Informatics in Medicine Unlocked最新文献

{"title":"Patient2Trial: From patient to participant in clinical trials using large language models","authors":"Surabhi Datta ,&nbsp;Kyeryoung Lee ,&nbsp;Liang-Chin Huang ,&nbsp;Hunki Paek ,&nbsp;Roger Gildersleeve ,&nbsp;Jonathan Gold ,&nbsp;Deepak Pillai ,&nbsp;Jingqi Wang ,&nbsp;Mitchell K. Higashi ,&nbsp;Lizheng Shi ,&nbsp;Percio S. Gulko ,&nbsp;Hua Xu ,&nbsp;Chunhua Weng ,&nbsp;Xiaoyan Wang","doi":"10.1016/j.imu.2025.101615","DOIUrl":"10.1016/j.imu.2025.101615","url":null,"abstract":"<div><h3>Purpose</h3><div>Large language models (LLMs) exhibit promising language understanding and generation capabilities and have been adopted for various clinical use cases. Investigating the feasibility of leveraging LLMs in building a clinical trial retrieval system for patients is crucial as it can greatly enhance the patient enrollment process by prioritizing the most suitable trials pertaining to a patient. In this work, we develop an LLM-assisted system focused on a patient-initiated approach, allowing patients with specific conditions to directly find eligible trials by completing disorder-specific questionnaires.</div></div><div><h3>Methods</h3><div>We obtained clinical trial eligibility criteria (from ClinicalTrials.gov) and simulated patient questionnaires (or topics) from the Text REtrieval Conference (TREC) 2023 Clinical Trials Track conducted by the National Institute of Standards and Technology (NIST), in which we also participated. These topics cover eight disorders across diverse domains, namely glaucoma, anxiety, chronic obstructive pulmonary disease, breast cancer, Covid-19, rheumatoid arthritis, sickle cell anemia, and type 2 diabetes. A Generative Pre-trained Transformer model (GPT-4) was employed for system development. We conducted both quantitative and qualitative evaluation using 37 patient topics.</div></div><div><h3>Results</h3><div>The system achieved an overall Precision@10 (proportion of relevant trials) of 0.7351 and NDCG@10 (considers ranking order of relevant trials) of 0.8109, indicating its effectiveness in retrieving ranked lists of suitable trials for patients. Notably, for eight out of 37 patient topics, all the top 10 retrieved trials were relevant. The system scored the highest on breast cancer (NDCG@10 = 0.9347, Precision@10 = 0.84) and the lowest on type 2 diabetes (NDCG@10 = 0.61, Precision@10 = 0.475). One probable reason could be that the information in breast cancer topics is relatively straightforward to match. Qualitative error analysis classified errors into four categories (e.g., difficulty in correctly matching inclusion criteria) and further highlighted strengths (e.g., ability to make clinical inference).</div></div><div><h3>Conclusion</h3><div>We demonstrated the feasibility of integrating LLMs in identifying and ranking suitable trials for patients across multiple disorders. Further work is required to assess the system's generalizability on other disorders and patient information sources. This system has the potential to expedite the patient-trial matching process by suggesting a ranked list of applicable trials to patients and clinicians.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"53 ","pages":"Article 101615"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143103205","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learning unbiased risk prediction based algorithms in healthcare: A case study with primary care patients","authors":"Vibhuti Gupta ,&nbsp;Julian Broughton ,&nbsp;Ange Rukundo ,&nbsp;Lubna J. Pinky","doi":"10.1016/j.imu.2025.101627","DOIUrl":"10.1016/j.imu.2025.101627","url":null,"abstract":"<div><div>The proliferation of Artificial Intelligence (AI) has revolutionized the healthcare domain with technological advancements in conventional diagnosis and treatment methods. These advancements lead to faster disease detection, and management and provide personalized healthcare solutions. However, most of the clinical AI methods developed and deployed in hospitals have algorithmic and data-driven biases due to insufficient representation of specific race, gender, and age group which leads to misdiagnosis, disparities, and unfair outcomes. Thus, it is crucial to thoroughly examine these biases and develop computational methods that can mitigate biases effectively. This paper critically analyzes this problem by exploring different types of data and algorithmic biases during both pre-processing and post-processing phases to uncover additional, previously unexplored biases in a widely used real-world healthcare dataset of primary care patients. Additionally, effective strategies are proposed to address gender, race, and age biases, ensuring that risk prediction outcomes are equitable and impartial. Through experiments with various machine learning algorithms leveraging the Fairlearn tool, we have identified biases in the dataset, compared the impact of these biases on the prediction performance, and proposed effective strategies to mitigate these biases. Our results demonstrate clear evidence of racial, gender-based, and age-related biases in the healthcare dataset used to guide resource allocation for patients and have profound impact on the prediction performance which leads to unfair outcomes. Thus, it is crucial to implement mechanisms to detect and address unintended biases to ensure a safe, reliable, and trustworthy AI system in healthcare.