challenge arises when dealing with medical images generated by different systems or even the same system with varying parameter settings. Such images often contain diverse textures and noise patterns, violating the assumption. Consequently, models trained on data from one machine or setting usually struggle to perform effectively on data from another. To address this issue in ultrasound images, we proposed a Generative Adversarial Network (GAN) based model in this paper. We formulated image harmonization and denoising tasks as an image-to-image translation task, wherein we adapt the texture pattern and reduced noise in Carotid ultrasound images while keeping the image content (the anatomy) unchanged. The performance was evaluated using feature distribution and pixel-space similarity metrics. In addition, blood-to-tissue contrast and influence on computed risk markers (Grey scale median, GSM) were evaluated. The results showed that domain adaptation was achieved in both tasks (histogram correlation 0.920 (0.043) and 0.844 (0.062)), as compared to no adaptation (0.890 (0.077) and 0.707 (0.098)), and that the anatomy of the images was retained (structure similarity index measure e.g. the arterial wall 0.71 (0.09) and 0.80 (0.08)). In addition, the image noise level (contrast) did not change in the image harmonization task (-34.1 (3.8) vs -35.2 (4.1) dB) but was improved in the noise reduction task (-23.5 (3.2) vs -46.7 (18.1) dB). To validate the performance of the proposed model, we compare its results with CycleGAN, the current state-of-the-art model. Our model outperformed CycleGAN in both tasks. Finally, the risk marker GSM was significantly changed in the noise reduction but not in the image harmonization task. We conclude that domain translation models are powerful tools for improving ultrasound image while retaining the underlying anatomy, but downstream calculations of risk markers may be affected.

Detecting and segmenting arterial plaques in ultrasound images is essential for the early diagnosis and prevention of cardiovascular diseases. This paper presents Hybrid Stable AnoGAN (HS-AnoGAN), an enhanced anomaly detection framework based on AnoGAN (Anomaly Generative Adversarial Network), which utilizes generative adversarial learning to model normal anatomical structures and identify abnormal regions indicative of pathology. The proposed approach introduces key improvements, including direct latent space encoding, hybrid reconstruction loss, feature matching in the discriminator, and adaptive thresholding, leading to more precise anomaly localization. Additionally, spectral normalization and Wasserstein loss with gradient penalty are incorporated to improve training stability and prevent mode collapse. To the best of our knowledge, this is the first attempt to apply anomaly detection techniques for arterial plaque detection and segmentation in ultrasound images. Comparative experiments demonstrate that HS-AnoGAN outperforms state-of-the-art methods, achieving a 9.8% increase in detection accuracy, and a 7.5% enhancement in Dice score for segmentation quality. These results highlight the effectiveness of HS-AnoGAN in improving both plaque detection and segmentation in ultrasound imaging, making it a promising tool for clinical applications.

Atherosclerosis is a leading cardiovascular disease typified by fatty streaks accumulating within arterial walls, culminating in potential plaque ruptures and subsequent strokes. Existing clinical risk scores, such as systematic coronary risk estimation and Framingham risk score, profile cardiovascular risks based on factors like age, cholesterol, and smoking, among others. However, these scores display limited sensitivity in early disease detection. Parallelly, ultrasound-based risk markers, such as the carotid intima media thickness, while informative, only offer limited predictive power. Notably, current models largely focus on either ultrasound image-derived risk markers or clinical risk factor data without combining both for a comprehensive, multimodal assessment. This study introduces a multimodal ensemble learning framework to assess atherosclerosis severity, especially in its early sub-clinical stage. We utilize a multi-objective optimization targeting both performance and diversity, aiming to integrate features from each modality effectively. Our objective is to measure the efficacy of models using multimodal data in assessing vascular aging, i.e., plaque presence and vascular age, over a six-year period. We also delineate a procedure for optimal model selection from a vast pool, focusing on best-suited models for classification tasks. Additionally, through eXplainable Artificial Intelligence techniques, this work delves into understanding key model contributors and discerning unique subject subgroups.

Background/Introduction: Cardiovascular diseases (CVDs) are the leading cause of death worldwide, with atherosclerosis being a major contributor. Early detection of carotid artery plaques is crucial for CVD prevention and management. While ultrasound imaging is widely used, its manual analysis is time-consuming and prone to variability. Automating plaque detection with deep learning can improve diagnostic consistency and efficiency. Recent advancements in unsupervised anomaly detection models offer a promising approach to detect plaque-related anomalies in carotid ultrasound images without labeled data. This study is, to the best of our knowledge, the first to apply these models for plaque detection in ultrasound images.

Purpose: This work aims to evaluate the effectiveness of three unsupervised anomaly detection methods—Autoencoder (AE), Variational Autoencoder (VAE), and Anomaly Detection Generative Adversarial Networks (AnoGAN)—for detecting plaques in longitudinal carotid artery ultrasound images. We used carotid ultrasound images from a large Randomized Controlled Trial (RCT) on subclinical atherosclerosis, comprising images from a diverse group of participants collected at both baseline and follow-up periods.

Methods: The generative models were trained to recognize normal anatomical structures using 3,000 plaque-free (normal) images. These models were then tested on 400 images, which included both normal and abnormal images, to distinguish them and detect anomalies in abnormal images. As shown in Figure 1, the anomaly scores were computed by comparing the original images to the generated ones, with a distribution-based cutoff applied to classify the images as normal or abnormal. The threshold for this cutoff was determined based on the distribution of the anomaly scores.

Results: AnoGAN achieved the highest accuracy, as shown in Figure 2, which compares the input image and generated output of AnoGAN, VAE, and AE for an abnormal image. Also, AnoGAN achieved the highest Area Under the Curve of the Receiver Operating Characteristic (AUC-ROC) at 91%, followed by VAE at 79% and AE at 66%. Additionally, AnoGAN achieved the best sensitivity-specificity balance, resulting in the highest AUC-ROC, while VAE demonstrated moderate performance, and AE, despite lower sensitivity, maintained competitive specificity.

