Hasil untuk "Computer applications to medicine. Medical informatics"

A. Esteva, Alexandre Robicquet, Bharath Ramsundar et al.

R. Bellazzi, B. Zupan

BACKGROUND The widespread availability of new computational methods and tools for data analysis and predictive modeling requires medical informatics researchers and practitioners to systematically select the most appropriate strategy to cope with clinical prediction problems. In particular, the collection of methods known as 'data mining' offers methodological and technical solutions to deal with the analysis of medical data and construction of prediction models. A large variety of these methods requires general and simple guidelines that may help practitioners in the appropriate selection of data mining tools, construction and validation of predictive models, along with the dissemination of predictive models within clinical environments. PURPOSE The goal of this review is to discuss the extent and role of the research area of predictive data mining and to propose a framework to cope with the problems of constructing, assessing and exploiting data mining models in clinical medicine. METHODS We review the recent relevant work published in the area of predictive data mining in clinical medicine, highlighting critical issues and summarizing the approaches in a set of learned lessons. RESULTS The paper provides a comprehensive review of the state of the art of predictive data mining in clinical medicine and gives guidelines to carry out data mining studies in this field. CONCLUSIONS Predictive data mining is becoming an essential instrument for researchers and clinical practitioners in medicine. Understanding the main issues underlying these methods and the application of agreed and standardized procedures is mandatory for their deployment and the dissemination of results. Thanks to the integration of molecular and clinical data taking place within genomic medicine, the area has recently not only gained a fresh impulse but also a new set of complex problems it needs to address.

Raquel Dias, A. Torkamani

Artificial intelligence (AI) is the development of computer systems that are able to perform tasks that normally require human intelligence. Advances in AI software and hardware, especially deep learning algorithms and the graphics processing units (GPUs) that power their training, have led to a recent and rapidly increasing interest in medical AI applications. In clinical diagnostics, AI-based computer vision approaches are poised to revolutionize image-based diagnostics, while other AI subtypes have begun to show similar promise in various diagnostic modalities. In some areas, such as clinical genomics, a specific type of AI algorithm known as deep learning is used to process large and complex genomic datasets. In this review, we first summarize the main classes of problems that AI systems are well suited to solve and describe the clinical diagnostic tasks that benefit from these solutions. Next, we focus on emerging methods for specific tasks in clinical genomics, including variant calling, genome annotation and variant classification, and phenotype-to-genotype correspondence. Finally, we end with a discussion on the future potential of AI in individualized medicine applications, especially for risk prediction in common complex diseases, and the challenges, limitations, and biases that must be carefully addressed for the successful deployment of AI in medical applications, particularly those utilizing human genetics and genomics data.

R. Haux

Clement Adebamowo, Adeola Akintola, Oluchi C Maduka et al.

