Selected work

Representative Papers

PolypVision: A Three-Stage Hierarchical Deep Learning Framework for Classification and Segmentation of Colorectal Polyps

Aug 11, 2026

This study addresses the challenge of accurate detection, segmentation, and endoscopy–histology joint classification of colorectal polyps by proposing a three-stage hierarchical deep learning framework. The first stage employs EfficientNetV2-M for polyp type and morphology classification; the second stage utilizes UNet++ for high-precision segmentation and recommends resection strategies; and the final stage refines adenoma subtyping through transfer learning. Innovatively integrating a task-progressive architecture, task-specific loss functions (Focal Loss and Dice+BCE), and cross-stage knowledge transfer, the framework achieves device-agnostic fully automated analysis. Evaluated on PolypGen, Kvasir-SEG, and CVC-ClinicDB datasets, it attains an AUC of 0.99 and mAP@50 of 94.4%, matching or surpassing current state-of-the-art methods, and has been deployed as a publicly accessible web application.

0 citationsRead paper

GIM-ENDO: A Multimodal Endoscopic Image and Video Dataset for Gastric Intestinal Metaplasia Morphology and Pathology

Jun 18, 2026

This study addresses a critical gap in artificial intelligence research on gastric intestinal metaplasia (GIM) by introducing the first publicly available, pathology-validated, multimodal endoscopic dataset. The dataset integrates high-definition white-light endoscopy, narrow-band imaging (NBI), and magnifying NBI (M-NBI) images and videos from 24 patients (22 GIM-positive), comprehensively annotated with histopathological subtypes, OLGA/OLGIM staging, standardized endoscopic features, and full clinical metadata. By providing a rigorously curated benchmark resource, this work enables AI-driven real-time GIM detection, subtype classification, and phenotypic analysis, thereby filling a significant void in standardized multimodal data for GIM research.

0 citationsRead paper

A Hybrid Computational Intelligence Framework with Metaheuristic Optimization for Drug-Drug Interaction Prediction

Oct 08, 2025

Drug–drug interactions (DDIs) are a leading cause of preventable adverse drug events, making accurate DDI prediction critical for clinical medication safety. To address this, we propose a high-accuracy, interpretable DDI prediction framework: (1) dual molecular embeddings are constructed by integrating Mol2Vec and SMILES-BERT; (2) a rule-based clinical scoring metric (RBScore) is designed—requiring no interaction labels—to explicitly incorporate pharmacological knowledge; and (3) a three-stage metaheuristic optimization algorithm (RSmpl-ACO-PSO) automatically tunes hyperparameters of a lightweight classifier. Evaluated on the DrugBank dataset, our model achieves 0.911 ROC-AUC and 0.867 PR-AUC. Furthermore, it demonstrates strong generalizability in a real-world cohort of type 2 diabetes patients, validating its clinical applicability and robustness.

0 citationsRead paper

Attention Xception UNet (AXUNet): A Novel Combination of CNN and Self-Attention for Brain Tumor Segmentation

Mar 26, 2025

To address the challenge of precise multi-sequence (T1CE/T2/FLAIR) MRI segmentation for gliomas, this paper proposes AXUNet—a novel UNet-based architecture that integrates an Xception backbone with a large-language-model-inspired dot-product self-attention mechanism into both encoder and decoder pathways, thereby enhancing joint spatial-contextual modeling. This design represents the first integration in medical image segmentation of a lightweight, high-efficiency backbone with a highly interpretable attention mechanism, enabling end-to-end co-optimization of multi-sequence features. Evaluated on the BraTS 2021 test set, AXUNet achieves a mean Dice score of 93.73—surpassing state-of-the-art methods including Inception-UNet (90.88) and Xception-UNet (93.24)—and demonstrates superior performance across all three tumor subregions: whole tumor (WT), tumor core (TC), and enhancing tumor (ET). The method thus provides more reliable imaging support for clinical diagnosis and treatment planning.

0 citationsRead paper

ExKG-LLM: Leveraging Large Language Models for Automated Expansion of Cognitive Neuroscience Knowledge Graphs

Mar 09, 2025

To address low accuracy, narrow coverage, and infrequent updates in Cognitive Neuroscience Knowledge Graph (CNKG) expansion, this paper proposes the first LLM-driven incremental construction framework tailored for CNKGs. The method integrates large language model–based joint entity-relation extraction, semantic consistency–constrained relation inference, knowledge graph embedding optimization, and dynamic graph structure integration—enabling efficient, scalable, and semantically faithful graph updates. Experimental results demonstrate significant improvements: precision reaches 0.80 (+6.67%), recall 0.81 (+15.71%), and F1-score increases by 11.81%. Node and edge counts grow by over 21%, while user interaction rates in clinical decision support scenarios rise by 20%. This work establishes a reusable methodology and technical paradigm for automated, iterative evolution of domain-specific knowledge graphs.

