Logo: to the web site of Uppsala University

The study of associations between an individual’s age and imaging and non-imaging data is an active research area that attempts to aid understanding of the effects and patterns of aging. In this work, we have conducted a supervoxel-wise association study between both volumetric and tissue attenuation features in coronary computed tomography angiograms (CCTA) and the chronological age of a subject, to understand the localized changes in morphology and CT attenuation (as a measure of tissue density) with age. To enable a supervoxel-wise correlation study, we developed a novel method based on image segmentation, inter-subject image registration, and robust supervoxel-based correlation analysis to achieve a statistical association study between the images and age. We evaluated the registration methodology in terms of the Dice coefficient for the heart chambers and myocardium, and the inverse consistency of the transformations, showing that the method works well in most cases with high overlap and inverse consistency. In a sex-stratified study conducted on a subset of n = 1388 images from the SCAPIS study, the supervoxel-wise analysis was able to find localized associations with age outside of the commonly segmented and analyzed sub-regions, and several substantial differences between the sexes in the association of age and volume.

The UK Biobank is a large-scale study collecting whole-body MR imaging and non-imaging health data. Robust and accurate inter-subject image registration of these whole-body MR images would enable their body-wide spatial standardization, and region-/voxel-wise correlation analysis of non-imaging data with image-derived parameters (e.g., tissue volume or fat content).

We propose a sex-stratified inter-subject whole-body MR image registration approach that uses subcutaneous adipose tissue- and muscle-masks from the state-of-the-art VIBESegmentator method to augment intensity-based graph-cut registration. The proposed method (that we refer to as mask-supported) was evaluated on a subset of 4000 subjects by comparing it to an intensity-only method as well as two previously published registration methods, uniGradICON and MIRTK. The evaluation consisted of a comparison of Jacobian Determinant (JD) folding frequency, Dice scores, and voxel-wise label error frequency calculated from the 71 VIBESegmentator masks. The 40 masks from MRSegmentator and 50 masks from TotalSegmentator were also used for independent Dice score evaluations. Additionally, voxel-wise correlation between age and each of fat content and tissue volume was studied to exemplify the usefulness for medical research.

The proposed method showed 7percentage points (pp) / 11pp lower frequency of JD folding for males / females when compared to the intensity-based method. The mask-supported method exhibited a mean Dice score of 0.773 / 0.744 across the cohort when evaluated on all VIBESegmentator masks, excluding the two used in the registration, for males / females, respectively. In comparison to the intensity-only registration, the mean values were 6 pp higher for both sexes, and the label error frequency was decreased in most tissue regions. These differences were 9pp / 8pp against uniGradICON and 12pp / 13pp against MIRTK. The mask-supported method achieved a mean Dice score of 0.736 / 0.676 when evaluated with MRSegmentator and 0.683/ 0.617 when evaluated with TotalSegmentator, showing an increase that ranged between 0.7pp and 12pp from the other three methods. Using the proposed method, the age-correlation maps were less noisy and showed higher anatomical alignment.

In conclusion, the image registration method using two tissue masks improves whole-body registration of UK Biobank images.

Background

Carotid and coronary atherosclerosis are critical determinants of cardiovascular risk, yet their interrelationship in middle-aged populations is incompletely understood. This study assessed carotid plaque composition, risk-factor associations, coronary disease, and sex differences in a subclinical cohort.

Methods

Within the Swedish CArdioPulmonary bioImage Study, 533 asymptomatic individuals aged 50–64 years with carotid plaque ≥2.7 mm on ultrasound underwent 3T multi-contrast carotid cardiovascular magnetic resonance (CMR) and coronary computed tomography angiography. Carotid plaque characteristics were determined manually using established criteria on multi-contrast weighted carotid CMR. Bayesian regression models evaluated associations between cardiovascular risk factors and coronary atherosclerosis.

Results

Lipid-rich necrotic core (LRNC) was present in 60% (320/533) and intraplaque hemorrhage (IPH) in 5.4% (29/533); calcification occurred in 48.6% (259/533). Maximum carotid wall thickness was 1.8 (1.6-2.0) mm, and mean lumen area 31.3 (26.7-36.1) mm². Coronary atherosclerosis was present in 63.6% (339/533) of participants, with ≥50% stenosis in 12.9% (69/533), and coronary artery calcium score >400 in 12.8% (68/533). Men (N = 367) had larger carotid lumen area, mean wall area, and maximum wall thickness (all p < 0.001) than women (N = 166), differences that persisted after body-surface-area adjustment (all p < 0.01). LRNC was present in 66% (242/367) of men compared to 47% (78/166) of women (p < 0.001). LRNC presence was not associated with coronary atherosclerosis, whereas IPH was associated with coronary involvement.

Conclusion

In middle-aged individuals, distinct cardiovascular risk factors were positively linked to the presence and volume of LRNC and calcified plaques. The substantial prevalence of high-risk plaque features, particularly LRNC and especially in men, highlights a significant subclinical carotid disease burden.

BACKGROUND/PURPOSE: To evaluate the advantages of including versus excluding the tumor periphery and combining diffusion-weighted imaging (DWI) with T2-weighted imaging (T2w) for outcome predictions of preoperative radio(chemo)therapy in rectal cancer.