BackgroundData science health research (DSHR) uses novel computational methods and high-performance computing to analyze big data from conventional and nonconventional health and related sources to generate novel insights and communications. DSHR creates assets but generates ethical, legal, and social challenges. Key gaps in current ethical oversight of DSHR include blurred boundaries between research and nonresearch data use, inadequate protection of data donors, power imbalances that risk extractive research practices, algorithmic biases, and regulatory inadequacies. Nigeria, a typical low- and middle-income country with rapidly expanding DSHR, exemplifies this environment and concerns. ObjectiveThis study will elicit answers from Nigerian DSHR stakeholders and contribute to understanding the ethical, legal, and social implications (ELSI) of DSHR and developing novel ethical oversight frameworks. MethodsBetween October 2024 and January 2025, we conducted Key Informant Interviews with 65 stakeholders of 87 individuals. The Key Informant Interview guide comprised 11 construct-based question domains addressing awareness of policies and laws, ethical oversight processes, ELSI considerations in policy development, experiences addressing DSHR challenges, organizational and procedural frameworks, ideal oversight components, stakeholder roles, research impact on ethics and policy, regulatory influences on research practices, equity-enhancing policies, and balanced regulations. The interviews lasted 60-90 minutes and were transcribed. We analyzed the transcripts using a hybrid deductive-inductive approach. A priori codes derived from research objectives provided the analytical framework while allowing for the identification of emergent concepts. The iterative 3-level coding process involved initial code generation, evaluation, and refinement, with codes grouped into thematic families and semantic networks representing hierarchical concept relationships. Query tools and Boolean operators were used to interrogate the codes to extract findings. ResultsOf 87 invited individuals, 22 (25%) were unable to participate. The 65 participants (age: mean 47.9, SD 7.9 years; 50/65, 77% male) included data science health researchers (25/65, 39%), biomedical researchers (17/65, 26%), Health Research Ethics Committee members (12/65, 19%), and policymakers (11/65, 17%). Most held doctoral degrees (38/65, 57%) and were affiliated with academic institutions (45/65, 69%) and government organizations (26/65, 40%), and had received general research ethics training (50/65, 77%). However, only 12% (8/65) had received predominantly short-duration ethics-specific DSHR training, while 92% (60/65) acknowledged the need for specialized DSHR ethics education. As of January 2025, the interview transcripts have been generated, with checking completed, with qualitative analysis scheduled for completion by March 2025 and completion of primary manuscripts by the end of 2025. ConclusionsThis study will generate stakeholder-informed recommendations for ethical oversight of DSHR that address issues relating to broad consent, ELSI, data ownership, benefit-sharing, and donor protection in resource-limited settings. Our findings will inform global DSHR and research ethics communities on the development of contextually appropriate oversight mechanisms that promote equitable partnerships, co-ownership, and tiered data governance. International Registered Report Identifier (IRRID)DERR1-10.2196/78557

Prottay Kumar Adhikary, Isha Motiyani, Gayatri Oke et al.

Abstract BackgroundThe quality and accessibility of menstrual health education (MHE) in low- and middle-income countries, including India, remain inadequate due to persistent challenges (eg, poverty, social stigma, and gender inequality). While community-driven initiatives have sought to raise awareness, artificial intelligence offers a scalable and efficient solution for disseminating accurate information. However, existing general-purpose large language models (LLMs) are often ill-suited for this task, tending to exhibit low accuracy, cultural insensitivity, and overly complex responses. To address these limitations, we developed MenstLLaMA—a specialized LLM tailored to the Indian context and designed to deliver MHE empathetically, supportively, and accessibly. ObjectiveWe aimed to develop and evaluate MenstLLaMA—a specialized LLM tailored to deliver accurate, culturally sensitive MHE—and assess its effectiveness in comparison to existing general-purpose models. MethodsWe curated MENST—a novel, domain-specific dataset comprising 23,820 question-answer pairs aggregated from medical websites, government portals, and health education resources. This dataset was systematically annotated with metadata capturing age groups, regions, topics, and sociocultural contexts. MenstLLaMA was developed by fine-tuning Meta-LLaMA-3-8B-Instruct, using parameter-efficient fine-tuning with low-rank adaptation to achieve domain alignment while minimizing computational overhead. We benchmarked MenstLLaMA against 9 state-of-the-art general-purpose LLMs, including GPT-4o, Claude-3, Gemini 1.5 Pro, and Mistral. The evaluation followed a multilayered framework: (1) automatic evaluation using standard natural language processing metrics (BLEU [Bilingual Evaluation Understudy], METEOR [Metric for Evaluation of Translation with Explicit Ordering], ROUGE-L [Recall-Oriented Understudy for Gisting Evaluation-Longest Common Subsequence], and BERTScore [Bidirectional Encoder Representations from Transformers Score]); (2) evaluation by clinical experts (N=18), who rated 200 expert-curated queries for accuracy and appropriateness; (3) medical practitioner interaction through the ISHA (Intelligent System for Menstrual Health Assistance) interactive chatbot, assessing qualitative dimensions (eg, relevance, understandability, preciseness, correctness,context sensitivity ResultsMenstLLaMA achieved the highest scores in BLEU (0.059) and BERTScore (0.911), outperforming GPT-4o (BLEU: 0.052, BERTScore: 0.896) and Claude-3 (BERTScore: 0.888). Clinical experts preferred MenstLLaMA’s responses over gold-standard answers in several culturally sensitive cases. In medical practitioners’ evaluations using the ISHA—the chat interface powered by MenstLLaMA—the model scored 3.5 in relevanceunderstandabilityprecisenesscorrectnesscontext sensitivityunderstandabilityrelevanceprecisenesscorrectnesstoneflowcontext sensitivity ConclusionsMenstLLaMA demonstrates exceptional accuracy, empathy, and user satisfaction within the domain of MHE, bridging critical gaps left by general-purpose LLMs. Its potential for integration into broader health education platforms positions it as a transformative tool for menstrual well-being. Future research could explore its long-term impact on public perception and menstrual hygiene practices, while expanding demographic representation, enhancing context sensitivity, and integrating multimodal and voice-based interactions to improve accessibility across diverse user groups.

