Content-based mammographic image retrieval systems require exact BIRADS categorical matching across five distinct classes, presenting significantly greater complexity than binary classification tasks commonly addressed in literature. Current medical image retrieval studies suffer from methodological limitations including inadequate sample sizes, improper data splitting, and insufficient statistical validation that hinder clinical translation. We developed a comprehensive evaluation framework systematically comparing CNN architectures (DenseNet121, ResNet50, VGG16) with advanced training strategies including sophisticated fine-tuning, metric learning, and super-ensemble optimization. Our evaluation employed rigorous stratified data splitting (50%/20%/30% train/validation/test), 602 test queries, and systematic validation using bootstrap confidence intervals with 1,000 samples. Advanced fine-tuning with differential learning rates achieved substantial improvements: DenseNet121 (34.79% precision@10, 19.64% improvement) and ResNet50 (34.54%, 19.58% improvement). Super-ensemble optimization combining complementary architectures achieved 36.33% precision@10 (95% CI: [34.78%, 37.88%]), representing 24.93% improvement over baseline and providing 3.6 relevant cases per query. Statistical analysis revealed significant performance differences between optimization strategies (p<0.001) with large effect sizes (Cohen's d>0.8), while maintaining practical search efficiency (2.8milliseconds). Performance significantly exceeds realistic expectations for 5-class medical retrieval tasks, where literature suggests 20-25% precision@10 represents achievable performance for exact BIRADS matching. Our framework establishes new performance benchmarks while providing evidence-based architecture selection guidelines for clinical deployment in diagnostic support and quality assurance applications.

Imaging inverse problems are commonly addressed by minimizing measurement consistency and signal prior terms. While huge attention has been paid to developing high-performance priors, even the most advanced signal prior may lose its effectiveness when paired with an ill-conditioned sensing matrix that hinders convergence and degrades reconstruction quality. In optimization theory, preconditioners allow improving the algorithm's convergence by transforming the gradient update. Traditional linear preconditioning techniques enhance convergence, but their performance remains limited due to their dependence on the structure of the sensing matrix. Learning-based linear preconditioners have been proposed, but they are optimized only for data-fidelity optimization, which may lead to solutions in the null-space of the sensing matrix. This paper employs knowledge distillation to design a nonlinear preconditioning operator. In our method, a teacher algorithm using a better-conditioned (synthetic) sensing matrix guides the student algorithm with an ill-conditioned sensing matrix through gradient matching via a preconditioning neural network. We validate our nonlinear preconditioner for plug-and-play FISTA in single-pixel, magnetic resonance, and super-resolution imaging tasks, showing consistent performance improvements and better empirical convergence.

Accurate estimation of intravoxel incoherent motion (IVIM) parameters from diffusion-weighted MRI remains challenging due to the ill-posed nature of the inverse problem and high sensitivity to noise, particularly in the perfusion compartment. In this work, we propose a probabilistic deep learning framework based on Deep Ensembles (DE) of Mixture Density Networks (MDNs), enabling estimation of total predictive uncertainty and decomposition into aleatoric (AU) and epistemic (EU) components. The method was benchmarked against non probabilistic neural networks, a Bayesian fitting approach and a probabilistic network with single Gaussian parametrization. Supervised training was performed on synthetic data, and evaluation was conducted on both simulated and an in vivo dataset. The reliability of the quantified uncertainties was assessed using calibration curves, output distribution sharpness, and the Continuous Ranked Probability Score (CRPS). MDNs produced more calibrated and sharper predictive distributions for the diffusion coefficient D and fraction f parameters, although slight overconfidence was observed in pseudo-diffusion coefficient D*. The Robust Coefficient of Variation (RCV) indicated smoother in vivo estimates for D* with MDNs compared to Gaussian model. Despite the training data covering the expected physiological range, elevated EU in vivo suggests a mismatch with real acquisition conditions, highlighting the importance of incorporating EU, which was allowed by DE. Overall, we present a comprehensive framework for IVIM fitting with uncertainty quantification, which enables the identification and interpretation of unreliable estimates. The proposed approach can also be adopted for fitting other physical models through appropriate architectural and simulation adjustments.

