Abstract
Gastric intestinal metaplasia (GIM) characterization is challenging for humans and AI models. Deep learning solutions for this task are sensitive to training data, which is particularly concerning given the wide range of acquisition conditions, sampling biases, and overall scarcity of high-quality datasets.In this paper, we set forth the GIM self-similarity hypothesis where we assume that an underlying stationary self-similar process governs the structural changes observed in the mucosa. To validate this hypothesis we show that a deep learning model can map an adequately placed patch to the endoscopic grading of GIM (EGGIM) of the entire still frame.To evaluate our approach, we collected and annotated both retrospective and prospective datasets with EGGIM scores. Our results are promising: using leave-one-patient-out cross-validation, the predictions from a ResNet-50 model can be used to correctly stratify the risk for 57 out of 65 patients with perfect sensitivity on an extremely biased dataset.

Abstract

Abstract
Cardiovascular diseases are one of the main causes of death in the world. The predominant form of cardiovascular disease is coronary artery disease. Coronary artery calcium scanning is a non-contrast computed tomography exam that is considered the most reliable predictor of coronary events. Deep learning models have been developed for the segmentation of coronary artery calcium but the results have limited interpretability due to the black-box nature of these models. This work proposes an image retrieval pipeline based on a supervised contrastive framework that is capable of enhancing this interpretability by providing similar visual examples of coronary calcifications. In the COCA dataset, it is shown that this retrieval presents a label precision of 0.944 ± 0.230 regarding artery labels of retrieved images, with moderate similarity in terms of calcification area and Agatston score. It is also shown that the retrieval can be used to correct a deep CAC segmentation model by passing predictions from a segmentation model through the retrieval system, improving robustness and explainability.Clinical relevance- This study enhances CAC segmentation through image retrieval, improving both explainability and artery-specific labeling. By providing clinicians with more interpretable and anatomically accurate results, our approach aims to increase confidence in AI-assisted diagnostics leading to better-informed clinical decision-making in coronary artery disease diagnosis.

Abstract
This paper explores a Cycle-GAN architecture based on diffusion models for translating cardiac CT images with and without contrast, aiming to enhance the quality and accuracy of medical imaging. The combination of GANs and diffusion models has demonstrated promising results, particularly in generating high-quality, visually similar contrast-enhanced cardiac images. This effectiveness is evidenced by metrics such as a PSNR of 32.85, an SSIM of 0.766, and an FID of 42.348, highlighting the model’s capability for accurate and detailed image generation. Although these results indicate substantial potential for improving diagnostic accuracy, challenges remain, particularly concerning the generation of image artefacts and brightness inconsistencies, which could affect the clinical validation of these images. These issues have important implications for the reliability of the images in real medical diagnoses. The results of this study suggest that future research should focus on optimizing these aspects, improving the handling of artefacts, and investigating alternative architectures further to enhance the quality and reliability of the generated images, ensuring their applicability in clinical settings © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Abstract
Background The Endoscopic Grading of Gastric Intestinal Metaplasia (EGGIM) classification correlates with histological assessment of gastric intestinal metaplasia and enables stratification of gastric cancer risk. We developed and evaluated an artificial intelligence (AI) approach for EGGIM estimation. Methods Two datasets (A and B) with 1280 narrow-band imaging images were used for per-image analysis. Still images with manually selected patches of 224 x 224 pixels, annotated by experts, were used. Dataset A was retrospectively collected from clinical routine; Dataset B (used for per-patient analysis) was prospectively collected and included 65 fully documented patients. To mimic clinical practice, a deep neural network classified image patches into three EGGIM classes (0, 1, 2) and calculated the total per-patient EGGIM score (0-10). Results On per-image analysis, an accuracy of 87% (95%CI 71%-100%) was obtained. Per-patient EGGIM estimation had an average error of 1.15 (out of 10) and showed 88% (95%CI 80%-96%) accurate clinical decisions for surveillance (EGGIM >= 5), with 85% (95%CI 75%-94%) specificity, no false negatives, and positive and negative predictive values of 62% (95%CI 32%-92%) and 100% (95%CI 100%-100%), respectively. Conclusions EGGIM was estimated with high accuracy using AI tools in endoscopic image analyses. Automated assessment of EGGIM may provide a greener strategy for gastric cancer risk stratification, prospective studies, and interventional trials.

Abstract