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"54 ","pages":"Article 101627"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143519066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using implementation science to develop and deploy an oncology electronic health record","authors":"Carla Taramasco ,&nbsp;Rene Noel ,&nbsp;Gastón Márquez ,&nbsp;Diego Robles","doi":"10.1016/j.imu.2025.101625","DOIUrl":"10.1016/j.imu.2025.101625","url":null,"abstract":"<div><div>The management of oncology clinical processes involves the efficient management of data using electronic clinical records to effectively monitor and treat oncology patients. As the process of treating and monitoring cancer patients involves multiple stakeholders with differing perspectives, the implementation and deployment of oncology clinical registries represent a significant challenge. In this study, we address this complexity by employing a technique that helps translate implementation strategies into requirement identification methods, which are subsequently disseminated throughout the implementation and deployment phases of health information systems. We applied this technique to develop an electronic health record for the national cancer plan in Chile. The findings indicate that six implementation strategies are essential to addressing stakeholder needs, as well as three requirement identification techniques to describe the underlying problem. Furthermore, a study conducted with 27 stakeholders revealed that the perception of the oncology electronic clinical record has considerable acceptance in three critical functionalities related to the clinical process of oncology patient management. The use of implementation science strategies provides an alternative approach to understanding the underlying problem that stakeholders face when they require healthcare technologies.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"53 ","pages":"Article 101625"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143420327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A lightweight classification system for the early detection of diabetic retinopathy","authors":"Ashim Chakraborty,&nbsp;George Wilson,&nbsp;Cristina Luca","doi":"10.1016/j.imu.2025.101655","DOIUrl":"10.1016/j.imu.2025.101655","url":null,"abstract":"<div><div>The eye disease known as Diabetic Retinopathy is one of the leading causes of permanent blindness in people of working age worldwide. Early identification is crucial for the treatment and management of the condition and this study presents a trustworthy approach for identifying the early stages of the disease from fundus images. A comparative analysis of a supervised machine learning algorithm and manual classification conducted by qualified optometrists is used to evaluate the work. Diabetic Retinopathy features such as Hard Exudates, Microaneurysms and Blood Vessels are extracted from the retinal images by a number of feature extraction methods. The performance and robustness of the proposed novel system are assessed using confusion matrix data and AUC-ROC curves. The findings demonstrate the validity of the decision-based system for the early detection of diabetic retinopathy, with the potential to be deployed on a portable screening system that can be used by people living in remote areas of the world.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"57 ","pages":"Article 101655"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144230269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Novel hybrid intelligence model for early Alzheimer's diagnosis utilizing multimodal biomarker fusion","authors":"Shehu Mohammed ,&nbsp;Neha Malhotra ,&nbsp;Arun Singh ,&nbsp;Awad M. Awadelkarim ,&nbsp;Shakeel Ahmed ,&nbsp;Saiprasad Potharaju","doi":"10.1016/j.imu.2025.101668","DOIUrl":"10.1016/j.imu.2025.101668","url":null,"abstract":"<div><div>One of the significant causes of dementia and a leading peril to global public health is Alzheimer's disease (AD), which calls for early and accurate diagnosis. The paper proposes a novel hybrid machine learning framework that integrates Gradient Boosting Machine (GBM) and Deep Neural Networks (DNN) for predicting Alzheimer's disease from multimodal biomarkers. The database comprises 35 demographic, behavioral, and clinical features. Feature selection procedures produced key predicting variables (i.e., MMSE scores, performance in Activities of Daily Living (ADL), cholesterol level, and behavior problems). A hybrid model was created by combining individual models, and it proved to be the most effective compared to particular models, achieving 92.6 % accuracy and a 0.94 AUC score on the database. The synergy between the capability of GBM for tabular data and the ability of DNN for complex interaction gives a good outcome. The research demonstrates the efficacy of blending machine learning techniques for supporting Alzheimer's disease (AD) identification and provides a method for early identification at a broader level. It is hoped that more biomarkers will be incorporated, and the model will be validated on larger and more phenotypically diverse databases to achieve clinical usability and generalizability.