The explosive growth of available high-quality imaging data coupled with new progress in hardware capabilities has enabled a new era of unprecedented performance in brain segmentation tasks. Despite the explosion of new data released by consortiums and groups around the world, most published, closed, or openly available segmentation models have either a limited or an unknown role in pediatric brains. This study explores the utility of state-of-the-art automated ventricular segmentation tools applied to pediatric hydrocephalus. Two popular, fast, whole-brain segmentation tools were used (FastSurfer and QuickNAT) to automatically segment the lateral ventricles and evaluate their accuracy in children with hydrocephalus. Forty scans from 32 patients were included in this study. The patients underwent imaging at the University of Minnesota Medical Center or satellite clinics, were between 0 and 18 years old, had an ICD-10 diagnosis that included the word hydrocephalus, and had at least one T1-weighted pre- or postcontrast MPRAGE sequence. Patients with poor quality scans were excluded. Dice similarity coefficient (DSC) scores were used to compare segmentation outputs against manually segmented lateral ventricles. Overall, both models performed poorly with DSCs of 0.61 for each segmentation tool. No statistically significant difference was noted between model performance (p = 0.86). Using a multivariate linear regression to examine factors associated with higher DSC performance, male gender (p = 0.66), presence of ventricular catheter (p = 0.72), and MRI magnet strength (p = 0.23) were not statistically significant factors. However, younger age (p = 0.03) and larger ventricular volumes (p = 0.01) were significantly associated with lower DSC values. A large-scale visualization of 196 scans in both models showed characteristic patterns of segmentation failure in larger ventricles. Significant gaps exist in current cutting-edge segmentation models when applied to pediatric hydrocephalus. Researchers will need to address these types of gaps in performance through thoughtful consideration of their training data before reaching the ultimate goal of clinical deployment.

This study aims to develop and validate a deep learning radiomics signature (DLRS) that integrates radiomics and deep learning features for the non-invasive prediction of microvascular invasion (MVI) in patients with colon cancer (CC). Furthermore, the study explores the potential association between DLRS and tumor immune heterogeneity. This study is a multi-center retrospective study that included a total of 1007 patients with colon cancer (CC) from three medical centers and The Cancer Genome Atlas (TCGA-COAD) database. Patients from Medical Centers 1 and 2 were divided into a training cohort (n = 592) and an internal validation cohort (n = 255) in a 7:3 ratio. Medical Center 3 (n = 135) and the TCGA-COAD database (n = 25) were used as external validation cohorts. Radiomics and deep learning features were extracted from contrast-enhanced venous-phase CT images. Feature selection was performed using machine learning algorithms, and three predictive models were developed: a radiomics model, a deep learning (DL) model, and a combined deep learning radiomics (DLR) model. The predictive performance of each model was evaluated using multiple metrics, including the area under the curve (AUC), sensitivity, and specificity. Additionally, differential gene expression analysis was conducted on RNA-seq data from the TCGA-COAD dataset to explore the association between the DLRS and tumor immune heterogeneity within the tumor microenvironment. Compared to the standalone radiomics and deep learning models, DLR fusion model demonstrated superior predictive performance. The AUC for the internal validation cohort was 0.883 (95% CI: 0.828-0.937), while the AUC for the external validation cohort reached 0.855 (95% CI: 0.775-0.935). Furthermore, stratifying patients from the TCGA-COAD dataset into high-risk and low-risk groups based on the DLRS revealed significant differences in immune cell infiltration and immune checkpoint expression between the two groups (P < 0.05). The contrast-enhanced CT-based DLR fusion model developed in this study effectively predicts the MVI status in patients with CC. This model serves as a non-invasive preoperative assessment tool and reveals a potential association between the DLRS and immune heterogeneity within the tumor microenvironment, providing insights to optimize individualized treatment strategies.

End-stage renal disease is characterized by an irreversible decline in kidney function. Despite a risk of chronic dysfunction of the transplanted kidney, renal transplantation is considered the most effective solution among available treatment options. Clinical attributes of graft survival prediction, such as allocation variables or results of pathological examinations, have been widely studied. Nevertheless, medical imaging is clinically used only to assess current transplant status. This study investigated the use of unsupervised deep learning-based algorithms to identify rich radiomic features that may be linked to graft survival from early dynamic contrast-enhanced magnetic resonance imaging data of renal transplants. A retrospective cohort of 108 transplanted patients (mean age 50 +/- 15, 67 men) undergoing systematic magnetic resonance imaging follow-up examinations (2013 to 2015) was used to train deep convolutional neural network models based on an unsupervised contrastive learning approach. 5-year graft survival analysis was performed from the obtained artificial intelligence radiomics features using penalized Cox models and Kaplan-Meier estimates. Using a validation set of 48 patients (mean age 54 +/- 13, 30 men) having 1-month post-transplantation magnetic resonance imaging examinations, the proposed approach demonstrated promising 5-year graft survival capability with a 72.7% concordance index from the artificial intelligence radiomics features. Unsupervised clustering of these radiomics features enabled statistically significant stratification of patients (p=0.029). This proof-of-concept study exposed the promising capability of artificial intelligence algorithms to extract relevant radiomics features that enable renal transplant survival prediction. Further studies are needed to demonstrate the robustness of this technique, and to identify appropriate procedures for integration of such an approach into multimodal and clinical settings.