Praveenbalaji Rajendran, Mojtaba Safari, Wenfeng He et al.

Recent advancements in artificial intelligence (AI), particularly foundation models (FMs), have revolutionized medical image analysis, demonstrating strong zero- and few-shot performance across diverse medical imaging tasks, from segmentation to report generation. Unlike traditional task-specific AI models, FMs leverage large corpora of labeled and unlabeled multimodal datasets to learn generalized representations that can be adapted to various downstream clinical applications with minimal fine-tuning. However, despite the rapid proliferation of FM research in medical imaging, the field remains fragmented, lacking a unified synthesis that systematically maps the evolution of architectures, training paradigms, and clinical applications across modalities. To address this gap, this review article provides a comprehensive and structured analysis of FMs in medical image analysis. We systematically categorize studies into vision-only and vision-language FMs based on their architectural foundations, training strategies, and downstream clinical tasks. Additionally, a quantitative meta-analysis of the studies was conducted to characterize temporal trends in dataset utilization and application domains. We also critically discuss persistent challenges, including domain adaptation, efficient fine-tuning, computational constraints, and interpretability along with emerging solutions such as federated learning, knowledge distillation, and advanced prompting. Finally, we identify key future research directions aimed at enhancing the robustness, explainability, and clinical integration of FMs, thereby accelerating their translation into real-world medical practice.

Pedro M. Gordaliza, Nataliia Molchanova, Jaume Banus et al.

Deep learning models for medical image segmentation suffer significant performance drops due to distribution shifts, but the causal mechanisms behind these drops remain poorly understood. We extend causal attribution frameworks to high-dimensional segmentation tasks, quantifying how acquisition protocols and annotation variability independently contribute to performance degradation. We model the data-generating process through a causal graph and employ Shapley values to fairly attribute performance changes to individual mechanisms. Our framework addresses unique challenges in medical imaging: high-dimensional outputs, limited samples, and complex mechanism interactions. Validation on multiple sclerosis (MS) lesion segmentation across 4 centers and 7 annotators reveals context-dependent failure modes: annotation protocol shifts dominate when crossing annotators (7.4% $\pm$ 8.9% DSC attribution), while acquisition shifts dominate when crossing imaging centers (6.5% $\pm$ 9.1%). This mechanism-specific quantification enables practitioners to prioritize targeted interventions based on deployment context.

