Umeå University's logo

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.

Background: The ABO blood group system has previously been associated with cardiovascular disease (CVD), where non-O blood group individuals have shown an increased risk. Studies assessing early atherosclerotic disease while also including RhD are few. We aimed to determine whether the ABO and RhD blood groups are associated with subclinical atherosclerosis in a healthy population.

Methods: We included 3532 participants from the VIPVIZA trial with available carotid ultrasonography results to assess subclinical disease. Information about blood groups was obtained from the SCANDAT-3 database, where 85% of VIPVIZA participants were registered.

Results: RhD− individuals aged 40 years showed increased carotid intima–media thickness (B 1.09 CI 95% 1.03; 1.14) compared to RhD+ individuals. For ABO, there were no differences in ultrasonography results when assessing the whole study population. However, 60-year-old individuals with heredity for CVD and a non-O blood group had decreased odds for carotid plaques (OR 0.54 CI 95% 0.33; 0.88).

Conclusions: RhD blood group is associated with subclinical atherosclerosis in younger individuals, indicating a role as a mediator in the atherosclerotic process. In addition, a non-O blood group was associated with decreased subclinical atherosclerosis in individuals aged 60 and with heredity (corresponding to the group with the highest atherosclerotic burden).

Purpose: Aortic stiffness, assessed as estimated aortic pulse wave velocity (aPWV), and carotid intima-media thickness (cIMT) are markers of vascular age, and carotid plaques are a marker of early atherosclerosis. In this cross-sectional study we aimed to investigate the association between aPWV, cIMT and plaques across different age groups and in women and men, in a middle-aged healthy population.

Materials and methods: Participants in the 6.5-year follow-up of the VIPVIZA trial who were aged 47, 57 and 67 underwent an oscillometric measurement which estimates aPWV between 2020 and 2023. Carotid ultrasound examinations were also performed. Linear and ordinal regression models were used to investigate how aPWV associates with cIMT and with carotid plaques, for the overall study group and stratified for age groups and sex.

Results: A total of 1046 subjects were included in the analyses. Linear associations between aPWV and cIMT (β = 0.018, 95% CI: 0.006–0.030, p = 0.003), and between aPWV and plaques (OR: 1.19, 95% CI: 1.03–1.38, p = 0.018), were seen in the 57-year-olds. In the 47-year-olds a significant association was seen between aPWV and plaques (OR: 2.98 95% CI: 1.44–6.14, p = 0.003). No significant associations were seen in the 67-year-olds. For women, a significant association between aPWV and cIMT (β = 0.011, 95% CI: 0.004–0.017, p = 0.002) was shown.

Conclusion: Estimated aPWV was positively associated with increasing cIMT and the presence of carotid plaques in younger middle-aged individuals, and with cIMT in women, suggesting that measurement of estimated aPWV may improve cardiovascular risk assessment in younger middle-aged individuals and women.

Clinical Trial Registration date 8 May 2013: URL: www.clinicaltrials.gov. Unique identifier: NCT01849575.

Background and aims: Studies on the influence of fasting plasma glucose (FPG) on the development of carotid plaque (CP) and intima media thickness (CIMT) mainly focused on single FPG measures. We investigated whether changes in FPG (ΔFPG) are associated with incident CP and CIMT change (ΔCIMT) over time.

Methods: Analyses were based on information from 1896 participants from the VIPVIZA trial (Visualization of asymptomatic atherosclerotic disease for optimum cardiovascular prevention), with baseline and 3-year follow-up data on FPG, ultrasonographic CP (none or ≥1 lesion/s) and CIMT assessments. We studied the association between baseline FPG (prior to intervention) or 3-year ΔFPG (mmol/L) and incident CP (logistic regression) or ΔCIMT (linear regression). Analyses were adjusted for multiple potential confounders.

