Pub Date : 2025-07-10DOI: 10.1016/j.artmed.2025.103187
Xiaobo Li , Yijia Zhang , Xiaodi Hou , Shilong Wang , Hongfei Lin
<div><h3>Background:</h3><div>The automatic International Classification of Diseases (ICD) coding task assigns unique medical codes to diseases in clinical texts for further data statistics, quality control, billing and other tasks. The efficiency and accuracy of medical code assignment is a significant challenge affecting healthcare. However, in clinical practice, Electronic Health Records (EHRs) data are usually complex, heterogeneous, non-standard and unstructured, and the manual coding process is time-consuming, laborious and error-prone. Traditional machine learning methods struggle to extract significant semantic information from clinical texts accurately, but the latest progress in Deep Learning (DL) has shown promising results to address these issues.</div></div><div><h3>Objective:</h3><div>This paper comprehensively reviewed recent advancements in utilizing deep learning for automatic ICD coding, which aimed to reveal prominent challenges and emerging development trends by summarizing and analyzing the model’s year, design motivation, deep neural networks, and auxiliary data.</div></div><div><h3>Methods:</h3><div>This review introduced systematic literature on automatic ICD coding methods based on deep learning. We screened 5 online databases, including Web of Science, SpringerLink, PubMed, ACM, and IEEE digital library, and collected 53 published articles related to deep learning-based ICD coding from 2017 to 2023.</div></div><div><h3>Results:</h3><div>These deep neural network methods aimed to overcome some challenges, such as lengthy and noisy clinical text, high dimensionality and functional relationships of medical codes, and long-tail label distribution. The Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), attention mechanisms, Transformers, Pre-trained Language Models (PLMs), etc, have become popular to address prominent issues in ICD coding. Meanwhile, introducing medical ontology within the ICD coding system (code description and code hierarchy) and external knowledge (Wikipedia articles, tabular data, Clinical Classification Software (CCS), fine-tuning PLMs based on biomedical corpus, entity recognition and concept extraction) has become an emerging trend for automatic ICD coding.</div></div><div><h3>Conclusion:</h3><div>This paper provided a comprehensive review of recent literature on applying deep learning technology to improve medical code assignment from a unique perspective. Multiple neural network methods (CNNs, RNNs, Transformers, PLMs, especially attention mechanisms) have been successfully applied in ICD tasks and achieved excellent performance. Various medical auxiliary data has also proven valuable in enhancing model feature representation and classification performance. Our in-depth and systematic analysis suggested that the automatic ICD coding method based on deep learning has a bright future in healthcare. Finally, we discussed some major challenges and outlined future development directio
背景：国际疾病自动分类（ICD）编码任务为临床文本中的疾病分配唯一的医学代码，用于进一步的数据统计、质量控制、计费等任务。医疗代码分配的效率和准确性是影响医疗保健的重大挑战。然而，在临床实践中，电子健康记录（EHRs）数据通常是复杂的、异构的、非标准的和非结构化的，并且人工编码过程耗时、费力且容易出错。传统的机器学习方法难以准确地从临床文本中提取重要的语义信息，但深度学习（DL）的最新进展显示出解决这些问题的有希望的结果。目的：通过对模型年份、设计动机、深度神经网络和辅助数据的总结和分析，全面回顾了近年来利用深度学习进行ICD自动编码的进展，揭示了当前面临的突出挑战和新兴发展趋势。方法：系统介绍了基于深度学习的ICD自动编码方法。我们筛选了Web of Science、SpringerLink、PubMed、ACM、IEEE数字图书馆等5个在线数据库，收集了2017 - 2023年间发表的53篇基于深度学习的ICD编码相关文章。结果：这些深度神经网络方法旨在克服临床文本冗长和嘈杂、医疗代码的高维数和功能关系以及长尾标签分布等挑战。卷积神经网络（cnn）、循环神经网络（rnn）、注意机制、变形器、预训练语言模型（PLMs）等，已经成为解决ICD编码中突出问题的流行方法。同时，在ICD编码系统内引入医学本体（代码描述和代码层次）和外部知识（维基百科文章、表格数据、临床分类软件（CCS）、基于生物医学语料库的微调PLMs、实体识别和概念提取）已成为ICD自动编码的新兴趋势。结论：本文从一个独特的角度全面回顾了近年来应用深度学习技术改善医疗代码分配的文献。多种神经网络方法（cnn、rnn、transformer、plm，尤其是注意力机制）已成功应用于ICD任务中，并取得了优异的性能。各种医疗辅助数据在增强模型特征表示和分类性能方面也被证明是有价值的。我们深入系统的分析表明，基于深度学习的ICD自动编码方法在医疗保健领域具有广阔的应用前景。最后，我们讨论了一些主要挑战，并概述了未来的发展方向。
{"title":"Deep learning for automatic ICD coding: Review, opportunities and challenges","authors":"Xiaobo Li , Yijia Zhang , Xiaodi Hou , Shilong Wang , Hongfei Lin","doi":"10.1016/j.artmed.2025.103187","DOIUrl":"10.1016/j.artmed.2025.103187","url":null,"abstract":"<div><h3>Background:</h3><div>The automatic International Classification of Diseases (ICD) coding task assigns unique medical codes to diseases in clinical texts for further data statistics, quality control, billing and other tasks. The efficiency and accuracy of medical code assignment is a significant challenge affecting healthcare. However, in clinical practice, Electronic Health Records (EHRs) data are usually complex, heterogeneous, non-standard and unstructured, and the manual coding process is time-consuming, laborious and error-prone. Traditional machine learning methods struggle to extract significant semantic information from clinical texts accurately, but the latest progress in Deep Learning (DL) has shown promising results to address these issues.