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"57 ","pages":"Article 101668"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144563334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hybrid quantum neural networks for computer-aided sex diagnosis in forensic and physical anthropology","authors":"Asel Sagingalieva ,&nbsp;Luca Lusnig ,&nbsp;Fabio Cavalli ,&nbsp;Alexey Melnikov","doi":"10.1016/j.imu.2025.101682","DOIUrl":"10.1016/j.imu.2025.101682","url":null,"abstract":"<div><div>The determination of sex from skeletal remains is essential for forensic science and the reconstruction of demographic patterns in ancient communities. This study aims to develop and evaluate hybrid deep-quantum neural network architectures to improve the accuracy and robustness of sex estimation from calvarium shape data. Deep learning has reached new heights in a number of scientific fields, while quantum techniques have the potential to process data in unique ways, offering benefits such as parallel processing and superposition. In the study, we investigate how an integration of deep learning and quantum computing will optimize sex estimation based on the shape complexity of calvarium, specifically through the algorithm based on Fast Fourier Transform of calvarium shapes. In previous studies with the same data, the highest achieved accuracy was 82.25%, utilizing classical machine learning and neural networks, such as Multilayer Perceptron. To improve the performance, we used four different neural network models: the classical Multilayer Perceptron and Convolutional Neural Network, along with Hybrid Quantum Classical Neural Networks, including Hybrid Quantum Classical Convolutional Networks on the morphological variations of the curve representing the sagittal profile of the calvarium. All the models outperformed the experts and improved the previous findings, achieving over 82.4% accuracy. Furthermore, the experiments on stability and accuracy over small datasets showed that one of proposed hybrid networks outperforms classical analogue on a small amount of data. The final performance of the hybrid models is better than the classical results and the best result achieved is 87.4%. We also launched the best hybrid model on the QPU, and the achieved accuracy of 90.71% is comparable to the classical simulation at 92.14%. Our research benefited significantly from new tools for analyzing quantum variational algorithms, which are inaccessible to classical methods, enabling us to reach higher results. This study not only demonstrates success in solving the specific task but also opens new possibilities for the application of quantum technologies in anthropology.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"58 ","pages":"Article 101682"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144907898","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced ROI guided deep learning model for Alzheimer’s detection using 3D MRI images","authors":"Israt Jahan Khan ,&nbsp;Md. Fahim Bin Amin ,&nbsp;Md. Delwar Shahadat Deepu ,&nbsp;Hazera Khatun Hira ,&nbsp;Asif Mahmud ,&nbsp;Anas Mashad Chowdhury ,&nbsp;Salekul Islam ,&nbsp;Md. Saddam Hossain Mukta ,&nbsp;Swakkhar Shatabda ,&nbsp;Alzheimer’s Disease Neuroimaging Initiative","doi":"10.1016/j.imu.2025.101650","DOIUrl":"10.1016/j.imu.2025.101650","url":null,"abstract":"<div><div>Alzheimer’s disease is an incurable condition that predominantly affects the human brain, leading to the shrinkage of various brain regions and the disruption of neuronal connections. Current state-of-the-art methods for detecting Alzheimer’s disease using 3D MRI images are resource-intensive and time-consuming. In this paper, we propose a Regions of Interest (ROI)-guided detection paradigm to address these challenges. We employ a 3D ResNet integrated with a Convolutional Block Attention Module (CBAM), demonstrating that emphasising ROIs in brain imaging can substantially reduce both computational expenditure and training time. Our model exhibits robust performance in discriminating Alzheimer’s disease from mild cognitive impairment, achieving an accuracy of 88% across the entire brain and 92% within targeted ROIs on the ADNI dataset. The accuracy on the OASIS dataset is even higher, reaching 98% for all regions and 98.33% for the ROIs. When distinguishing Alzheimer’s disease from cognitively normal individuals, the accuracy improves further, achieving 93.33% for the ROIs on the ADNI dataset and 97.8% on the OASIS dataset. In differentiating cognitively normal individuals from those with mild cognitive impairment, the model attains an accuracy of 88.2% for the ROIs on the ADNI dataset and 98.6% on the OASIS dataset. These findings highlight a notable enhancement in detection accuracy through the utilisation of fewer, yet more salient brain regions, underscoring the efficacy of our ROI-guided approach.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"56 ","pages":"Article 101650"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144124751","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The love-hate state of mobile device management in healthcare: An international survey","authors":"George A. Gellert,&nbsp;Gabriel L. Gellert,&nbsp;Rachel Pickering,&nbsp;Sean P. Kelly","doi":"10.1016/j.imu.2024.101603","DOIUrl":"10.1016/j.imu.2024.101603","url":null,"abstract":"<div><h3>Objective</h3><div>To gather insights regarding mobile device fleet deployment, management and security in healthcare delivery organizations (HDOs), including unmet needs and gaps in capabilities, across four nations.