METHODS: Four analysis strategies, based on two segmentation methods and two magnetic resonance imaging (MRI) sequences, were evaluated in 106 patients examined with pretreatment MRI. One segmentation method included the tumor periphery in the region of interest (ROI) encompassing the whole tumor (wROI), considered as the reference segmentation approach, and one included only the central part (cROI). Relevant radiomics imaging features were extracted from either T2w alone or from both T2w and DWI and used by a machine learning algorithm for the prediction of pathologic complete response (pCR), neoadjuvant rectal (NAR) score, and disease recurrence. The area under the curve (AUC) was the performance measure. AUCs were compared with a bootstrapping method based on 10⁴ bootstraps.

RESULTS: cROI applied to both T2w and DWI provided the highest numerical prediction of pCR (AUC 0.76), however, not significantly superior to the other strategies (p ≥ 0.138). cROI applied to both T2w and DWI also yielded the highest numerical prediction of NAR score (AUC 0.84), showing advantages over wROI-based analysis strategies (AUC 0.66 and 0.69; p ≤ 0.008). When compared to cROI applied to T2w alone (AUC 0.73), the benefit was borderline statistically significant (p = 0.053). For prediction of disease recurrence, no differences were found between the analysis strategies.

CONCLUSIONS: Inclusion of the tumor periphery in radiomic analysis of magnetic resonance images does not improve predictions of the preoperative therapy response in patients with rectal cancer. Excluding tumor periphery while adding DWI to T2w improves prediction of the NAR score, although it does not affect pCR or recurrence prediction.

Background: Accurate prediction of clinical outcomes is challenging yet important for patient care. The aim of the study was to evaluate a deep learning-based methodology using tissue-wise information, as a proof of concept, for predicting parameters known to be associated with clinical outcomes.

Methods: We utilized the publicly available autoPET cohort, consisting of 1014 FDG-PET/CT examinations. Tissue-wise projections were extracted, representing specific tissues (bone, lean tissue, adipose tissue, and air) at different angles. A deep regression and classification framework was trained to predict total metabolic tumor volume (TMTV), lesion count, patient age, sex, and diagnosis status (cancer vs. no cancer). Saliency analysis was performed to identify image regions contributing most to each prediction.

Results: Here we show that the best model predicts TMTV (MAE = 77 ml; R² = 0.84; (p <0.05)) and lesion count (MAE = 5.18, R² = 0.90), when using all tissue-wise projections. Age prediction improves when multiple projection angles are included (MAE = 6.57 years; R² = 0.70 (p <0.05)). The model also predicts sex (AUC = 1.00 (p <0.05)) and diagnosis status with high accuracy (AUC = 0.95 (p <0.05)).

Conclusions: This proof-of-concept study demonstrates that tissue-wise projections can be used for efficient and automated prediction of parameters related to clinical outcomes, highlighting their potential for future prediction of clinical outcomes.

Background: This study aimed to evaluate whether early renal and bone marrow effects - previously identified through laboratory tests - can be detected using MRI and ¹⁸F-FLT PET-MR in neuroendocrine tumor patients treated with ¹⁷⁷Lu-DOTATATE. Despite known early changes, precise clinical tools to monitor these effects remain limited.

Methods: Ten patients with metastatic NETs, scheduled to initiate at least four treatment cycles of ¹⁷⁷Lu-DOTATATE, were included. MRI was performed at all designated treatment time points, whereas F-FLT PET was acquired only at baseline and on day 2 following treatment initiation. Bone marrow (BM) fat conversion, proliferative activity, and renal effects were studied with MRI and ¹⁸F-FLT PET-MR. Laboratory tests for hematology and kidney and liver function were taken.

Results: Six patients completed the full imaging protocol with MRI, of whom ¹⁸F-FLT PET-MR examinations were performed in two patients. All subjects demonstrated increased BM fat content during the study interval, with a mean of +12.8%. The magnitude of the increase varied substantially, from +3.8% to +23.5%. There was a strong correlation between increased BM fat content and reduced peripheral blood cell counts. Decreased FLT uptake was found on day 2 after treatment. An increase in renal arterial flow and parenchymal volume was detected during the first week after treatment. No sustained renal effects were observed.

Conclusions: The hematological effects observed with laboratory tests after ¹⁷⁷Lu-DOTATATE treatment in neuroendocrine tumor patients can be detected by MRI as an increase in bone marrow fat content. This finding was supported by decreased FLT uptake on ¹⁸F-FLT PET-MR, indicating decreased bone marrow cell proliferation. MRI can also identify renal effects associated with this therapy.