</div></div><div><h3>Objective:</h3><div>This paper comprehensively reviewed recent advancements in utilizing deep learning for automatic ICD coding, which aimed to reveal prominent challenges and emerging development trends by summarizing and analyzing the model’s year, design motivation, deep neural networks, and auxiliary data.</div></div><div><h3>Methods:</h3><div>This review introduced systematic literature on automatic ICD coding methods based on deep learning. We screened 5 online databases, including Web of Science, SpringerLink, PubMed, ACM, and IEEE digital library, and collected 53 published articles related to deep learning-based ICD coding from 2017 to 2023.</div></div><div><h3>Results:</h3><div>These deep neural network methods aimed to overcome some challenges, such as lengthy and noisy clinical text, high dimensionality and functional relationships of medical codes, and long-tail label distribution. The Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), attention mechanisms, Transformers, Pre-trained Language Models (PLMs), etc, have become popular to address prominent issues in ICD coding. Meanwhile, introducing medical ontology within the ICD coding system (code description and code hierarchy) and external knowledge (Wikipedia articles, tabular data, Clinical Classification Software (CCS), fine-tuning PLMs based on biomedical corpus, entity recognition and concept extraction) has become an emerging trend for automatic ICD coding.</div></div><div><h3>Conclusion:</h3><div>This paper provided a comprehensive review of recent literature on applying deep learning technology to improve medical code assignment from a unique perspective. Multiple neural network methods (CNNs, RNNs, Transformers, PLMs, especially attention mechanisms) have been successfully applied in ICD tasks and achieved excellent performance. Various medical auxiliary data has also proven valuable in enhancing model feature representation and classification performance. Our in-depth and systematic analysis suggested that the automatic ICD coding method based on deep learning has a bright future in healthcare. Finally, we discussed some major challenges and outlined future development directio","PeriodicalId":55458,"journal":{"name":"Artificial Intelligence in Medicine","volume":"168 ","pages":"Article 103187"},"PeriodicalIF":6.1,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144614723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-10DOI: 10.1016/j.artmed.2025.103218
Aliaksandra Sikirzhytskaya , Ilya Tyagin , S. Scott Sutton , Michael D. Wyatt , Ilya Safro , Michael Shtutman
Neurodegenerative diseases like Alzheimer's, Parkinson's, and HIV-associated neurocognitive disorder severely impact patients and healthcare systems. While effective treatments remain limited, researchers are actively developing ways to slow progression and improve patient outcomes, requiring innovative approaches to handle huge volumes of new scientific data. To enable the automatic analysis of biomedical data we introduced AGATHA, an effective AI-based literature mining tool that can navigate massive scientific literature databases. The overarching goal of this effort is to adapt AGATHA for drug repurposing by revealing hidden connections between FDA-approved medications and a health condition of interest. Our tool converts the abstracts of peer-reviewed papers from PubMed into multidimensional space where each gene and health condition are represented by specific metrics. We implemented advanced statistical analysis to reveal distinct clusters of scientific terms within the virtual space created using AGATHA-calculated parameters for selected health conditions and genes. Partial Least Squares Discriminant Analysis was employed for categorizing and predicting samples (122 diseases and 20,889 genes) fitted to specific classes. Advanced statistics were employed to build a discrimination model and extract lists of genes specific to each disease class. We focused on repurposing drugs for dementia by identifying dementia-associated genes highly ranked in other disease classes. The method was developed for detection of genes that shared across multiple conditions and classified them based on their roles in biological pathways. This led to the selection of six primary drugs for further study.