IntroductionStructural neuroimaging analyses require research quality images, acquired with costly MRI acquisitions. Isotropic (3D-T1) images are desirable for quantitative analyses, however a routine compromise in the clinical setting is to acquire anisotropic (2D-T1) analogues for qualitative visual inspection. ML (Machine learning-based) software have shown promise in addressing some of the limitations of 2D-T1 scans in research applications, yet their efficacy in quantitative research is generally poorly understood. Pathology-related abnormalities of the subcortical structures have previously been identified in idiopathic generalised epilepsy (IGE), which have been overlooked based on visual inspection, through the use of quantitative morphometric analyses. As such, IGE biomarkers present a suitable model in which to evaluate the applicability of image preprocessing methods. This study therefore explores subcortical structural biomarkers of IGE, first in our silver standard 3D-T1 scans, then in 2D-T1 scans that were either untransformed, resampled using a classical interpolation approach, or synthesised with a resolution and contrast agnostic ML model (the latter of which is compared to a separate model). Methods2D-T1 and 3D-T1 MRI scans were acquired during the same scanning session for 33 individuals with drug-responsive IGE (age mean 32.16 {+/-} SD = 14.20, male n = 14) and 42 individuals with drug-resistant IGE (31.76 {+/-} 11.12, 17), all diagnosed at the Walton Centre NHS Foundation Trust Liverpool, alongside 39 age- and sex-matched healthy controls (32.32 {+/-} 8.65, 16). The untransformed 2D-T1 scans were resampled into isotropic images using NiBabel (res-T1), and preprocessed into synthetic isotropic images using SynthSR (syn-T1). For the 3D-T1, 2D-T1, res-T1, and syn-T1 images, the recon-all command from FreeSurfer 8.0.0 was used to create parcellations of 174 anatomical regions (equivalent to the 174 regional parcellations provided as part of the DL+DiReCT pipeline), defined by the aseg and Destrieux atlases, and FSL run_first_all was used to segment subcortical surface shapes. The new ML FreeSurfer pipeline, recon-all-clinical, was also tested in the 2D-T1, 3D-T1, and res-T1 images. As a model comparison for SynthSR, the DL+DiReCT pipeline was used to provide segmentations of the 2D-T1 and res-T1 images, including estimates of regional volume and thickness. Spatial overlap and intraclass correlations between the morphometrics of the eight resulting parcellations were first determined, then subcortical surface shape abnormalities associated with IGE were identified by comparing the FSL run_first_all outputs of patients with controls. ResultsWhen standardised to the metrics derived from the 3D-T1 scans, cortical volume and thickness estimates trended lower for the 2D-T1, res-T1, syn-T1, and DL+DiReCT outputs, whereas subcortical volume estimates were more coherent. Dice coefficients revealed an acceptable spatial similarity between the cortices of the 3D-T1 scans and the other images overall, and was higher in the subcortical structures. Intraclass correlation coefficients were consistently lowest when metrics were computed for model-derived inputs, and estimates of thickness were less similar to the ground truth than those of volume. For the people with epilepsy, the 3D-T1 scans showed significant surface deflations across various subcortical structures when compared to healthy controls. Analysis of the 2D-T1 scans enabled the reliable detection of a subset of subcortical abnormalities, whereas analyses of the res-T1 and syn-T1 images were more prone to false-positive results. ConclusionsResampling and ML image synthesis methods do not currently attenuate partial volume effects resulting from low through plane resolution in anisotropic MRI scans, instead quantitative analyses using 2D-T1 scans should be interpreted with caution, and researchers should consider the potential implications of preprocessing. The recon-all-clinical pipeline is promising, but requires further evaluation, especially when considered as an alternative to the classical pipeline. Key PointsO_LISurface deviations indicative of regional atrophy and hypertrophy were identified in people with idiopathic generalised epilepsy. C_LIO_LIPartial volume effects are likely to attenuate subtle morphometric abnormalities, increasing the likelihood of erroneous inference. C_LIO_LIPriors in synthetic image creation models may render them insensitive to subtle biomarkers. C_LIO_LIResampling and machine-learning based image synthesis are not currently replacements for research quality acquisitions in quantitative MRI research. C_LIO_LIThe results of studies using synthetic images should be interpreted in a separate context to those using untransformed data. C_LI