Membrane-bound neuropilin-1 (Nrp-1) has been implicated as a regulator of insulin resistance-related lipid metabolism, but little is known about its circulating form, soluble Nrp-1 (sNrp-1), in a metabolic context. Here we investigated the association between plasma levels of sNrp-1 and metabolic variables in 501 participants from the Prospective investigation of Obesity, Energy and Metabolism (POEM) study. Metabolic phenotyping included blood sampling, dual energy X-ray absorptiometry scan, magnetic resonance imaging for fat distribution, and a glucose tolerance test. Associations were examined using linear regression models and Spearman's rank correlation. Plasma sNrp-1 levels were inversely associated with levels of total fatty acids, triglycerides and lactate, and positively associated with markers of insulin sensitivity. In human Simpson-Golabi-Behmel Syndrome (SGBS) adipocytes, we showed that treatment with recombinant Nrp-1 reduced phosphorylation of hormone-sensitive lipase in the basal state and attenuated isoproterenol-stimulated glycerol release, a measure of total lipolytic activity. Based on these findings, we thus postulate that sNrp-1-mediated suppression of lipolysis could explain our observed association between plasma sNrp-1 and a favorable lipid profile in humans. Further studies are warranted to establish causality and to clarify the role of sNrp-1 in promoting insulin sensitivity in human adipose tissue.

Purpose

Automated medical image-based prediction of clinical outcomes, such as overall survival (OS), has great potential in improving patient prognostics and personalized treatment planning. We developed a deep regression framework using tissue-wise FDG-PET/CT projections as input, along with a temporal input representing a scalar time horizon (in days) to predict OS in patients with Non-Small Cell Lung Cancer (NSCLC).

Methods

The proposed framework employed a ResNet-50 backbone to process input images and generate corresponding image embeddings. The embeddings were then combined with temporal data to produce OS probabilities as a function of time, effectively parameterizing the predictions based on time. The overall framework was developed using the U-CAN cohort (n = 556) and evaluated by comparing with a baseline method on the test set (n = 292). The baseline utilized the ResNet-50 architecture, processing only the images as input and providing OS predictions at pre-specified intervals, such as 2 or 5 year.

Results

The incorporation of temporal data with image embeddings demonstrated an advantage in predicting OS, outperforming the baseline method with an improvement in AUC of 4.3%. The proposed model using clinical + IDP features achieved strong performance, and an ensemble of imaging and clinical + IDP models achieved the best overall performance (0.788), highlighting the complementary value of multimodal inputs. The proposed method also enabled risk stratification of patients into distinct categories (high vs. low risk). Heat maps from the saliency analysis highlighted tumor regions as key structures for the prediction.

Conclusion

Our method provided an automated framework for predicting OS as a function of time, and demonstrates the potential of combining imaging and tabular data for improved survival prediction.

Breast cancer responds heterogeneously to treatment. This study evaluated the efficacy of image registration-based lesion tracking for assessing lesion-wise response in metastatic breast cancer (MBC). [18F]FDG PET/CT images from 15 patients with MBC from the multi-center IMPACT-MBC study were analyzed. Manual segmentation identified 644 lesions in baseline scans and 804 lesions in 2-week post-treatment scans. A deformable image registration method, validated for whole-body PET/CT scans of patients with cancer, was applied. Lesion tracking was performed using spatial overlap resulting from within-subject image registration, yielding a precision of 0.93 ± 0.07, a sensitivity of 0.87 ± 0.17, and an F1 score of 0.89 ± 0.10. Performance was consistent across lesion sites and lesion count-based disease burden. Sensitivity decreased for lesions under 1 mL. Image registration direction did not affect results. Between-subject registration enabled quantitative visualization and analysis of metabolic response patterns in bone lesions across the cohort. Results demonstrate the potential for automated, accurate lesion-wise response assessment, which may improve treatment monitoring in MBC.

BACKGROUND: Obesity has been associated with onset and progression of chronic kidney disease (CKD) but causal relationship remains uncertain. This study investigated how obesity causally affects estimated glomerular filtration rate.

METHODS: Cross-sectional and magnetic resonance imaging (MRI) data analyses were performed within the Prospective Investigation of Obesity, Energy, and Metabolism (POEM) study (502 participants, all aged 50 years). Additionally Mendelian randomization was performed using published summary data. Outcomes were creatinine- and cystatin C-based eGFR. Body mass index (BMI) and waist circumference (WC) were used as exposure variables in the cross-sectional and Mendelian randomization analyses. In the imaging data analyses, eGFR was regressed non-parametrically on tissue volume for each 3D voxel and visualized as a correlation "Imiomics" map.

RESULTS: Negative correlations were shown between cystatin C-based eGFR and BMI [beta = -0.190 (95% CI: -0.280 to -0.100)] and WC [beta = -0.160 (95% CI: -0.250 to -0.060)] in an adjusted model. In contrast, a positive association was found for creatinine-based eGFR [BMI beta = 1.20 (95% CI: 0.030 to 0.210) and WC beta = 0.160 (95% CI: 0.070 to 0.260)]. Similar patterns were found using MRI analysis (Imiomics map). Mendelian randomization implied a negative causal effect of obesity-related measures on cystatin C-based eGFR [BMI beta = -0.031 (95% CI: -0.037 to -0.026) and WC beta = -0.038 (95% CI: -0.045 to -0.031)], but no statistically significant effect was found for creatinine-based eGFR.

CONCLUSION: This study suggests a causal negative effect of obesity on cystatin C-based, but not creatinine-based eGFR. These findings warrant further research regarding estimations of kidney function when assessing obesity and CKD.