{"title":"AI-based mining of biomedical literature: Applications for drug repurposing for the treatment of dementia","authors":"Aliaksandra Sikirzhytskaya , Ilya Tyagin , S. Scott Sutton , Michael D. Wyatt , Ilya Safro , Michael Shtutman","doi":"10.1016/j.artmed.2025.103218","DOIUrl":"10.1016/j.artmed.2025.103218","url":null,"abstract":"<div><div>Neurodegenerative diseases like Alzheimer's, Parkinson's, and HIV-associated neurocognitive disorder severely impact patients and healthcare systems. While effective treatments remain limited, researchers are actively developing ways to slow progression and improve patient outcomes, requiring innovative approaches to handle huge volumes of new scientific data. To enable the automatic analysis of biomedical data we introduced AGATHA, an effective AI-based literature mining tool that can navigate massive scientific literature databases. The overarching goal of this effort is to adapt AGATHA for drug repurposing by revealing hidden connections between FDA-approved medications and a health condition of interest. Our tool converts the abstracts of peer-reviewed papers from PubMed into multidimensional space where each gene and health condition are represented by specific metrics. We implemented advanced statistical analysis to reveal distinct clusters of scientific terms within the virtual space created using AGATHA-calculated parameters for selected health conditions and genes. Partial Least Squares Discriminant Analysis was employed for categorizing and predicting samples (122 diseases and 20,889 genes) fitted to specific classes. Advanced statistics were employed to build a discrimination model and extract lists of genes specific to each disease class. We focused on repurposing drugs for dementia by identifying dementia-associated genes highly ranked in other disease classes. The method was developed for detection of genes that shared across multiple conditions and classified them based on their roles in biological pathways. This led to the selection of six primary drugs for further study.</div></div>","PeriodicalId":55458,"journal":{"name":"Artificial Intelligence in Medicine","volume":"168 ","pages":"Article 103218"},"PeriodicalIF":6.1,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-07-01DOI: 10.1016/j.artmed.2025.103202
Weidun Xie , Xingjian Chen , Lei Huang , Zetian Zheng , Yuchen Wang , Ruoxuan Zhang , Xiao Zhang , Zhichao Liu , Chengbin Peng , Monika Gullerova , Ka-chun Wong
Imbalanced datasets have been a persistent challenge in bioinformatics, particularly in the context of drug-drug interaction (DDI) risk level datasets. Such imbalance can lead to biased models that perform poorly on underrepresented classes. To address this issue, one strategy is to construct a balanced dataset, while another involves employing more advanced features and models. In this study, we introduce a novel approach called DDintensity, which leverages pre-trained deep learning models as embedding generators combined with LSTM-attention models to address the imbalance in DDI risk level datasets. We tested embeddings from various domains, including images, graphs, and textual corpus. Among these, embeddings generated by BioGPT achieved the highest performance, with an Area Under the Curve (AUC) of 0.97 and an Area Under the Precision-Recall curve (AUPR) of 0.92. Our model was trained on the DDinter and further validated using the MecDDI dataset. Additionally, case studies on chemotherapeutic drugs, DB00398 (Sorafenib) and DB01204 (Mitoxantrone) used in oncology, were conducted to demonstrate the specificity and effectiveness of the this methods. Our approach demonstrates high scalability across DDI modalities, as well as the discovery of novel interactions. In summary, we introduce DDIntensity as a solution for imbalanced datasets in bioinformatics with pre-trained deep-learning embeddings.