Dynamic Positron Emission Tomography (PET) scans offer rich spatiotemporal data for detecting malignancies, but their high-dimensionality and noise pose significant challenges. We introduce a novel framework, the Equivariant Spatiotemporal Transformer with MDL-Guided Feature Selection (EST-MDL), which integrates group-theoretic symmetries, Kolmogorov complexity, and Minimum Description Length (MDL) principles. By enforcing spatial and temporal symmetries (e.g., translations and rotations) and leveraging MDL for robust feature selection, our model achieves improved generalization and interpretability. Evaluated on three realworld PET datasets--LUNG-PET, BRAIN-PET, and BREAST-PET--our approach achieves AUCs of 0.94, 0.92, and 0.95, respectively, outperforming CNNs, Vision Transformers (ViTs), and Graph Neural Networks (GNNs) in AUC, sensitivity, specificity, and computational efficiency. This framework offers a robust, interpretable solution for malignancy detection in clinical settings.

BackgroundLarge Language Models (LLMs) represent an ever-emerging and rapidly evolving generative artificial intelligence (AI) modality with promising developments in the field of medical education. LLMs can provide automated feedback services to medical trainees (i.e. medical students, residents, fellows, etc.) and possibly serve a role in medical imaging education. AimThis systematic review aims to comprehensively explore the current applications and educational outcomes of LLMs in providing automated feedback on medical imaging reports. MethodsThis study employs a comprehensive systematic review strategy, involving an extensive search of the literature (Pubmed, Scopus, Embase, and Cochrane), data extraction, and synthesis of the data. ConclusionThis systematic review will highlight the best practices of LLM use in automated feedback of medical imaging reports and guide further development of these models.

Retinal detachment (RD) is a vision-threatening condition that requires timely intervention to preserve vision. Macular involvement -- whether the macula is still intact (macula-intact) or detached (macula-detached) -- is the key determinant of visual outcomes and treatment urgency. Point-of-care ultrasound (POCUS) offers a fast, non-invasive, cost-effective, and accessible imaging modality widely used in diverse clinical settings to detect RD. However, ultrasound image interpretation is limited by a lack of expertise among healthcare providers, especially in resource-limited settings. Deep learning offers the potential to automate ultrasound-based assessment of RD. However, there are no ML ultrasound algorithms currently available for clinical use to detect RD and no prior research has been done on assessing macular status using ultrasound in RD cases -- an essential distinction for surgical prioritization. Moreover, no public dataset currently supports macular-based RD classification using ultrasound video clips. We introduce Eye Retinal DEtachment ultraSound, ERDES, the first open-access dataset of ocular ultrasound clips labeled for (i) presence of retinal detachment and (ii) macula-intact versus macula-detached status. The dataset is intended to facilitate the development and evaluation of machine learning models for detecting retinal detachment. We also provide baseline benchmarks using multiple spatiotemporal convolutional neural network (CNN) architectures. All clips, labels, and training code are publicly available at https://osupcvlab.github.io/ERDES/.