</div></div><div><h3>Methods</h3><div>An exploratory online survey of health information technology leaders working in HDOs to gather information about respondents’ organizational deployment of mobile devices as well as existing and needed mobile management capabilities.</div></div><div><h3>Results</h3><div>HDO mobile device losses were high, with 42% reporting average annual loss rates of 11–30%. Reported organizational effectiveness in protecting confidential information on lost mobile devices was low, with 50% of respondents ranking at six or below on a 10-point scale. Perception of end user satisfaction accessing applications/data on mobile devices was low, with 56–60% ranking satisfaction at six or below on a 10-point scale. Less than half of HDOs reported seven core mobile device management capabilities. Reported costs of mobile device information security breach across nations were between $100,000 and $1 million (USD). Respondents estimated aggregate weekly downtime exceeds 500h among 28% in Australia, 49% in Germany, 45% in the UK, and 47% in the US.</div></div><div><h3>Conclusions</h3><div>HDOs reported substantial perceived gaps and challenges in effectively managing mobility. System leaders desire what mobile device workflows add to care delivery, but effectively and efficiently managing a mobile device fleet remains a significant challenge. Mobility management tools are needed to facilitate rapid mobile device authentication, and efficiency of information access, while reducing clinician friction. Existing shared mobile device management solutions can help HDOs reduce costs and improve access security, user experience and workflow flexibility.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"52 ","pages":"Article 101603"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143178372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dual attention mechanisms with patch-level significance embedding for ischemic stroke classification in brain CT images","authors":"Mahesh Anil Inamdar ,&nbsp;Anjan Gudigar ,&nbsp;U. Raghavendra ,&nbsp;Massimo Salvi ,&nbsp;Nithin Raj ,&nbsp;J. Pooja ,&nbsp;Ajay Hegde ,&nbsp;Girish R. Menon ,&nbsp;U. Rajendra Acharya","doi":"10.1016/j.imu.2025.101678","DOIUrl":"10.1016/j.imu.2025.101678","url":null,"abstract":"<div><div>Stroke is currently a major contributor to disability and mortality across the globe, with ischemic stroke being the most predominant subtype. Accurate and timely diagnosis is critical for effective treatment. This study introduces a novel deep learning framework that leverages patch-level significance analysis for precise identification of ischemic strokes in Computed Tomography (CT) images. Our approach integrates a dual attention mechanism dynamic and cross attention with hybrid convolutional kernels to analyze the relative importance of brain regions in stroke diagnosis. The proposed architecture captures both fine-grained and contextual features to identify significant regions through attention-weighted feature embedding. The framework is evaluated on a dataset of 2023 CT of four different classes (i.e., acute: 361, chronic: 267, subacute: 382, and normal: 1013 images), employing both four and nine non-overlapping patch configurations. Experimental results demonstrate that the light gradient boosted machine classifier achieved the highest patch identification accuracy of 94.81 % and the extra tree classifier achieved an accuracy of 99.51 % for classification using 4-patch configuration analysis. The study highlights the importance of features obtained from dense layers in mitigating overfitting and improving generalization. In addition, the study reveals the potential of attention modules with interpretable factors for patch identification of cerebral infarction, suggesting the potential of artificial intelligence in aiding medical diagnosis.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"57 ","pages":"Article 101678"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144779596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Glaucoma identification with retinal fundus images using deep learning: Systematic review","authors":"Dulani Meedeniya ,&nbsp;Thisara Shyamalee ,&nbsp;Gilbert Lim ,&nbsp;Pratheepan Yogarajah","doi":"10.1016/j.imu.2025.101644","DOIUrl":"10.1016/j.imu.2025.101644","url":null,"abstract":"<div><div>Glaucoma is a leading cause of blindness, affecting millions of people worldwide. It is a chronic eye condition that damages the optic nerve and, if left untreated, can lead to vision loss and a decreased quality of life. Therefore, there is a need to explore practical and reliable mechanisms for glaucoma identification. This study systematically reviews deep-learning approaches for glaucoma identification using retinal fundus images from 2018 to 2024. Compared to existing survey studies, we cover the latest research, including several public retinal fundus image datasets, and focus on segmentation, classification based on convolutional neural networks and vision transformers, and explainability. The findings of this study, including comparisons of existing methods and key insights, will assist researchers and developers in identifying the most suitable techniques for glaucoma detection.</div></div>","PeriodicalId":13953,"journal":{"name":"Informatics in Medicine Unlocked","volume":"56 ","pages":"Article 101644"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144083890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}