{"title":"DDintensity: Addressing imbalanced drug-drug interaction risk levels using pre-trained deep learning model embeddings","authors":"Weidun Xie , Xingjian Chen , Lei Huang , Zetian Zheng , Yuchen Wang , Ruoxuan Zhang , Xiao Zhang , Zhichao Liu , Chengbin Peng , Monika Gullerova , Ka-chun Wong","doi":"10.1016/j.artmed.2025.103202","DOIUrl":"10.1016/j.artmed.2025.103202","url":null,"abstract":"<div><div>Imbalanced datasets have been a persistent challenge in bioinformatics, particularly in the context of drug-drug interaction (DDI) risk level datasets. Such imbalance can lead to biased models that perform poorly on underrepresented classes. To address this issue, one strategy is to construct a balanced dataset, while another involves employing more advanced features and models. In this study, we introduce a novel approach called DDintensity, which leverages pre-trained deep learning models as embedding generators combined with LSTM-attention models to address the imbalance in DDI risk level datasets. We tested embeddings from various domains, including images, graphs, and textual corpus. Among these, embeddings generated by BioGPT achieved the highest performance, with an Area Under the Curve (AUC) of 0.97 and an Area Under the Precision-Recall curve (AUPR) of 0.92. Our model was trained on the DDinter and further validated using the MecDDI dataset. Additionally, case studies on chemotherapeutic drugs, DB00398 (Sorafenib) and DB01204 (Mitoxantrone) used in oncology, were conducted to demonstrate the specificity and effectiveness of the this methods. Our approach demonstrates high scalability across DDI modalities, as well as the discovery of novel interactions. In summary, we introduce DDIntensity as a solution for imbalanced datasets in bioinformatics with pre-trained deep-learning embeddings.</div></div>","PeriodicalId":55458,"journal":{"name":"Artificial Intelligence in Medicine","volume":"168 ","pages":"Article 103202"},"PeriodicalIF":6.1,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144549994","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
IF 6.1 2区 医学Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE
Pub Date : 2025-06-26DOI: 10.1016/j.artmed.2025.103191
Ahmad Berjaoui , Eduardo Hugo Sanchez , Louis Roussel , Elizabeth Cohen-Jonathan Moyal
Glioblastoma is a highly aggressive form of brain cancer characterized by rapid progression and poor prognosis. Despite advances in treatment, the underlying genetic mechanisms driving this aggressiveness remain poorly understood. In this study, we employed multimodal deep learning approaches to investigate glioblastoma heterogeneity using joint image/RNA-seq analysis. Our results reveal novel genes associated with glioblastoma. By leveraging a combination of whole-slide images and RNA-seq, as well as introducing novel methods to encode RNA-seq data, we identified specific genetic profiles that may explain different patterns of glioblastoma progression. These findings provide new insights into the genetic mechanisms underlying glioblastoma heterogeneity and highlight potential targets for therapeutic intervention. Code and data downloading instructions are available at: https://github.com/ma3oun/gbheterogeneity.
{"title":"Uncovering the genetic basis of glioblastoma heterogeneity through multimodal analysis of whole slide images and RNA sequencing data","authors":"Ahmad Berjaoui , Eduardo Hugo Sanchez , Louis Roussel , Elizabeth Cohen-Jonathan Moyal","doi":"10.1016/j.artmed.2025.103191","DOIUrl":"10.1016/j.artmed.2025.103191","url":null,"abstract":"<div><div>Glioblastoma is a highly aggressive form of brain cancer characterized by rapid progression and poor prognosis. Despite advances in treatment, the underlying genetic mechanisms driving this aggressiveness remain poorly understood. In this study, we employed multimodal deep learning approaches to investigate glioblastoma heterogeneity using joint image/RNA-seq analysis. Our results reveal novel genes associated with glioblastoma. By leveraging a combination of whole-slide images and RNA-seq, as well as introducing novel methods to encode RNA-seq data, we identified specific genetic profiles that may explain different patterns of glioblastoma progression. These findings provide new insights into the genetic mechanisms underlying glioblastoma heterogeneity and highlight potential targets for therapeutic intervention. Code and data downloading instructions are available at: <span><span>https://github.com/ma3oun/gbheterogeneity</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55458,"journal":{"name":"Artificial Intelligence in Medicine","volume":"168 ","pages":"Article 103191"},"PeriodicalIF":6.1,"publicationDate":"2025-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144563602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-26DOI: 10.1016/j.artmed.2025.103200
Antonio Lopez-Martinez-Carrasco , Hugo M. Proença , Jose M. Juarez , Matthijs van Leeuwen , Manuel Campos
Antibiotic resistance is one of the major global threats to human health and occurs when antibiotics lose their ability to combat bacterial infections. In this problem, a clinical decision support system could use phenotypes in order to alert clinicians of the emergence of patterns of antibiotic resistance in patients. Patient phenotyping is the task of finding a set of patient characteristics related to a specific medical problem such as the one described in this work. However, a single explanation of a medical phenomenon might be useless in the eyes of a clinical expert and be discarded. The discovery of multiple patient phenotypes for the same medical phenomenon would be useful in such cases. Therefore, in this work, we define the problem of mining diverse top-k phenotypes and propose the EDSLM algorithm, which is based on the Subgroup Discovery technique, the subgroup list model, and the Minimum Description Length principle. Our proposal provides clinicians with a method with which to obtain multiple and diverse phenotypes of a set of patients. We show a real use case of phenotyping in antimicrobial resistance using the well-known MIMIC-III dataset.
