Accurate polyp segmentation from colonoscopy images is critical for colorectal cancer prevention, yet the generalization of deep learning models under domain shift remains insufficiently explored. We propose Boundary-Explicit Guided Attention U-Net (BEGA-UNet), a boundary-aware segmentation architecture that introduces explicit edge modeling as a structural inductive bias to enhance both segmentation accuracy and cross-domain robustness. The framework integrates three components: an Edge-Guided Module (EGM) with learnable Sobel-initialized operators to capture boundary cues, a Dual-Path Attention (DPA) module that processes channel and spatial attention in parallel, and a Multi-Scale Feature Aggregation (MSFA) module to encode contextual information across multiple receptive fields. Evaluated on the combined Kvasir-SEG and CVC-ClinicDB benchmarks, BEGA-UNet achieves 88.53% Dice and 82.51% IoU, outperforming representative convolutional and transformer-based baselines. More importantly, cross-dataset evaluation demonstrates strong robustness under domain shift, with BEGA-UNet retaining 83.2% of its in-distribution performance--substantially higher than U-Net (64.5%), Attention U-Net (47.5%), and TransUNet (53.1%). In a zero-shot setting on an entirely unseen dataset, the model further maintains 72.6% performance retention. Comprehensive ablation studies indicate that explicit boundary modeling plays a central role in improving generalization, while multi-scale context aggregation further stabilizes performance across domains. Feature distribution analyses support this observation by showing that edge-oriented representations exhibit markedly reduced cross-domain variability compared to appearance-driven features. Overall, BEGA-UNet provides an effective and interpretable solution for robust polyp segmentation, demonstrating that explicit boundary modeling serves as a critical inductive bias for ensuring reliability under clinical domain shifts.
Strong non-invasive tests (NITs) are needed to predict decompensation in patients with compensated advanced chronic liver disease (cACLD) and improve personalized patient care. We conducted a comprehensive review of the studies evaluating the effectiveness of NITs in predicting decompensation or liver-related death in patients with cACLD. A literature search was conducted in the PubMed database up to August 2025. Prospective or retrospective studies that included patients with cACLD and evaluated NITs for predicting decompensation or death or liver transplantation were included. Studies evaluating elastography and blood-based tests were analysed separately. The majority of studies assessed liver stiffness measurement (LSM), primarily using transient elastography (TE-LSM). There is a strong association between higher LSM values and an increased risk of decompensation, allowing classification of patients at low risk of decompensation from those with a higher risk. However, none of the studies reported data calibration, thereby limiting the ability to accurately predict the individual risk of decompensation. Higher FIB-4, ELF, and MELD values have been associated with an increased occurrence of decompensation. However, their performance was modest, with an area under the curve (AUC) below 0.75. Innovative approaches to improving the non-invasive prediction of decompensation may include levels of extracellular vesicles, genetic polymorphisms, or imaging-derived variables. Furthermore, since decompensation is the result of numerous interacting factors, artificial intelligence has the potential to improve the clinical relevance of predictive models by incorporating and processing a high-dimensional set of variables that reflect the underlying pathophysiological complexity.
Oligometastatic prostate cancer (oligoPCa) represents a clinical state of limited metastatic spread in which metastasis-directed therapy (MDT) may offer meaningful disease control either alone or with systemic therapy. As imaging, systemic therapy, and biologic characterization evolve, management strategies for both synchronous and metachronous presentations continue to undergo significant refinement.
AI-based image analysis is increasingly applied in pathology. Excluding fungal elements in PAS-stained skin sections is labor-intensive and well suited for AI assistance. While fungal detection in nails has been studied, skin-biopsy applications and real-world benefit remain limited. This study aimed to develop an AI algorithm for fungal detection in skin histology, create intuitive visualization software, and assess its diagnostic utility for pathologists.
To systematically evaluate the contribution of clinical heterogeneity to chronic kidney disease (CKD) progression in gout patients using data-driven phenotyping, and to assess whether incorporating genetic risk improves prediction of renal outcomes.
One of the most common consequences in individuals with diabetes is the diabetic foot, which can cause foot ulcers and even lead to limb amputation. Since an increase of the temperature in the plantar region is directly correlated with an increased risk of ulceration, infrared thermography (IRT) has been used in multiple studies as an automatic tool for detecting problems in diabetic foot. Artificial intelligence-based computer-aided diagnosis systems are being more frequently used to improve decision-making and minimize errors. These technologies are designed to increase examination accuracy, consistency in image interpretation, prognosis evaluation support, and examination accuracy. They also have the ability to offer insightful information and help medical professionals to manage diabetic foot issues successfully.
Artificial intelligence research in orthopedics has grown rapidly, yet a substantial gap remains between technical development and clinical translation. This narrative review summarizes current applications of artificial intelligence in orthopedic practice and highlights barriers to implementation.
Lateral gene transfer (LGT) has contributed to the genetic makeup of various eukaryotic lineages, yet its prevalence and long-term significance remain poorly understood, particularly for transfers between eukaryotes. Here, we investigate LGT across 29 species of Rhizaria, an ancient and ecologically diverse clade of predominantly free-living, single-celled phagotrophs. Using phylogenetic analyses of over 40,000 gene families complemented by machine learning-based prediction, we estimate that 8-20% of protein-coding genes in contemporary rhizarian genomes were acquired through LGT at various points during their billion-year history, with ~2,000 transfer events shared between at least two species across the rhizarian tree of life. Gene duplications outnumber LGTs across most lineages, yet LGT-derived genes themselves duplicate more frequently than vertically inherited ones, amplifying the genomic impact of each transfer event. Notably, transfers from other eukaryotes outnumber those from prokaryotes and show distinct signatures: prokaryote-derived LGTs are enriched among extracellular proteins, whereas eukaryote-derived LGTs are overrepresented in nuclear and informational processes. Prokaryote-derived LGT genes progressively acquire introns over evolutionary time, confirming their genomic integration and long-term retention. Our findings establish LGT as a pervasive force in rhizarian genome evolution and highlight eukaryote-to-eukaryote transfer as a substantial but often overlooked component of eukaryotic genetic innovation.
Deciphering the mechanisms of protein-protein interactions (PPIs) and protein structural changes within the native cellular environment is crucial for advancing drug discovery. In-vivo chemical cross-linking coupled with mass spectrometry (XL-MS) captures weak, transient and higher-order interactions that are often dysregulated under altered physiological conditions and remain challenging to detect using conventional methods. Applications of in-vivo XL-MS range from targeted mapping of PPIs to large-scale identification of interactome networks within the cells. The integration of quantitative approaches further facilitates comparison across different physiological conditions. The recent incorporation of machine learning (ML) tools into XL-MS workflows is transforming the depth and efficiency of this technology. AI-driven algorithms now enable more accurate identification of cross-linked peptides and the mapping of interaction topologies. Furthermore, the synergistic coupling of in-vivo XL-MS data with AI-assisted structural modeling platforms such as AlphaFold allows dynamic and high-throughput prediction of protein networks. This review discusses the broader applications of in-vivo XL-MS in complex biological samples, ranging from organelles and cells to whole tissues and highlights how AI integration is expanding structural biology toward a systems-level understanding of proteome architecture.
