Hasil untuk "Computer applications to medicine. Medical informatics"

Johannes Allgaier, Lena Mulansky, R. Draelos et al.

BACKGROUND Medical use cases for machine learning (ML) are growing exponentially. The first hospitals are already using ML systems as decision support systems in their daily routine. At the same time, most ML systems are still opaque and it is not clear how these systems arrive at their predictions. METHODS In this paper, we provide a brief overview of the taxonomy of explainability methods and review popular methods. In addition, we conduct a systematic literature search on PubMed to investigate which explainable artificial intelligence (XAI) methods are used in 450 specific medical supervised ML use cases, how the use of XAI methods has emerged recently, and how the precision of describing ML pipelines has evolved over the past 20 years. RESULTS A large fraction of publications with ML use cases do not use XAI methods at all to explain ML predictions. However, when XAI methods are used, open-source and model-agnostic explanation methods are more commonly used, with SHapley Additive exPlanations (SHAP) and Gradient Class Activation Mapping (Grad-CAM) for tabular and image data leading the way. ML pipelines have been described in increasing detail and uniformity in recent years. However, the willingness to share data and code has stagnated at about one-quarter. CONCLUSIONS XAI methods are mainly used when their application requires little effort. The homogenization of reports in ML use cases facilitates the comparability of work and should be advanced in the coming years. Experts who can mediate between the worlds of informatics and medicine will become more and more in demand when using ML systems due to the high complexity of the domain.

Yoshiyasu Ito, Mika Moriyama, Akemi Nasu et al.

Abstract BackgroundShared decision-making (SDM) is a collaborative process that integrates patients’ values and preferences into health care decisions. In intensive care units, patients who are critically ill often lack the capacity to make decisions, necessitating surrogates to make complex choices regarding life-sustaining treatments (LSTs). ObjectiveThis scoping review aims to assess the range of research conducted on surrogate SDM for LSTs among patients who are critically ill over the past decade and highlight areas where current research remains limited. MethodsThis scoping review will follow the Joanna Briggs Institute methodology and adhere to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) reporting guidelines. Studies will be included if they examine SDM involving surrogates of adult patients who are critically ill in relation to LST decisions within intensive care unit settings. SDM is defined using 4 criteria: participation of both health care professionals and surrogates, mutual information sharing, consensus building, and agreement on treatment based on the patient’s values and preferences. A comprehensive search will be performed across PubMed, CINAHL, PsycInfo, CENTRAL, and Ichushi-Web for English- and Japanese-language studies published between 2016 and 2025. Eligible study designs will include quantitative, qualitative, and mixed methods research. Title and abstract screening, as well as full-text selection, will be conducted independently by 2 reviewers using Rayyan. Data will be extracted on study characteristics, SDM definitions, participant roles, and key findings. Results will be synthesized descriptively and presented in tables and narrative summaries to identify research gaps and inform future investigations. ResultsAs of June 13, 2025, the literature search has been completed. A total of 2899 citations were identified through the specified database searches, and 527 (18.2%) duplicates were removed. Title and abstract screening are currently in progress, and full-text review is expected to be completed by September 2025. ConclusionsThis scoping review will systematically map recent evidence on surrogate SDM in the context of LST decisions for patients who are critically ill. By synthesizing diverse studies, it will identify challenges faced by surrogates and summarize existing interventions that aim to improve SDM processes. The findings are expected to inform future interventions and policies and advance patient- and family-centered care in critical care settings.

Joseph Paillard, Antoine Collas, Denis A. Engemann et al.

Recent advances in machine learning have greatly expanded the repertoire of predictive methods for medical imaging. However, the interpretability of complex models remains a challenge, which limits their utility in medical applications. Recently, model-agnostic methods have been proposed to measure conditional variable importance and accommodate complex non-linear models. However, they often lack power when dealing with highly correlated data, a common problem in medical imaging. We introduce Hierarchical-CPI, a model-agnostic variable importance measure that frames the inference problem as the discovery of groups of variables that are jointly predictive of the outcome. By exploring subgroups along a hierarchical tree, it remains computationally tractable, yet also enjoys explicit family-wise error rate control. Moreover, we address the issue of vanishing conditional importance under high correlation with a tree-based importance allocation mechanism. We benchmarked Hierarchical-CPI against state-of-the-art variable importance methods. Its effectiveness is demonstrated in two neuroimaging datasets: classifying dementia diagnoses from MRI data (ADNI dataset) and analyzing the Berger effect on EEG data (TDBRAIN dataset), identifying biologically plausible variables.

Gemma Postill, Neesha Hussain-Shamsy, Stephanie Dephoure et al.

