PurposeWe report on a deep learning-based approach to the segmentation of geographic atrophy (GA) in patients with advanced age-related macular degeneration (AMD). MethodThree-dimensional (3D) optical coherence tomography (OCT) data was collected from two instruments at two different retina practices. This totaled 367 and 348 volumes, respectively, of routinely collected clinical data. For all data, the accuracy of a 3D-to-2D segmentation model was assessed relative to ground-truth manual labeling. ResultsDice Similarity Scores (DSC) averaged 0.824 and 0.826 for each data set. Correlations (r2) between manual and automated areas were 0.883 and 0.906, respectively. The inclusion of near Infra-red imagery as an additional information channel to the algorithm did not notably improve performance. ConclusionAccurate assessment of GA in real-world clinical OCT data can be achieved using deep learning. In the advent of therapeutics to slow the rate of GA progression, reliable, automated assessment is a clinical objective and this work validates one such method.

The success of machine learning algorithms is inherently related to the extraction of meaningful features, as they play a pivotal role in the performance of these algorithms. Central to this challenge is the quality of data representation. However, the ability to generalize and extract these features effectively from unseen datasets is also crucial. In light of this, we introduce a novel concept: the polyrepresentation. Polyrepresentation integrates multiple representations of the same modality extracted from distinct sources, for example, vector embeddings from the Siamese Network, self-supervised models, and interpretable radiomic features. This approach yields better performance metrics compared to relying on a single representation. Additionally, in the context of X-ray images, we demonstrate the transferability of the created polyrepresentation to a smaller dataset, underscoring its potential as a pragmatic and resource-efficient approach in various image-related solutions. It is worth noting that the concept of polyprepresentation on the example of medical data can also be applied to other domains, showcasing its versatility and broad potential impact.

Deep neural networks are increasingly applied for automated histopathology. Yet, whole-slide images (WSIs) are often acquired at gigapixel sizes, rendering it computationally infeasible to analyze them entirely at high resolution. Diagnostic labels are largely available only at the slide-level, because expert annotation of images at a finer (patch) level is both laborious and expensive. Moreover, regions with diagnostic information typically occupy only a small fraction of the WSI, making it inefficient to examine the entire slide at full resolution. Here, we propose SASHA -- {\it S}equential {\it A}ttention-based {\it S}ampling for {\it H}istopathological {\it A}nalysis -- a deep reinforcement learning approach for efficient analysis of histopathological images. First, SASHA learns informative features with a lightweight hierarchical, attention-based multiple instance learning (MIL) model. Second, SASHA samples intelligently and zooms selectively into a small fraction (10-20\%) of high-resolution patches, to achieve reliable diagnosis. We show that SASHA matches state-of-the-art methods that analyze the WSI fully at high-resolution, albeit at a fraction of their computational and memory costs. In addition, it significantly outperforms competing, sparse sampling methods. We propose SASHA as an intelligent sampling model for medical imaging challenges that involve automated diagnosis with exceptionally large images containing sparsely informative features.

The aim of the present article is to enrich the comprehension of iterative magnetic resonance imaging (MRI) reconstructions, including compressed sensing (CS) and iterative deep learning (DL) reconstructions, by describing them in the general framework of finite-dimensional inner-product spaces. In particular, we show that image-space preconditioning (ISP) and data-space preconditioning (DSP) can be formulated as non-conventional inner-products. The main gain of our reformulation is an embedding of ISP in the variational formulation of the MRI reconstruction problem (in an algorithm-independent way) which allows in principle to naturally and systematically propagate ISP in all iterative reconstructions, including many iterative DL and CS reconstructions where preconditioning is lacking. The way in which we apply linear algebraic tools to MRI reconstructions as presented in this article is a novelty. A secondary aim of our article is to offer a certain didactic material to scientists who are new in the field of MRI reconstruction. Since we explore here some mathematical concepts of reconstruction, we take that opportunity to recall some principles that may be understood for experts, but which may be hard to find in the literature for beginners. In fact, the description of many mathematical tools of MRI reconstruction is fragmented in the literature or sometimes missing because considered as a general knowledge. Further, some of those concepts can be found in mathematic manuals, but not in a form that is oriented toward MRI. For example, we think of the conjugate gradient descent, the notion of derivative with respect to non-conventional inner products, or simply the notion of adjoint. The authors believe therefore that it is beneficial for their field of research to dedicate some space to such a didactic material.