We present a fully automated, anatomically guided deep learning pipeline for prostate cancer (PCa) risk stratification using routine MRI. The pipeline integrates three key components: an nnU-Net module for segmenting the prostate gland and its zones on axial T2-weighted MRI; a classification module based on the UMedPT Swin Transformer foundation model, fine-tuned on 3D patches with optional anatomical priors and clinical data; and a VAE-GAN framework for generating counterfactual heatmaps that localize decision-driving image regions. The system was developed using 1,500 PI-CAI cases for segmentation and 617 biparametric MRIs with metadata from the CHAIMELEON challenge for classification (split into 70% training, 10% validation, and 20% testing). Segmentation achieved mean Dice scores of 0.95 (gland), 0.94 (peripheral zone), and 0.92 (transition zone). Incorporating gland priors improved AUC from 0.69 to 0.72, with a three-scale ensemble achieving top performance (AUC = 0.79, composite score = 0.76), outperforming the 2024 CHAIMELEON challenge winners. Counterfactual heatmaps reliably highlighted lesions within segmented regions, enhancing model interpretability. In a prospective multi-center in-silico trial with 20 clinicians, AI assistance increased diagnostic accuracy from 0.72 to 0.77 and Cohen's kappa from 0.43 to 0.53, while reducing review time per case by 40%. These results demonstrate that anatomy-aware foundation models with counterfactual explainability can enable accurate, interpretable, and efficient PCa risk assessment, supporting their potential use as virtual biopsies in clinical practice.

Purpose: To evaluate the impact of harmonization and multi-region CT image feature integration on survival prediction in non-small cell lung cancer (NSCLC) patients, using handcrafted radiomics, pretrained foundation model (FM) features, and clinical data from a multicenter dataset. Methods: We analyzed CT scans and clinical data from 876 NSCLC patients (604 training, 272 test) across five centers. Features were extracted from the whole lung, tumor, mediastinal nodes, coronary arteries, and coronary artery calcium (CAC). Handcrafted radiomics and FM deep features were harmonized using ComBat, reconstruction kernel normalization (RKN), and RKN+ComBat. Regularized Cox models predicted overall survival; performance was assessed using the concordance index (C-index), 5-year time-dependent area under the curve (t-AUC), and hazard ratio (HR). SHapley Additive exPlanations (SHAP) values explained feature contributions. A consensus model used agreement across top region of interest (ROI) models to stratify patient risk. Results: TNM staging showed prognostic utility (C-index = 0.67; HR = 2.70; t-AUC = 0.85). The clinical + tumor radiomics model with ComBat achieved a C-index of 0.7552 and t-AUC of 0.8820. FM features (50-voxel cubes) combined with clinical data yielded the highest performance (C-index = 0.7616; t-AUC = 0.8866). An ensemble of all ROIs and FM features reached a C-index of 0.7142 and t-AUC of 0.7885. The consensus model, covering 78% of valid test cases, achieved a t-AUC of 0.92, sensitivity of 97.6%, and specificity of 66.7%. Conclusion: Harmonization and multi-region feature integration improve survival prediction in multicenter NSCLC data. Combining interpretable radiomics, FM features, and consensus modeling enables robust risk stratification across imaging centers.

Deep learning (DL)-based methods show promise in detecting pulmonary embolism (PE) on CT pulmonary angiography (CTPA), potentially improving diagnostic accuracy and workflow efficiency. This meta-analysis aimed to (1) determine pooled performance estimates of DL algorithms for PE detection; and (2) compare the diagnostic efficacy of convolutional neural network (CNN)- versus U-Net-based architectures. Following PRISMA guidelines, we searched PubMed and EMBASE through April 15, 2025 for English-language studies (2010-2025) reporting DL models for PE detection with extractable 2 × 2 data or performance metrics. True/false positives and negatives were reconstructed when necessary under an assumed 50 % PE prevalence (with 0.5 continuity correction). We approximated AUROC as the mean of sensitivity and specificity if not directly reported. Sensitivity, specificity, accuracy, PPV and NPV were pooled using a DerSimonian-Laird random-effects model with Freeman-Tukey transformation; AUROC values were combined via a fixed-effect inverse-variance approach. Heterogeneity was assessed by Cochran's Q and I². Subgroup analyses contrasted CNN versus U-Net models. Twenty-four studies (n = 22,984 patients) met inclusion criteria. Pooled estimates were: AUROC 0.895 (95 % CI: 0.874-0.917), sensitivity 0.894 (0.856-0.923), specificity 0.871 (0.831-0.903), accuracy 0.857 (0.833-0.882), PPV 0.832 (0.794-0.869) and NPV 0.902 (0.874-0.929). Between-study heterogeneity was high (I² ≈ 97 % for sensitivity/specificity). U-Net models exhibited higher sensitivity (0.899 vs 0.893) and CNN models higher specificity (0.926 vs 0.900); subgroup Q-tests confirmed significant differences for both sensitivity (p = 0.0002) and specificity (p < 0.001). DL algorithms demonstrate high diagnostic accuracy for PE detection on CTPA, with complementary strengths: U-Net architectures excel in true-positive identification, whereas CNNs yield fewer false positives. However, marked heterogeneity underscores the need for standardized, prospective validation before routine clinical implementation.