Results: 1896 and 873 individuals, respectively, were included in the analysis on incident CP and ΔCIMT. Participants were 60 years old at baseline and 61% and 54% were females, in the CP and CIMT analyses, respectively. Every mmol/L increase in FPG was associated with an increased odds of incident CP (odds ratio: 1.42, 95% confidence interval [CI]: 1.17, 1.73), but there was no association with ΔCIMT (mean difference: 0.002 mm, 95% CI: −0.003, 0.008) after 3 years. Baseline FPG was not associated with incident CP nor ΔCIMT progression.

Conclusions: In middle-aged individuals with low to moderate risk for cardiovascular diseases, 3-year ΔFPG was positively associated with the risk of incident CP, but not with ΔCIMT. Single measures of FPG may not be sufficient in estimating cardiovascular risk among individuals with low to moderate risk.

Objective: To test the hypothesis of low emotional support being associated with lifestyle and biomedical cardiovascular disease (CVD) risk factors, estimated risk of CVD morbidity and mortality, and subclinical atherosclerosis in middle-aged healthy adults.

Methods and measures: Cross-sectional data were obtained from participants aged 40–60 years who had one or more conventional CVD risk factor. They underwent assessment based on questionnaires, clinical examination, blood sampling, and carotid ultrasound of plaque formation and carotid intima-media wall thickness (cIMT). Based on the Interview Schedule for Social Interaction, the participants were categorised as either low in emotional support (n = 884) or as a referent (n = 2570). Logistic regression analyses were conducted to study the associations.

Results: Logistic regression analyses showed that low emotional support was significantly associated with smoking, alcohol consumption and physical inactivity (OR = 1.53 − 1.94), estimated risk of CVD morbidity and mortality (OR = 1.56 − 1.68), and plaque formation (OR = 1.39). No significant associations were found regarding biomedical CVD risk factors or cIMT.

Conclusion: The findings suggest that low social support is associated with lifestyle CVD risk factors, estimated risk of CVD morbidity and mortality, and subclinical atherosclerosis in middle-aged healthy adults, encouraging causal evaluation with longitudinal data investigating an impact of emotional support on mechanisms underlying CVD.

Ultrasonic echolucent carotid intima-media (IM) complex and accelerated progression of carotid intima mediathickness (cIMT) have both separately been shown to predict future cardiovascular events. The aim of this studywas to evaluate if the echogenicity of the IM-complex is associated with the 3-year progression of cIMT. B-modeultrasound images captured at baseline and 3-year follow-up in the ‘Visualization of asymptomatic atherosclerotic disease for optimum cardiovascular prevention’ (VIPVIZA) trial were included (n: 3154). The bilateral mean cIMT and IM-echogenicity by greyscale median (GSM) were measured in the common carotid artery. Associations between IM-GSM at baseline and the 3-year cIMT progression were investigated using linear regression models for the whole population and stratified by sex, age and VIPVIZA study group (intervention versus control). In addition, adjusted analyses for confounding factors were performed. Unadjusted analysis showed that decreased IM-GSM at baseline was associated with increased progression of cIMT (p < 0.001). Stratified by age, the association was significant among 40 (p < 0.001) and 60 years old (p < 0.001). The association was statistically significant in both sexes and on comparison of VIPVIZA study subgroups. Adjustments for confounding factors did not alter the estimated relationship between IM-GSM and cIMT progression. Echolucent carotid intima media at baseline associates with increased 3-year cIMT progression among an asymptomatic, middle-aged population. Echogenicity of the intima media may identify individuals at risk for accelerated vascular aging.

Objectives: The risk of cardiovascular disease is associated with the echo intensity of carotid plaques in ultrasound images and their cardiac cycle-induced intensity variations. In this study, we aimed to 1) explore the underlying origin of echo intensity variations by using simulations and 2) evaluate the association between the two-dimensional (2D) spatial distribution of these echo intensity variations and plaque vulnerability.

Methods: First, we analyzed how out-of-plane motion and compression of simulated scattering spheres of different sizes affect the ultrasound echo intensity. Next, we propose a method to analyze the features of the 2D spatial distribution of interframe plaque echo intensity in carotid ultrasound image sequences and explore their associations with plaque vulnerability in experimental data.