{"title":"Discovering multiple antibiotic resistance phenotypes using diverse top-k subgroup list discovery","authors":"Antonio Lopez-Martinez-Carrasco , Hugo M. Proença , Jose M. Juarez , Matthijs van Leeuwen , Manuel Campos","doi":"10.1016/j.artmed.2025.103200","DOIUrl":"10.1016/j.artmed.2025.103200","url":null,"abstract":"<div><div>Antibiotic resistance is one of the major global threats to human health and occurs when antibiotics lose their ability to combat bacterial infections. In this problem, a clinical decision support system could use phenotypes in order to alert clinicians of the emergence of patterns of antibiotic resistance in patients. Patient phenotyping is the task of finding a set of patient characteristics related to a specific medical problem such as the one described in this work. However, a single explanation of a medical phenomenon might be useless in the eyes of a clinical expert and be discarded. The discovery of multiple patient phenotypes for the same medical phenomenon would be useful in such cases. Therefore, in this work, we define the problem of mining diverse top-k phenotypes and propose the EDSLM algorithm, which is based on the Subgroup Discovery technique, the subgroup list model, and the Minimum Description Length principle. Our proposal provides clinicians with a method with which to obtain multiple and diverse phenotypes of a set of patients. We show a real use case of phenotyping in antimicrobial resistance using the well-known MIMIC-III dataset.</div></div>","PeriodicalId":55458,"journal":{"name":"Artificial Intelligence in Medicine","volume":"167 ","pages":"Article 103200"},"PeriodicalIF":6.1,"publicationDate":"2025-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144514073","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-25DOI: 10.1016/j.artmed.2025.103195
Zhirui Liao , Lei Xie , Shanfeng Zhu
Drug–target interaction (DTI) identification is one of the crucial issues in the field of drug discovery. Machine learning approaches offer efficient ways to address this issue, reducing expensive and time-consuming laboratory experiments. However, the scarcity of annotated drug data with labels restricts supervised machine learning applications to DTI prediction. Drawing inspiration from recent advances in contrastive learning, we present ContraDTI—a novel framework that adopts multi-view contrastive learning to overcome data limitations in this paper. Our model considers the molecular graph of a drug as the main view and the SMILES string of a drug as the side view, employing two types of loss functions for the contrast of the main view and the cross-view alignment between the main and the side views. Extensive experiments on both single-target and multi-target DTI datasets demonstrate that ContraDTI enhances the classification performance of DTI prediction, particularly when labeled data is scarce. ContraDTI can be a powerful tool for DTI prediction in data-limited scenarios. The code of this paper is available at https://github.com/zhiruiliao/ContraDTI.