Ana Carolina de Jesus Paniza, Fabio Ynoe Moraes

Pathology and oncology education are at an inflection point. Beyond abbreviated preclinical blocks, the central problem is pedagogical misalignment with learners who expect relevance, interactivity, and clinical application. We advocate a shift from content delivery to concept integration anchored in clinical reasoning and data literacy. In oncology, trainees must learn to interpret next‑generation sequencing and biomarker profiles, participate in molecular tumor boards, sequence precision therapies, manage toxicities, and incorporate patient‑reported outcomes-competencies rarely taught in a structured way. The digitization of histopathology and the integration of artificial intelligence demand exposure to digital pathology and critical appraisal of algorithmic outputs, including AI‑supported IHC quantification, variant classification, and methylation‑based classifiers. Large language models may enhance self‑directed learning but require faculty oversight, instruction in appraisal and ethics, and safeguards against inaccuracy and overconfidence. Operationalizing these reforms requires institutional commitment, curriculum redesign that integrates pathology, oncology, genomics, and decision‑making, and expanded residency time to acquire competencies in informatics and AI (machine learning, deep learning, supervised and unsupervised methods, and validation). Faculty development, adoption of digital platforms and virtual microscopy, competency‑based assessment, and collaboration with computer scientists, bioinformaticians, and ethicists are essential. Implementation barriers-including limited faculty time, resource constraints, and accreditation requirements-can be mitigated by pilot programs, strategic partnerships, phased integration, and attention to transparency, equity, and accountability. Absent deliberate reform within LCME and ACGME frameworks that currently do not mandate genomics or AI literacy, future physicians will enter practice unprepared for precision medicine. Modernizing curricula to meet the genomics and AI era is therefore urgent.

Anush Karampuri, Shyam Perugu

Breast cancer is the most prevalent and heterogeneous form of cancer affecting women worldwide. Various therapeutic strategies are in practice based on the extent of disease spread, such as surgery, chemotherapy, radiotherapy, and immunotherapy. Combinational therapy is another strategy that has proven to be effective in controlling cancer progression. Administration of Anchor drug, a well-established primary therapeutic agent with known efficacy for specific targets, with Library drug, a supplementary drug to enhance the efficacy of anchor drugs and broaden the therapeutic approach. Our work focused on harnessing regression-based Machine learning (ML) and deep learning (DL) algorithms to develop a structure-activity relationship between the molecular descriptors of drug pairs and their combined biological activity through a QSAR (Quantitative structure-activity relationship) model. 11 popularly known machine learning and deep learning algorithms were used to develop QSAR models. A total of 52 breast cancer cell lines, 25 anchor drugs, and 51 library drugs were considered in developing the QSAR model. It was observed that Deep Neural Networks (DNNs) achieved an impressive R2 (Coefficient of Determination) of 0.94, with an RMSE (Root Mean Square Error) value of 0.255, making it the most effective algorithm for developing a structure-activity relationship with strong generalization capabilities. In conclusion, applying combinational therapy alongside ML and DL techniques represents a promising approach to combating breast cancer.

Ana Echarri, Carmen Pradera, Gema Molina et al.

Introduction: Skin lesions are a common extraintestinal manifestation associated with inflammatory bowel disease (IBD), although they may also appear as a complication of IBD treatment. Prompt referral to the dermatologist can be very helpful in practice. Teledermatology complements the traditional in-person health care modality, improving access to dermatological care. Objective: To evaluate the impact of a store-and-forward teledermatology electronic consultation (e-consult) program on the care of IBD patients. Methods: A retrospective study assessing the outcomes of our teledermatology program over its first 2 years of implementation. Results: A total of 39 consultations involving 33 patients (69.2% women, mean age 39.6 years [12–63]) were conducted. The mean number of teleconsultations was 2.8 per month in the initial implementation stage: 33 consultations were carried out in patients with Crohn's disease and 6 in ulcerative colitis. Only 18% of the patients had an active flare-up. The most frequent reason for the e-consult was paradoxical psoriasiform lesions (n = 13, 33.3%), commonly related with anti-tumor necrosis factor agents (70% of the patients) and hidradenitis suppurativa (n = 4, 10.3%). Resolution was achieved in 87% of patients, with a mean waiting time of 4.7 days (0–14). Almost all patients (97%) were satisfied with our program, and considered the referral through the program to be appropriate (92%). Best valued features were the reduced waiting time and the coordinated approach between the two departments involved. Conclusions: Dermatology e-consult is an efficient and useful means of optimizing IBD patient care.

Sarah Müller, Louisa Fay, Lisa M. Koch et al.

Medical imaging cohorts are often confounded by factors such as acquisition devices, hospital sites, patient backgrounds, and many more. As a result, deep learning models tend to learn spurious correlations instead of causally related features, limiting their generalizability to new and unseen data. This problem can be addressed by minimizing dependence measures between intermediate representations of task-related and non-task-related variables. These measures include mutual information, distance correlation, and the performance of adversarial classifiers. Here, we benchmark such dependence measures for the task of preventing shortcut learning. We study a simplified setting using Morpho-MNIST and a medical imaging task with CheXpert chest radiographs. Our results provide insights into how to mitigate confounding factors in medical imaging.