Results: The simulations showed that the magnitude of echo intensity changes was similar for both the out-of-plane motion and compression, but for scattering objects smaller than 1 mm radius, the out-of-plane motion dominated. In experimental data, maps of the 2D spatial distribution of the echo intensity variations had a low correlation with standard B-mode echo intensity distribution, indicating complementary information on plaque tissue composition. In addition, we found the existence of ∼1 mm diameter subregions with pronounced echo intensity variations associated with plaque vulnerability.

Conclusions: The results indicate that out-of-plane motion contributes to intra-plaque regions of high echo intensity variation. The 2D echo intensity variation maps may provide complementary information for assessing plaque composition and vulnerability. Further studies are needed to verify this method's role in identifying vulnerable plaques and predicting cardiovascular disease risk.

Objectives: Reduced progression of atherosclerosis can decrease the risk of cardiovascular disease (CVD). This study aimed at evaluating the effect of a pictorial intervention showing atherosclerotic severity on the progression of carotid atherosclerosis.

Methods: A prospective randomised open-label blinded end-point trial with participants aged 40–60 years enroled from a routine CVD prevention programme. The intervention group (n: 1575) and their treating physicians received an image based presentation of subclinical atherosclerotic severity measured by carotid ultrasound. The control group (n: 1579) did not receive any information about ultrasound results. Carotid ultrasound at baseline and at 3-year follow-up contained plaque detection and measurements of carotid intima media thickness (cIMT). The left, right and bilateral-mean-cIMT, plaque prevalence and total plaque area (TPA) at 3-year follow-up were compared between groups. Significance level was set to p = 0.01 to adjust for multiple comparisons.

Results: The intervention group revealed reduced cIMT progression in the left-mean-cIMT of −0.011 mm (p = 0.001) compared with the control group. The intervention effect on cIMT progression was most prominent in individuals with increased cIMT and plaque prevalence at baseline (−0.021 mm, p = 0.005). There were no differences in progression between groups for the right-and bilateral-mean-cIMT (−0.005 mm, p = 0.223 and −0.005 mm, p = 0.036, respectively), nor any differences between groups for plaque prevalence or TPA (odds ratio 0.88, p = 0.09 and 0.89, p = 0.21, respectively).

Conclusion: Pictorial presentation of subclinical atherosclerotic severity sent to both the individual and their treating physician resulted in significantly reduced left cIMT progression. Pictorial presentation has the potential to increase adherence in CVD prevention.

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.

Objective: Non-adherence to guidelines and preventive measures is a major challenge, particularly so to ob- tain long-term adherence to lifestyle changes and recommended medication. The objective was to investigate if pictorial information regarding subclinical carotid atherosclerosis provided to individuals and physicians gave sustained effects on cardiovascular risk beyond the previously reported effect after 1 year and up to 3 years. 

Methods: A Prospective Randomized Open Blinded End-point (PROBE) trial. Within a CVD prevention program in Västerbotten County, Sweden, 3532 healthy individuals aged 40, 50 or 60 years were enrolled and 1:1 ran- domized to intervention ( n = 1749; pictorial information with additional prevention materials to participants and physicians) or control group ( n = 1783; no pictorial information to participants and physicians). Preventive measures were managed within primary care. Participants were investigated at baseline during 2013–2016 and at follow-up after 1 and 3 years. 

Results: A beneficial effect on cardiovascular risk was observed at 3-year follow-up; Framingham Risk Score (FRS) was 13.38 for the intervention group and 14.08 for the control group ( p = 0.047) and SCORE was 1.69 vs. 1.82 ( p = 0.022). The effect observed at 1-year was sustained over 3 years after adjustment for sex and education and more pronounced among participants with a severe atherosclerotic picture at baseline.

Conclusions: This study provides evidence of sustained beneficial effects on the adherence to prevention guidelines over 3 years of pictorial information about subclinical carotid atherosclerosis, resulting in lower cardiovascular risk regardless of sex and educational level. Direct visualization of the underlying still subclinical atherosclerotic disease, rather than just indirect information about risk factors and statistical risk of future myocardial infarction, stroke and death, is one way to tackle the problem of non-adherence to prevention of cardiovascular diseases.