{"title":"ContraDTI: Improved drug–target interaction prediction via multi-view contrastive learning","authors":"Zhirui Liao , Lei Xie , Shanfeng Zhu","doi":"10.1016/j.artmed.2025.103195","DOIUrl":"10.1016/j.artmed.2025.103195","url":null,"abstract":"<div><div>Drug–target interaction (DTI) identification is one of the crucial issues in the field of drug discovery. Machine learning approaches offer efficient ways to address this issue, reducing expensive and time-consuming laboratory experiments. However, the scarcity of annotated drug data with labels restricts supervised machine learning applications to DTI prediction. Drawing inspiration from recent advances in contrastive learning, we present ContraDTI—a novel framework that adopts multi-view contrastive learning to overcome data limitations in this paper. Our model considers the molecular graph of a drug as the main view and the SMILES string of a drug as the side view, employing two types of loss functions for the contrast of the main view and the cross-view alignment between the main and the side views. Extensive experiments on both single-target and multi-target DTI datasets demonstrate that ContraDTI enhances the classification performance of DTI prediction, particularly when labeled data is scarce. ContraDTI can be a powerful tool for DTI prediction in data-limited scenarios. The code of this paper is available at <span><span>https://github.com/zhiruiliao/ContraDTI</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55458,"journal":{"name":"Artificial Intelligence in Medicine","volume":"168 ","pages":"Article 103195"},"PeriodicalIF":6.1,"publicationDate":"2025-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144534645","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-24DOI: 10.1016/j.artmed.2025.103197
Yazhou Zhu , Minxian Li , Qiaolin Ye , Shidong Wang , Tong Xin , Haofeng Zhang
Few-shot medical image segmentation (FSMIS) aims to perform the limited annotated data learning in the medical image analysis scope. Despite the progress has been achieved, current FSMIS models are all trained and deployed on the same data domain, as is not consistent with the clinical reality that medical imaging data is always across different data domains (e.g. imaging modalities, institutions and equipment sequences). In this paper, we introduce Cross-domain Few-shot Medical Image Segmentation (CD-FSMIS) and propose a RobustEMD matching mechanism based on Earth Mover’s Distance (EMD) to enhance cross-domain generalization. Our approach includes three key components: (1) a channel-wise feature decomposition strategy that uniformly divides support and query features into local nodes, (2) a texture structure aware weights generation method that restrains domain-relevant nodes through Sobel-based gradient calculation, and (3) a boundary-aware Hausdorff distance measurement for transportation cost calculation. Extensive experiments across three scenarios (cross-modal, cross-sequence and cross-institution) show that our method significantly outperforms existing approaches. And ablation studies further confirm that each component of our RobustEMD mechanism contributes to the enhanced performance. The experimental outcomes highlight strong generalization capabilities of our model in real-world heterogeneous medical imaging environments. Code is available at https://github.com/YazhouZhu19/RobustEMD.
{"title":"RobustEMD: Domain robust matching for cross-domain few-shot medical image segmentation","authors":"Yazhou Zhu , Minxian Li , Qiaolin Ye , Shidong Wang , Tong Xin , Haofeng Zhang","doi":"10.1016/j.artmed.2025.103197","DOIUrl":"10.1016/j.artmed.2025.103197","url":null,"abstract":"<div><div>Few-shot medical image segmentation (FSMIS) aims to perform the limited annotated data learning in the medical image analysis scope. Despite the progress has been achieved, current FSMIS models are all trained and deployed on the same data domain, as is not consistent with the clinical reality that medical imaging data is always across different data domains (e.g. imaging modalities, institutions and equipment sequences). In this paper, we introduce Cross-domain Few-shot Medical Image Segmentation (CD-FSMIS) and propose a RobustEMD matching mechanism based on Earth Mover’s Distance (EMD) to enhance cross-domain generalization. Our approach includes three key components: (1) a channel-wise feature decomposition strategy that uniformly divides support and query features into local nodes, (2) a texture structure aware weights generation method that restrains domain-relevant nodes through Sobel-based gradient calculation, and (3) a boundary-aware Hausdorff distance measurement for transportation cost calculation. Extensive experiments across three scenarios (cross-modal, cross-sequence and cross-institution) show that our method significantly outperforms existing approaches. And ablation studies further confirm that each component of our RobustEMD mechanism contributes to the enhanced performance. The experimental outcomes highlight strong generalization capabilities of our model in real-world heterogeneous medical imaging environments. Code is available at <span><span>https://github.com/YazhouZhu19/RobustEMD</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55458,"journal":{"name":"Artificial Intelligence in Medicine","volume":"167 ","pages":"Article 103197"},"PeriodicalIF":6.1,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144481460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-24DOI: 10.1016/j.artmed.2025.103196
Dianlei Gao, Fei Zhu
Drug–target interaction (DTI) is paramount in drug discovery and repurposing, which involves screening for effective candidate drugs by targeting specific proteins. Existing methods often focus on one or two representations of drugs or targets, and little has been explored regarding 3D structures. Moreover, how to capture interactions between multi-modal features comprehensively is also a key issue. A multi-modal interaction fusion method called GraphCF is proposed to overcome these limitations. Specifically, GraphCF uses a MixHop aggregator to gather higher-order neighborhood information between nodes in the DTI topological network and incorporate graph contrastive learning to capture more discriminative 2D representations of drugs and targets. Additionally, GraphCF utilizes convolutional neural networks and graph neural networks to extract the sequence and 3D structural features of drugs and targets, respectively. Then, GraphCF employs a cross-attention-based multi-feature fusion module to facilitate information interaction and fusion among multi-modal feature representations. GraphCF is evaluated and compared with some advanced methods on four public datasets, and the results demonstrate the competitive performance of GraphCF in DTI prediction.