0 citationsRead paper

Recent publications

Latest Papers

PolypVision: A Three-Stage Hierarchical Deep Learning Framework for Classification and Segmentation of Colorectal Polyps

Aug 11, 2026

This study addresses the challenge of accurate detection, segmentation, and endoscopy–histology joint classification of colorectal polyps by proposing a three-stage hierarchical deep learning framework. The first stage employs EfficientNetV2-M for polyp type and morphology classification; the second stage utilizes UNet++ for high-precision segmentation and recommends resection strategies; and the final stage refines adenoma subtyping through transfer learning. Innovatively integrating a task-progressive architecture, task-specific loss functions (Focal Loss and Dice+BCE), and cross-stage knowledge transfer, the framework achieves device-agnostic fully automated analysis. Evaluated on PolypGen, Kvasir-SEG, and CVC-ClinicDB datasets, it attains an AUC of 0.99 and mAP@50 of 94.4%, matching or surpassing current state-of-the-art methods, and has been deployed as a publicly accessible web application.

0 citationsRead paper

GIM-ENDO: A Multimodal Endoscopic Image and Video Dataset for Gastric Intestinal Metaplasia Morphology and Pathology

Jun 18, 2026

This study addresses a critical gap in artificial intelligence research on gastric intestinal metaplasia (GIM) by introducing the first publicly available, pathology-validated, multimodal endoscopic dataset. The dataset integrates high-definition white-light endoscopy, narrow-band imaging (NBI), and magnifying NBI (M-NBI) images and videos from 24 patients (22 GIM-positive), comprehensively annotated with histopathological subtypes, OLGA/OLGIM staging, standardized endoscopic features, and full clinical metadata. By providing a rigorously curated benchmark resource, this work enables AI-driven real-time GIM detection, subtype classification, and phenotypic analysis, thereby filling a significant void in standardized multimodal data for GIM research.

0 citationsRead paper

A Hybrid Computational Intelligence Framework with Metaheuristic Optimization for Drug-Drug Interaction Prediction

Oct 08, 2025

Drug–drug interactions (DDIs) are a leading cause of preventable adverse drug events, making accurate DDI prediction critical for clinical medication safety. To address this, we propose a high-accuracy, interpretable DDI prediction framework: (1) dual molecular embeddings are constructed by integrating Mol2Vec and SMILES-BERT; (2) a rule-based clinical scoring metric (RBScore) is designed—requiring no interaction labels—to explicitly incorporate pharmacological knowledge; and (3) a three-stage metaheuristic optimization algorithm (RSmpl-ACO-PSO) automatically tunes hyperparameters of a lightweight classifier. Evaluated on the DrugBank dataset, our model achieves 0.911 ROC-AUC and 0.867 PR-AUC. Furthermore, it demonstrates strong generalizability in a real-world cohort of type 2 diabetes patients, validating its clinical applicability and robustness.

0 citationsRead paper

Attention Xception UNet (AXUNet): A Novel Combination of CNN and Self-Attention for Brain Tumor Segmentation

Mar 26, 2025

To address the challenge of precise multi-sequence (T1CE/T2/FLAIR) MRI segmentation for gliomas, this paper proposes AXUNet—a novel UNet-based architecture that integrates an Xception backbone with a large-language-model-inspired dot-product self-attention mechanism into both encoder and decoder pathways, thereby enhancing joint spatial-contextual modeling. This design represents the first integration in medical image segmentation of a lightweight, high-efficiency backbone with a highly interpretable attention mechanism, enabling end-to-end co-optimization of multi-sequence features. Evaluated on the BraTS 2021 test set, AXUNet achieves a mean Dice score of 93.73—surpassing state-of-the-art methods including Inception-UNet (90.88) and Xception-UNet (93.24)—and demonstrates superior performance across all three tumor subregions: whole tumor (WT), tumor core (TC), and enhancing tumor (ET). The method thus provides more reliable imaging support for clinical diagnosis and treatment planning.

0 citationsRead paper

ExKG-LLM: Leveraging Large Language Models for Automated Expansion of Cognitive Neuroscience Knowledge Graphs

Mar 09, 2025

To address low accuracy, narrow coverage, and infrequent updates in Cognitive Neuroscience Knowledge Graph (CNKG) expansion, this paper proposes the first LLM-driven incremental construction framework tailored for CNKGs. The method integrates large language model–based joint entity-relation extraction, semantic consistency–constrained relation inference, knowledge graph embedding optimization, and dynamic graph structure integration—enabling efficient, scalable, and semantically faithful graph updates. Experimental results demonstrate significant improvements: precision reaches 0.80 (+6.67%), recall 0.81 (+15.71%), and F1-score increases by 11.81%. Node and edge counts grow by over 21%, while user interaction rates in clinical decision support scenarios rise by 20%. This work establishes a reusable methodology and technical paradigm for automated, iterative evolution of domain-specific knowledge graphs.

0 citationsRead paper