Wasif Khan, Seowung Leem, Kyle B. See et al.

Foundation models (FMs) are large-scale deep learning models trained on massive datasets, often using self-supervised learning techniques. These models serve as a versatile base for a wide range of downstream tasks, including those in medicine and healthcare. FMs have demonstrated remarkable success across multiple healthcare domains. However, existing surveys in this field do not comprehensively cover all areas where FMs have made significant strides. In this survey, we present a comprehensive review of FMs in medicine, focusing on their evolution, learning strategies, flagship models, applications, and associated challenges. We examine how prominent FMs, such as the BERT and GPT families, are transforming various aspects of healthcare, including clinical large language models, medical image analysis, and omics research. Additionally, we provide a detailed taxonomy of FM-enabled healthcare applications, spanning clinical natural language processing, medical computer vision, graph learning, and other biology- and omics- related tasks. Despite the transformative potentials of FMs, they also pose unique challenges. This survey delves into these challenges and highlights open research questions and lessons learned to guide researchers and practitioners. Our goal is to provide valuable insights into the capabilities of FMs in health, facilitating responsible deployment and mitigating associated risks.

Shahabedin Nabavi, Kian Anvari Hamedani, Mohsen Ebrahimi Moghaddam et al.

Background: Image classification can be considered one of the key pillars of medical image analysis. Deep learning (DL) faces challenges that prevent its practical applications despite the remarkable improvement in medical image classification. The data distribution differences can lead to a drop in the efficiency of DL, known as the domain shift problem. Besides, requiring bulk annotated data for model training, the large size of models, and the privacy-preserving of patients are other challenges of using DL in medical image classification. This study presents a strategy that can address the mentioned issues simultaneously. Method: The proposed domain adaptive model based on knowledge distillation can classify images by receiving limited annotated data of different distributions. The designed multiple teachers-meticulous student model trains a student network that tries to solve the challenges by receiving the parameters of several teacher networks. The proposed model was evaluated using six available datasets of different distributions by defining the respiratory motion artefact detection task. Results: The results of extensive experiments using several datasets show the superiority of the proposed model in addressing the domain shift problem and lack of access to bulk annotated data. Besides, the privacy preservation of patients by receiving only the teacher network parameters instead of the original data and consolidating the knowledge of several DL models into a model with almost similar performance are other advantages of the proposed model. Conclusions: The proposed model can pave the way for practical clinical applications of deep classification methods by achieving the mentioned objectives simultaneously.

Anna Reithmeir, Veronika Spieker, Vasiliki Sideri-Lampretsa et al.

Image registration is fundamental in medical imaging applications, such as disease progression analysis or radiation therapy planning. The primary objective of image registration is to precisely capture the deformation between two or more images, typically achieved by minimizing an optimization problem. Due to its inherent ill-posedness, regularization is a key component in driving the solution toward anatomically meaningful deformations. A wide range of regularization methods has been proposed for both conventional and deep learning-based registration. However, the appropriate application of regularization techniques often depends on the specific registration problem, and no one-fits-all method exists. Despite its importance, regularization is often overlooked or addressed with default approaches, assuming existing methods are sufficient. A comprehensive and structured review remains missing. This review addresses this gap by introducing a novel taxonomy that systematically categorizes the diverse range of proposed regularization methods. It highlights the emerging field of learned regularization, which leverages data-driven techniques to automatically derive deformation properties from the data. Moreover, this review examines the transfer of regularization methods from conventional to learning-based registration, identifies open challenges, and outlines future research directions. By emphasizing the critical role of regularization in image registration, we hope to inspire the research community to reconsider regularization strategies in modern registration algorithms and to explore this rapidly evolving field further.