Medical image super-resolution (SR) is a classic regression task in low-level vision. Limited by hardware limitations, acquisition time, low radiation dose, and other factors, the spatial resolution of some medical images is not sufficient. To address this problem, many different SR methods have been proposed. Especially in recent years, medical image SR networks based on deep learning have been vigorously developed. This survey provides a modular and detailed introduction to the key components of medical image SR technology based on deep learning. In this paper, we first introduce the background concepts of deep learning and medical image SR task. Subsequently, we present a comprehensive analysis of the key components from the perspectives of effective architecture, upsampling module, learning strategy, and image quality assessment of medical image SR networks. Furthermore, we focus on the urgent problems that need to be addressed in the medical image SR task based on deep learning. And finally we summarize the trends and challenges of future development.

Background: Accurate MRI-based identification of extramural vascular invasion (EVI) and mesorectal fascia invasion (MFI) is pivotal for risk-stratified management of rectal cancer, yet visual assessment is subjective and vulnerable to inter-institutional variability. Purpose: To develop and externally evaluate a multicenter, foundation-model-driven framework that automatically classifies EVI and MFI on axial and sagittal T2-weighted MRI. Methods: This retrospective study used 331 pre-treatment rectal cancer MRI examinations from three European hospitals. After TotalSegmentator-guided rectal patch extraction, a self-supervised frequency-domain harmonization pipeline was trained to minimize scanner-related contrast shifts. Four classifiers were compared: ResNet50, SeResNet, the universal biomedical pretrained transformer (UMedPT) with a lightweight MLP head, and a logistic-regression variant using frozen UMedPT features (UMedPT_LR). Results: UMedPT_LR achieved the best EVI detection when axial and sagittal features were fused (AUC = 0.82; sensitivity = 0.75; F1 score = 0.73), surpassing the Chaimeleon Grand-Challenge winner (AUC = 0.74). The highest MFI performance was attained by UMedPT on axial harmonized images (AUC = 0.77), surpassing the Chaimeleon Grand-Challenge winner (AUC = 0.75). Frequency-domain harmonization improved MFI classification but variably affected EVI performance. Conventional CNNs (ResNet50, SeResNet) underperformed, especially in F1 score and balanced accuracy. Conclusion: These findings demonstrate that combining foundation model features, harmonization, and multi-view fusion significantly enhances diagnostic performance in rectal MRI.

Medical image segmentation is vital for clinical diagnosis, yet current deep learning methods often demand extensive expert effort, i.e., either through annotating large training datasets or providing prompts at inference time for each new case. This paper introduces a zero-shot and automatic segmentation pipeline that combines off-the-shelf vision-language and segmentation foundation models. Given a medical image and a task definition (e.g., "segment the optic disc in an eye fundus image"), our method uses a grounding model to generate an initial bounding box, followed by a visual prompt boosting module that enhance the prompts, which are then processed by a promptable segmentation model to produce the final mask. To address the challenges of domain gap and result verification, we introduce a test-time adaptation framework featuring a set of learnable adaptors that align the medical inputs with foundation model representations. Its hyperparameters are optimized via Bayesian Optimization, guided by a proxy validation model without requiring ground-truth labels. Our pipeline offers an annotation-efficient and scalable solution for zero-shot medical image segmentation across diverse tasks. Our pipeline is evaluated on seven diverse medical imaging datasets and shows promising results. By proper decomposition and test-time adaptation, our fully automatic pipeline performs competitively with weakly-prompted interactive foundation models.

Identification of bony landmarks in medical images is of high importance for 3D planning in orthopaedic surgery. Automated landmark identification has the potential to optimize clinical routines and allows for the scientific analysis of large databases. To the authors' knowledge, no direct comparison of different methods for automated landmark detection on the same dataset has been published to date. We compared 3 methods for automated femoral landmark identification: an artificial neural network, a statistical shape model and a geometric approach. All methods were compared against manual measurements of two raters on the task of identifying 6 femoral landmarks on CT data or derived surface models of 202 femora. The accuracy of the methods was in the range of the manual measurements and comparable to those reported in previous studies. The geometric approach showed a significantly higher average deviation compared to the manually selected reference landmarks, while there was no statistically significant difference for the neural network and the SSM. All fully automated methods show potential for use, depending on the use case. Characteristics of the different methods, such as the input data required (raw CT/segmented bone surface models, amount of training data required) and/or the methods robustness, can be used for method selection in the individual application.