BackgroundNailfold capillaroscopy enables visualisation of structural abnormalities in the microvasculature of patients with systemic sclerosis (SSc). The objective of this feasibility study was to determine whether multispectral imaging could provide functional assessment (differences in haemoglobin concentration or oxygenation) of capillaries to aid discrimination between healthy controls and patients with SSc. MSI of nailfold capillaries visualizes the smallest blood vessels and the impact of SSc on angiogenesis and their deformation, making it suitable for evaluating oxygenation-sensitive imaging techniques. Imaging of the nailfold capillaries offers tissue-specific oxygenation information, unlike pulse oximetry, which measures arterial blood oxygenation as a single-point measurement. MethodsThe CAPoxy study was a single-centre, cross-sectional, feasibility study of nailfold capillary multispectral imaging, comparing a cohort of patients with SSc to controls. A nine-band multispectral camera was used to image 22 individuals (10 patients with SSc and 12 controls). Linear mixed-effects models and summary statistics were used to compare the different regions of the nailfold (capillaries, surrounding edges, and outside area) between SSc and controls. A machine learning model was used to compare the two groups. ResultsPatients with SSc exhibited higher indicators of haemoglobin concentration in the capillary and adjacent regions compared to controls, which were significant in the regions surrounding the capillaries (p<0.001). There were also spectral differences between the SSc and controls groups that could indicate differences in oxygenation of the capillaries and surrounding tissue. Additionally, a machine learning model distinguished SSc patients from healthy controls with an accuracy of 84%, suggesting potential for multispectral imaging to classify SSc based on structural and functional microvascular changes. ConclusionsData indicates that multispectral imaging differentiates between patients with SSc from controls based on differences in vascular function. Further work to develop a targeted spectral camera would further improve the contrast between patients with SSc and controls, enabling better imaging. Key messagesMultispectral imaging holds promise for providing functional oxygenation measurement in nailfold capillaroscopy. Significant oxygenation differences between individuals with systemic sclerosis and healthy controls can be detected with multispectral imaging in the tissue surrounding capillaries.

Accurate detection of anatomical landmarks in brain Magnetic Resonance Imaging (MRI) scans is essential for reliable spatial normalization, image alignment, and quantitative neuroimaging analyses. In this study, we introduce BrainSignsNET, a deep learning framework designed for robust three-dimensional (3D) landmark detection. Our approach leverages a multi-task 3D convolutional neural network that integrates an attention decoder branch with a multi-class decoder branch to generate precise 3D heatmaps, from which landmark coordinates are extracted. The model was trained and internally validated on T1-weighted Magnetization-Prepared Rapid Gradient-Echo (MPRAGE) scans from the Alzheimers Disease Neuroimaging Initiative (ADNI), the Baltimore Longitudinal Study of Aging (BLSA), and the Biomarkers of Cognitive Decline in Adults at Risk for AD (BIOCARD) datasets and externally validated on a clinical dataset from the Johns Hopkins Hydrocephalus Clinic. The study encompassed 14,472 scans from 6,299 participants, representing a diverse demographic profile with a significant proportion of older adult participants, particularly those over 70 years of age. Extensive preprocessing and data augmentation strategies, including traditional MRI corrections and tailored 3D transformations, ensured data consistency and improved model generalizability. Performance metrics demonstrated that on internal validation BrainSignsNET achieved an overall mean Euclidean distance of 2.32 {+/-} 0.41 mm and 94.8% of landmarks localized within their anatomically defined 3D volumes in the external validation dataset. This improvement in accurate anatomical landmark detection on brain MRI scans should benefit many imaging tasks, including registration, alignment, and quantitative analyses.

Cross-domain Few-shot Medical Image Segmentation (CD-FSMIS) is a potential solution for segmenting medical images with limited annotation using knowledge from other domains. The significant performance of current CD-FSMIS models relies on the heavily training procedure over other source medical domains, which degrades the universality and ease of model deployment. With the development of large visual models of natural images, we propose a training-free CD-FSMIS model that introduces the Multi-center Adaptive Uncertainty-aware Prompting (MAUP) strategy for adapting the foundation model Segment Anything Model (SAM), which is trained with natural images, into the CD-FSMIS task. To be specific, MAUP consists of three key innovations: (1) K-means clustering based multi-center prompts generation for comprehensive spatial coverage, (2) uncertainty-aware prompts selection that focuses on the challenging regions, and (3) adaptive prompt optimization that can dynamically adjust according to the target region complexity. With the pre-trained DINOv2 feature encoder, MAUP achieves precise segmentation results across three medical datasets without any additional training compared with several conventional CD-FSMIS models and training-free FSMIS model. The source code is available at: https://github.com/YazhouZhu19/MAUP.