Sicheng Wang, Che Liu, Rossella Arcucci

Recent advancements in medical vision-language pre-training (MedVLP) have significantly enhanced zero-shot medical vision tasks such as image classification by leveraging large-scale medical image-text pair pre-training. However, the performance of these tasks can be heavily influenced by the variability in textual prompts describing the categories, necessitating robustness in MedVLP models to diverse prompt styles. Yet, this sensitivity remains underexplored. In this work, we are the first to systematically assess the sensitivity of three widely-used MedVLP methods to a variety of prompts across 15 different diseases. To achieve this, we designed six unique prompt styles to mirror real clinical scenarios, which were subsequently ranked by interpretability. Our findings indicate that all MedVLP models evaluated show unstable performance across different prompt styles, suggesting a lack of robustness. Additionally, the models' performance varied with increasing prompt interpretability, revealing difficulties in comprehending complex medical concepts. This study underscores the need for further development in MedVLP methodologies to enhance their robustness to diverse zero-shot prompts.

E. Aramaki, Shoko Wakamiya, Shuntaro Yada et al.

Summary Objectives : Owing to the rapid progress of natural language processing (NLP), the role of NLP in the medical field has radically gained considerable attention from both NLP and medical informatics. Although numerous medical NLP papers are published annually, there is still a gap between basic NLP research and practical product development. This gap raises questions, such as what has medical NLP achieved in each medical field, and what is the burden for the practical use of NLP? This paper aims to clarify the above questions. Methods : We explore the literature on potential NLP products/services applied to various medical/clinical/healthcare areas. Results : This paper introduces clinical applications (bedside applications), in which we introduce the use of NLP for each clinical department, internal medicine, pre-surgery, post-surgery, oncology, radiology, pathology, psychiatry, rehabilitation, obstetrics, and gynecology. Also, we clarify technical problems to be addressed for encouraging bedside applications based on NLP. Conclusions : These results contribute to discussions regarding potentially feasible NLP applications and highlight research gaps for future studies.

Hanieh Rezazadeh Tamrin, Elham Saniei, Mehdi Salehi Barough

Introduction: Breast cancer is the most common cancer in women that causes more deaths than other cancers. Thermography is one of the methods of breast cancer diagnosis. The most important challenge in early detection of these images can be human error or lack of access to a skilled person. The use of artificial intelligence methods in image processing can be effective in early detection and reduction of human error. The main aim of this research was to introduce hybrid networks for intelligent diagnosis of breast cancer from thermographic images. Method: The thermographic images used in this study were collected from the DMR-IR database. First, the main features of the images were extracted by deep convolutional network (CNN). Then, FCNNs and SVM algorithms were used to classify breast cancer from thermographic images. Results: The accuracy rate for CNN_FC and CNN-SVM algorithms was 94.2% and 0.95%, respectively. In addition, the reliability parameters for these classifiers were calculated as 92.1%, and 97.5%, and the sensitivity for each of these classifiers as 95.5%, and 94.1%, respectively. Conclusion: The proposed model based on the deep hybrid network has good accuracy compared to similar algorithms; therefore, it can help doctors in the early diagnosis of breast cancer through thermographic images and minimize human error.

The Lancet Digital Health

Young Min Cho, Sunny Rai, Lyle Ungar et al.

Mental health conversational agents (a.k.a. chatbots) are widely studied for their potential to offer accessible support to those experiencing mental health challenges. Previous surveys on the topic primarily consider papers published in either computer science or medicine, leading to a divide in understanding and hindering the sharing of beneficial knowledge between both domains. To bridge this gap, we conduct a comprehensive literature review using the PRISMA framework, reviewing 534 papers published in both computer science and medicine. Our systematic review reveals 136 key papers on building mental health-related conversational agents with diverse characteristics of modeling and experimental design techniques. We find that computer science papers focus on LLM techniques and evaluating response quality using automated metrics with little attention to the application while medical papers use rule-based conversational agents and outcome metrics to measure the health outcomes of participants. Based on our findings on transparency, ethics, and cultural heterogeneity in this review, we provide a few recommendations to help bridge the disciplinary divide and enable the cross-disciplinary development of mental health conversational agents.