{"title":"GraphCF: Drug–target interaction prediction via multi-feature fusion with contrastive graph neural network","authors":"Dianlei Gao, Fei Zhu","doi":"10.1016/j.artmed.2025.103196","DOIUrl":"10.1016/j.artmed.2025.103196","url":null,"abstract":"<div><div>Drug–target interaction (DTI) is paramount in drug discovery and repurposing, which involves screening for effective candidate drugs by targeting specific proteins. Existing methods often focus on one or two representations of drugs or targets, and little has been explored regarding 3D structures. Moreover, how to capture interactions between multi-modal features comprehensively is also a key issue. A multi-modal interaction fusion method called GraphCF is proposed to overcome these limitations. Specifically, GraphCF uses a MixHop aggregator to gather higher-order neighborhood information between nodes in the DTI topological network and incorporate graph contrastive learning to capture more discriminative 2D representations of drugs and targets. Additionally, GraphCF utilizes convolutional neural networks and graph neural networks to extract the sequence and 3D structural features of drugs and targets, respectively. Then, GraphCF employs a cross-attention-based multi-feature fusion module to facilitate information interaction and fusion among multi-modal feature representations. GraphCF is evaluated and compared with some advanced methods on four public datasets, and the results demonstrate the competitive performance of GraphCF in DTI prediction.</div></div>","PeriodicalId":55458,"journal":{"name":"Artificial Intelligence in Medicine","volume":"167 ","pages":"Article 103196"},"PeriodicalIF":6.1,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144518568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-24DOI: 10.1016/j.artmed.2025.103199
Lei Cao , Ling Feng , Yang Ding , Huijun Zhang , Xin Wang , Kaisheng Zeng , Yi Dai
Suicide is a tragedy for family and society. With social media becoming an integral part of people’s life nowadays, assessing suicidal risk based on one’s social media behavior has drawn increasing research attentions. The majority of the works trained a machine learning model to classify user’s suicidal risk severity level in a batch learning setting on the entire training data. This is not a timely and scalable solution in the context of social media where new data arrives sequentially in a stream form. In this study, we formulate and address the continuous suicidal risk assessment problem through a three-layered joint memory network, consisting of a short-term personal memory and long-term personal and global memories. Unlike existing methods that rely on static classification, our model supports real-time, continuous learning from users’ emotional and behavioral dynamics without the need for full retraining. This allows for personalized and adaptive risk tracking over time. We also present a way to continuously capture users’ personal features and integrate them in suicidal risk assessment. The performance on the constructed dataset containing 95 suicidal and 95 non-suicidal social media users shows that 96% of accuracy can be achieved with the proposed method.
{"title":"Online continuous learning of users suicidal risk on social media","authors":"Lei Cao , Ling Feng , Yang Ding , Huijun Zhang , Xin Wang , Kaisheng Zeng , Yi Dai","doi":"10.1016/j.artmed.2025.103199","DOIUrl":"10.1016/j.artmed.2025.103199","url":null,"abstract":"<div><div>Suicide is a tragedy for family and society. With social media becoming an integral part of people’s life nowadays, assessing suicidal risk based on one’s social media behavior has drawn increasing research attentions. The majority of the works trained a machine learning model to classify user’s suicidal risk severity level in a batch learning setting on the entire training data. This is not a timely and scalable solution in the context of social media where new data arrives sequentially in a stream form. In this study, we formulate and address the continuous suicidal risk assessment problem through a three-layered joint memory network, consisting of a short-term personal memory and long-term personal and global memories. Unlike existing methods that rely on static classification, our model supports real-time, continuous learning from users’ emotional and behavioral dynamics without the need for full retraining. This allows for personalized and adaptive risk tracking over time. We also present a way to continuously capture users’ personal features and integrate them in suicidal risk assessment. The performance on the constructed dataset containing 95 suicidal and 95 non-suicidal social media users shows that 96% of accuracy can be achieved with the proposed method.</div></div>","PeriodicalId":55458,"journal":{"name":"Artificial Intelligence in Medicine","volume":"167 ","pages":"Article 103199"},"PeriodicalIF":6.1,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144501962","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：