Social media platforms, such as Instagram, are increasingly used as a source of health information; however, it is unclear how to effectively leverage these platforms during public health emergencies. @PandemicPregnancyGuide (PPG) was an Instagram account created by Canadian physicians to provide perinatal health information during the COVID-19 pandemic. We conducted a cross-sectional survey, and assessed Instagram analytics, to determine how and why users followed PPG and its impact on health decision-making. Respondents most valued posts explaining scientific articles in lay language and the delivery of content by medical experts. Topics of greatest interest were COVID-19 vaccination while pregnant (76%), COVID-19 infection during pregnancy (71%), and labour and delivery during the pandemic (69%). Respondents self-reported being more likely to use COVID-19 protective measures while pregnant (80%), receive COVID-19 vaccines in pregnancy (87%), and vaccinate their children against COVID-19 (58%) due to the information shared by PPG. Taken together, we demonstrate how healthcare professionals can effectively leverage social media to disseminate health information and improve uptake of public health recommendations. We recommend consideration of our findings in the development of future health-based social media platforms, particularly during public health emergencies or campaigns.

C. Vandelanotte, S. Trost, Danya Hodgetts et al.

OBJECTIVE Physical inactivity is a leading modifiable cause of death and disease worldwide. Population-based interventions to increase physical activity are needed. Existing automated expert systems (e.g., computer-tailored interventions) have significant limitations that result in low long-term effectiveness. Therefore, innovative approaches are needed. This special communication aims to describe and discuss a novel mHealth intervention approach that proactively offers participants with hyper-personalised intervention content adjusted in real-time. METHODS Using machine learning approaches, we propose a novel physical activity intervention approach that can learn and adapt in real-time to achieve high levels of personalisation and user engagement, underpinned by a likeable digital assistant. It will consist of three major components: 1) conversations: to increase user's knowledge on a wide range of activity-related topics underpinned by Natural Language Processing; 2) nudge engine: to provide users with hyper-personalised cues to action underpinned by reinforcement learning (i.e., contextual bandit) and integrating real-time data from activity tracking, GPS, GIS, weather, and user provided data; 3) Q&A: to facilitate users asking any physical activity related questions underpinned by generative AI (e.g., ChatGPT, Bard) for content generation. RESULTS The detailed concept of the proposed physical activity intervention platform demonstrates the practical application of a just-in-time adaptive intervention applying various machine learning techniques to deliver a hyper-personalised physical activity intervention in an engaging way. Compared to traditional interventions, the novel platform is expected to show potential for increased user engagement and long-term effectiveness due to: 1) using new variables to personalise content (e.g., GPS, weather), 2) providing behavioural support at the right time in real-time, 3) implementing an engaging digital assistant and 4) improving the relevance of content through applying machine learning algorithms. CONCLUSION The use of machine learning is on the rise in every aspect of today's society, however few attempts have been undertaken to harness its potential to achieve health behaviour change. By sharing our intervention concept, we contribute to the ongoing dialogue on creating effective methods for promoting health and well-being in the informatics research community. Future research should focus on refining these techniques and evaluating their effectiveness in controlled and real-world circumstances.

Liu Li, Hanchun Wang, Matthew Baugh et al.

Although existing medical image segmentation methods provide impressive pixel-wise accuracy, they often neglect topological correctness, making their segmentations unusable for many downstream tasks. One option is to retrain such models whilst including a topology-driven loss component. However, this is computationally expensive and often impractical. A better solution would be to have a versatile plug-and-play topology refinement method that is compatible with any domain-specific segmentation pipeline. Directly training a post-processing model to mitigate topological errors often fails as such models tend to be biased towards the topological errors of a target segmentation network. The diversity of these errors is confined to the information provided by a labelled training set, which is especially problematic for small datasets. Our method solves this problem by training a model-agnostic topology refinement network with synthetic segmentations that cover a wide variety of topological errors. Inspired by the Stone-Weierstrass theorem, we synthesize topology-perturbation masks with randomly sampled coefficients of orthogonal polynomial bases, which ensures a complete and unbiased representation. Practically, we verified the efficiency and effectiveness of our methods as being compatible with multiple families of polynomial bases, and show evidence that our universal plug-and-play topology refinement network outperforms both existing topology-driven learning-based and post-processing methods. We also show that combining our method with learning-based models provides an effortless add-on, which can further improve the performance of existing approaches.

Ehsan Abadi, Aldo Badano, Predrag Bakic et al.

This submission comprises the proceedings of the 1st Virtual Imaging Trials in Medicine conference, organized by Duke University on April 22-24, 2024. The listed authors serve as the program directors for this conference. The VITM conference is a pioneering summit uniting experts from academia, industry and government in the fields of medical imaging and therapy to explore the transformative potential of in silico virtual trials and digital twins in revolutionizing healthcare. The proceedings are categorized by the respective days of the conference: Monday presentations, Tuesday presentations, Wednesday presentations, followed by the abstracts for the posters presented on Monday and Tuesday.