Conclusion(s): This study shows the effectiveness of anomaly detection models, especially AnoGAN, in automating the detection of plaques in carotid ultrasound imaging. These models have the potential to reduce diagnostic variability, improve early detection of CVD risk factors, and enhance clinical efficiency. Future research should aim to expand the use of these models to larger datasets and explore their clinical integration to assess real-world applicability. Additionally, optimizing the computational performance and interpretability of these models will be crucial for their widespread adoption in routine clinical settings.

The silent progression of cardiovascular disease (CVD) is a major problem particularly in CVD prevention. New techniques enabled by the rise of electronic health records may facilitate CVD prevention. Both public health research and big data applications, such as digital twins, are dependent on access to longitudinal and sensitive data; a challenge which may be facilitated by access to longitudinal synthetic data. In this study, we establish a fidelity benchmark for longitudinal synthetic data by extending a well-known method for cross-sectional synthetic data to a longitudinal application within CVD. We find that the univariate distributional difference between the real and the synthetic data is kept low and that pairwise relations are preserved in the synthetic data. Further, we see that the variable-wise temporal trends are preserved, yet may be more extensively studied and have some room for improvement. The results of this study is important to enable future studies within public health prevention and cardiovascular digital twins.

Ultrasound (US) research scanners have recently been used to obtain motor unit (MU) activity. To improve clinical applicability, this study investigates whether similar MU activity can be detected using clinical US scanners.

A tibialis anterior muscle was simultaneously scanned using clinical US and surface electromyography (sEMG). Tissue velocity imaging (TVI) data was estimated from B-mode images, and spatial maps and twitch profiles were estimated from the TVI using spike-triggered averaging (STA).

The MU action potentials obtained from sEMG and US-derived spatial maps were estimated to be at approximately the same location. This demonstrates that clinical US may offer an accessible alternative for MU research

BACKGROUND: Elevated low-density lipoprotein (LDL) cholesterol levels represent a significant modifiable risk factor for atherosclerotic cardiovascular disease. However, a residual risk persists, possibly attributed to other atherogenic lipoproteins such as non-high-density lipoprotein (non-HDL) and remnant cholesterol. Nevertheless, few studies have explored the independent associations between these lipid biomarkers and early atherosclerotic disease.

OBJECTIVE: To evaluate the relative contributions of LDL, non-HDL, and remnant cholesterol to subclinical atherosclerosis, assessed by carotid ultrasonography.

METHOD: In this cross-sectional study, we included 1929 previously healthy individuals from the pragmatic VIPVIZA trial who had available lipid levels and carotid ultrasonography results to assess subclinical disease. Non-HDL, LDL, and remnant cholesterol were calculated from a standard lipid profile. Subclinical atherosclerosis was assessed by carotid intima-media thickness (cIMT) and the presence of carotid plaques.

RESULTS: We found that all lipid variables (LDL, non-HDL, and remnant cholesterol) were associated with subclinical atherosclerosis in univariable models (P < .01 across all models for cIMT and P < .001, P < .001, P = .003 respectively for carotid plaques). In multivariable-adjusted models, increasing LDL and non-HDL cholesterol levels were still significantly associated with increased odds of having carotid plaques (P < .001 for both) and increased cIMT (P < .001 for both). However, no independent association between remnant cholesterol and subclinical atherosclerosis was observed in the model adjusted for LDL cholesterol levels (P = .073 for cIMT and = .818 for plaque).

CONCLUSION: Increasing LDL and non-HDL cholesterol levels, but not remnant cholesterol, seem to contribute to carotid subclinical atherosclerosis.

Objectives: To evaluate the diagnostic performance of ultrasound guided attenuation parameter (UGAP) for evaluating liver fat content with different probe forces and body positions, in relation to sex, and compared with proton density fat fraction (PDFF).

Methods: We prospectively enrolled a metabolic dysfunction-associated steatotic liver disease (MASLD) cohort that underwent UGAP and PDFF in the autumn of 2022. Mean UGAP values were obtained in supine and 30° left decubitus body position with normal 4 N and increased 30 N probe force. The diagnostic performance was evaluated by the area under the receiver operating characteristic curve (AUC).

Results: Among 60 individuals (mean age 52.9 years, SD 12.9; 30 men), we found the best diagnostic performance with increased probe force in 30° left decubitus position (AUC 0.90; 95% CI 0.82–0.98) with a cut-off of 0.58 dB/cm/MHz. For men, the best performance was in supine (AUC 0.91; 95% CI 0.81–1.00) with a cut-off of 0.60 dB/cm/MHz, and for women, 30° left decubitus position (AUC 0.93; 95% CI 0.83–1.00), with a cut-off 0.56 dB/cm/MHz, and increased 30 N probe force for both genders. No difference was in the mean UGAP value when altering body position. UGAP showed good to excellent intra-reproducibility (Intra-class correlation 0.872; 95% CI 0.794–0.921).

Conclusion: UGAP provides excellent diagnostic performance to detect liver fat content in metabolic dysfunction-associated steatotic liver diseases, with good to excellent intra-reproducibility. Regardless of sex, the highest diagnostic accuracy is achieved with increased probe force with men in supine and women in 30° left decubitus position, yielding different cut-offs.

Clinical relevance statement: The ultrasound method ultrasound-guided attenuation parameter shows excellent diagnostic accuracy and performs with good to excellent reproducibility. There is a possibility to alter body position and increase probe pressure, and different performances for men and women should be considered for the highest accuracy.