Visual question answering (VQA) in medical imaging aims to support clinical diagnosis by automatically interpreting complex imaging data in response to natural language queries. Existing studies typically rely on distinct visual and textual encoders to independently extract features from medical images and clinical questions, which are subsequently combined to generate answers. Specifically, in computed tomography (CT), such approaches are similar to the conventional practices in medical image analysis. However, these approaches pay less attention to the spatial continuity and inter-slice correlations in the volumetric CT data, leading to fragmented and imprecise responses. In this paper, we propose a novel large language model (LLM)-based framework enhanced by a graph representation of salient features. Different from conventional multimodal encoding strategies, our approach constructs a cross-modal graph integrating both visual and textual features, treating individual CT slices and question tokens as nodes within the graph. We further leverage an attentive graph convolutional network to dynamically fuse information within this structure. The resulting aggregated graph features then serve as a soft prompt to guide a large language model in generating accurate answers. Extensive experiments on the M3D-VQA benchmark demonstrate that our approach consistently outperforms baselines across multiple evaluation metrics, offering more robust reasoning capabilities.

Photorefractive Keratectomy (PRK) is a widely used laser-assisted refractive surgical technique. In some cases, it leads to temporary subepithelial inflammation or fibrosis linked to visual haze. There are to our knowledge no physics based and quantitative tools to monitor these symptoms. We here present a comprehensive machine learning-based algorithm for the detection of fibrosis based on spectral domain optical coherence tomography images recorded in vivo on standard clinical devices. Because of the rarity of these phenomena, we trained the model on corneas presenting Fuchs dystrophy causing similar, but permanent, fibrosis symptoms, and applied it to images from patients who have undergone PRK surgery. Our study shows that the model output (probability of Fuchs dystrophy classification) provides a quantified and explainable indicator of corneal healing for post-operative follow-up.

BackgroundArtificial intelligence (AI) has potentially shown promise in interpreting ultrasound imaging through flexible pattern recognition and algorithmic learning, but implementation in clinical practice remains limited. This study aimed to investigate the current application of AI in prenatal ultrasounds to identify congenital anomalies, and to synthesise challenges and opportunities for the advancement of AI-assisted ultrasound diagnosis. This comprehensive analysis addresses the clinical translation gap between AI performance metrics and practical implementation in prenatal care. MethodsSystematic searches were conducted in eight electronic databases (CINAHL Plus, Ovid/EMBASE, Ovid/MEDLINE, ProQuest, PubMed, Scopus, Web of Science and Cochrane Library) and Google Scholar from inception to May 2025. Studies were included if they applied an AI-assisted ultrasound diagnostic tool to identify a congenital anomaly during pregnancy. This review adhered to PRISMA guidelines for systematic reviews. We evaluated study quality using the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) guidelines. FindingsOf 9,918 records, 224 were identified for full-text review and 20 met the inclusion criteria. The majority of studies (11/20, 55%) were conducted in China, with most published after 2020 (16/20, 80%). All AI models were developed as an assistive tool for anomaly detection or classification. Most models (85%) focused on single-organ systems: heart (35%), brain/cranial (30%), or facial features (20%), while three studies (15%) attempted multi-organ anomaly detection. Fifty percent of the included studies reported exceptionally high model performance, with both sensitivity and specificity exceeding 0.95, with AUC-ROC values ranging from 0.91 to 0.97. Most studies (75%) lacked external validation, with internal validation often limited to small training and testing datasets. InterpretationWhile AI applications in prenatal ultrasound showed potential, current evidence indicates significant limitations in their practical implementation. Much work is required to optimise their application, including the external validation of diagnostic models with clinical utility to have real-world implications. Future research should prioritise larger-scale multi-centre studies, developing multi-organ anomaly detection capabilities rather than the current single-organ focus, and robust evaluation of AI tools in real-world clinical settings.