Zihan Li, Yuan Zheng, Dandan Shan et al.

Most recent scribble-supervised segmentation methods commonly adopt a CNN framework with an encoder-decoder architecture. Despite its multiple benefits, this framework generally can only capture small-range feature dependency for the convolutional layer with the local receptive field, which makes it difficult to learn global shape information from the limited information provided by scribble annotations. To address this issue, this paper proposes a new CNN-Transformer hybrid solution for scribble-supervised medical image segmentation called ScribFormer. The proposed ScribFormer model has a triple-branch structure, i.e., the hybrid of a CNN branch, a Transformer branch, and an attention-guided class activation map (ACAM) branch. Specifically, the CNN branch collaborates with the Transformer branch to fuse the local features learned from CNN with the global representations obtained from Transformer, which can effectively overcome limitations of existing scribble-supervised segmentation methods. Furthermore, the ACAM branch assists in unifying the shallow convolution features and the deep convolution features to improve model's performance further. Extensive experiments on two public datasets and one private dataset show that our ScribFormer has superior performance over the state-of-the-art scribble-supervised segmentation methods, and achieves even better results than the fully-supervised segmentation methods. The code is released at https://github.com/HUANGLIZI/ScribFormer.

Yixuan Wu, Kaiyuan Hu, Danny Z. Chen et al.

With the rapid advance of computer graphics and artificial intelligence technologies, the ways we interact with the world have undergone a transformative shift. Virtual Reality (VR) technology, aided by artificial intelligence (AI), has emerged as a dominant interaction media in multiple application areas, thanks to its advantage of providing users with immersive experiences. Among those applications, medicine is considered one of the most promising areas. In this paper, we present a comprehensive examination of the burgeoning field of AI-enhanced VR applications in medical care and services. By introducing a systematic taxonomy, we meticulously classify the pertinent techniques and applications into three well-defined categories based on different phases of medical diagnosis and treatment: Visualization Enhancement, VR-related Medical Data Processing, and VR-assisted Intervention. This categorization enables a structured exploration of the diverse roles that AI-powered VR plays in the medical domain, providing a framework for a more comprehensive understanding and evaluation of these technologies. To our best knowledge, this is the first systematic survey of AI-powered VR systems in medical settings, laying a foundation for future research in this interdisciplinary domain.

Jingyu Xu, Binbin Wu, Jiaxin Huang et al.

The medical field is one of the important fields in the application of artificial intelligence technology. With the explosive growth and diversification of medical data, as well as the continuous improvement of medical needs and challenges, artificial intelligence technology is playing an increasingly important role in the medical field. Artificial intelligence technologies represented by computer vision, natural language processing, and machine learning have been widely penetrated into diverse scenarios such as medical imaging, health management, medical information, and drug research and development, and have become an important driving force for improving the level and quality of medical services.The article explores the transformative potential of generative AI in medical imaging, emphasizing its ability to generate syntheticACM-2 data, enhance images, aid in anomaly detection, and facilitate image-to-image translation. Despite challenges like model complexity, the applications of generative models in healthcare, including Med-PaLM 2 technology, show promising results. By addressing limitations in dataset size and diversity, these models contribute to more accurate diagnoses and improved patient outcomes. However, ethical considerations and collaboration among stakeholders are essential for responsible implementation. Through experiments leveraging GANs to augment brain tumor MRI datasets, the study demonstrates how generative AI can enhance image quality and diversity, ultimately advancing medical diagnostics and patient care.

Jianfei Liu, Christopher Parnell, Ronald M. Summers

Accurate training labels are a key component for multi-class medical image segmentation. Their annotation is costly and time-consuming because it requires domain expertise. This work aims to develop a dual-branch network and automatically improve training labels for multi-class image segmentation. Transfer learning is used to train the network and improve inaccurate weak labels sequentially. The dual-branch network is first trained by weak labels alone to initialize model parameters. After the network is stabilized, the shared encoder is frozen, and strong and weak decoders are fine-tuned by strong and weak labels together. The accuracy of weak labels is iteratively improved in the fine-tuning process. The proposed method was applied to a three-class segmentation of muscle, subcutaneous and visceral adipose tissue on abdominal CT scans. Validation results on 11 patients showed that the accuracy of training labels was statistically significantly improved, with the Dice similarity coefficient of muscle, subcutaneous and visceral adipose tissue increased from 74.2% to 91.5%, 91.2% to 95.6%, and 77.6% to 88.5%, respectively (p<0.05). In comparison with our earlier method, the label accuracy was also significantly improved (p<0.05). These experimental results suggested that the combination of the dual-branch network and transfer learning is an efficient means to improve training labels for multi-class segmentation.