Automatic medical image segmentation techniques are vital for assisting clinicians in making accurate diagnoses and treatment plans. Although the U-shaped network (U-Net) has been widely adopted in medical image analysis, it still faces challenges in capturing long-range dependencies, particularly in complex and textured medical images where anatomical structures often blend into the surrounding background. To address these limitations, a novel network architecture, called recursive transformer-based U-Net (ReT-UNet), which integrates recursive feature learning and transformer technology, is proposed. One of the key innovations of ReT-UNet is the multi-scale global feature fusion (Multi-GF) module, inspired by transformer models and multi-scale pooling mechanisms. This module captures long-range dependencies, enhancing the abstraction and contextual understanding of multi-level features. Additionally, a recursive feature accumulation block is introduced to iteratively update features across layers, improving the network's ability to model spatial correlations and represent deep features in medical images. To improve sensitivity to local details, a lightweight atrous spatial pyramid pooling (ASPP) module is appended after the Multi-GF module. Furthermore, the segmentation head is redesigned to emphasize feature aggregation and fusion. During the encoding phase, a hybrid pooling layer is employed to ensure comprehensive feature sampling, thereby enabling a broader range of feature representation and improving detailed information learning. Results: The proposed method has been evaluated through ablation experiments, demonstrating generally consistent performance across multiple trials. When applied to cardiac, pulmonary nodule, and polyp segmentation datasets, the method showed a reduction in mis-segmented regions. The experimental results suggest that the approach can improve segmentation accuracy and stability compared to competing state-of-the-art methods. Experimental findings highlight the superiority of the proposed ReT-UNet over related methods and demonstrate its potential for applications in medical image segmentation.

Speckle patterns in ultrasound images often obscure anatomical details, leading to diagnostic uncertainty. Recently, various deep learning (DL)-based techniques have been introduced to effectively suppress speckle; however, their high computational costs pose challenges for low-resource devices, such as portable ultrasound systems. To address this issue, EdgeSRIE, which is a lightweight hybrid DL framework for real-time speckle reduction and image enhancement in portable ultrasound imaging, is introduced. The proposed framework consists of two main branches: an unsupervised despeckling branch, which is trained by minimizing a loss function between speckled images, and a deblurring branch, which restores blurred images to sharp images. For hardware implementation, the trained network is quantized to 8-bit integer precision and deployed on a low-resource system-on-chip (SoC) with limited power consumption. In the performance evaluation with phantom and in vivo analyses, EdgeSRIE achieved the highest contrast-to-noise ratio (CNR) and average gradient magnitude (AGM) compared with the other baselines (different 2-rule-based methods and other 4-DL-based methods). Furthermore, EdgeSRIE enabled real-time inference at over 60 frames per second while satisfying computational requirements (< 20K parameters) on actual portable ultrasound hardware. These results demonstrated the feasibility of EdgeSRIE for real-time, high-quality ultrasound imaging in resource-limited environments.

Early and accurate detection of oral cancer plays a pivotal role in improving patient outcomes. This research introduces a custom-designed, 19-layer convolutional neural network (CNN) for the automated diagnosis of oral cancer using clinical images of the lips and tongue. The methodology integrates advanced preprocessing steps, including min-max normalization and histogram-based contrast enhancement, to optimize image features critical for reliable classification. The model is extensively validated on the publicly available Oral Cancer (Lips and Tongue) Images (OCI) dataset, which is divided into 80% training and 20% testing subsets. Comprehensive performance evaluation employs established metrics-accuracy, sensitivity, specificity, precision, and F1-score. Our CNN architecture achieved an accuracy of 99.54%, sensitivity of 95.73%, specificity of 96.21%, precision of 96.34%, and F1-score of 96.03%, demonstrating substantial improvements over prominent transfer learning benchmarks, including SqueezeNet, AlexNet, Inception, VGG19, and ResNet50, all tested under identical experimental protocols. The model's robust performance, efficient computation, and high reliability underline its practicality for clinical application and support its superiority over existing approaches. This study provides a reproducible pipeline and a new reference point for deep learning-based oral cancer detection, facilitating translation into real-world healthcare environments and promising enhanced diagnostic confidence.