Yanming Zhu, Xuefei Yin, Alan Wee-Chung Liew et al.

With the rapid advancement of artificial intelligence and deep learning, medical image analysis has become a critical tool in modern healthcare, significantly improving diagnostic accuracy and efficiency. However, AI-based methods also raise serious privacy concerns, as medical images often contain highly sensitive patient information. This review offers a comprehensive overview of privacy-preserving techniques in medical image analysis, including encryption, differential privacy, homomorphic encryption, federated learning, and generative adversarial networks. We explore the application of these techniques across various medical image analysis tasks, such as diagnosis, pathology, and telemedicine. Notably, we organizes the review based on specific challenges and their corresponding solutions in different medical image analysis applications, so that technical applications are directly aligned with practical issues, addressing gaps in the current research landscape. Additionally, we discuss emerging trends, such as zero-knowledge proofs and secure multi-party computation, offering insights for future research. This review serves as a valuable resource for researchers and practitioners and can help advance privacy-preserving in medical image analysis.

Daniel Schwabe, Katinka Becker, Martin Seyferth et al.

The adoption of machine learning (ML) and, more specifically, deep learning (DL) applications into all major areas of our lives is underway. The development of trustworthy AI is especially important in medicine due to the large implications for patients' lives. While trustworthiness concerns various aspects including ethical, technical and privacy requirements, we focus on the importance of data quality (training/test) in DL. Since data quality dictates the behaviour of ML products, evaluating data quality will play a key part in the regulatory approval of medical AI products. We perform a systematic review following PRISMA guidelines using the databases PubMed and ACM Digital Library. We identify 2362 studies, out of which 62 records fulfil our eligibility criteria. From this literature, we synthesise the existing knowledge on data quality frameworks and combine it with the perspective of ML applications in medicine. As a result, we propose the METRIC-framework, a specialised data quality framework for medical training data comprising 15 awareness dimensions, along which developers of medical ML applications should investigate a dataset. This knowledge helps to reduce biases as a major source of unfairness, increase robustness, facilitate interpretability and thus lays the foundation for trustworthy AI in medicine. Incorporating such systematic assessment of medical datasets into regulatory approval processes has the potential to accelerate the approval of ML products and builds the basis for new standards.

Kamal Kant Sharma, Jeeva Somasundaram, Ashish Sachdeva

BackgroundSmartphones have become integral to people’s lives, with a noticeable increase in the average screen time, both on a global scale and, notably, in India. Existing research links mobile consumption to sleep problems, poor physical and mental health, and lower subjective well-being. The comparative effectiveness of monetary incentives given for self-selected versus assigned targets on reducing screen time and thereby improving mental health remains unanswered. ObjectiveThis study aims to assess the impact of monetary incentives and target selection on mobile screen time reduction and mental health. MethodsWe designed a 3-armed randomized controlled trial conducted with employees and students at an educational institution in India. The study is conducted digitally over 12 weeks, including baseline (2 weeks), randomization (1 week), intervention (5 weeks), and postintervention (4 week) periods. We emailed the employees and students to inquire about their interest in participation. Those who expressed interest received detailed study information and consent forms. After securing consent, participants were asked to complete the initial survey and provide their mobile screen time during the baseline period. At the beginning of the intervention period, the participants were randomly allocated into 1 of 3 study groups in a 2:2:1 ratio (self-selected vs assigned vs control). Participants in the self-selected group were presented with 3 target options: 10%, 20%, and 30%, and they were asked to self-select a target to reduce their mobile screen time from their baseline average mobile screen time. Participants in the assigned group were given a target to reduce their mobile screen time from their baseline average mobile screen time. The assigned target was set as the average of the targets selected by participants in the self-selected group. During the intervention period, participants in the self-selected and assigned group were eligible to receive a monetary incentive of INR (Indian Rupee) 50 (US $0.61) per day for successfully attaining their target. Participants in the control group neither received nor selected a target for reducing their mobile screen time and did not receive any monetary incentives during the intervention period. All participants received information regarding the advantages of reducing mobile screen time. As an incentive, all participants would receive INR 500 (US $6.06) upon completion of the study and a chance to win 1 of 2 lotteries valued at INR 5000 (US $60.55) for consistently sharing their mobile screen time data. ResultsCurrently, the study intervention is being rolled out. Enrollment occurred between August 21, 2023, and September 2, 2023; data collection concluded in November 2023. We expect that results will be available by early 2024. ConclusionsThe monetary incentives and self-selected versus assigned targets might be effective interventions in reducing mobile screen time among working professionals and students. Trial RegistrationAsPredicted 142497; https://aspredicted.org/hr3nn.pdf International Registered Report Identifier (IRRID)DERR1-10.2196/53756

Franklin D. Gaylis, Bruno Emond, Ameur M. Manceur et al.

# Background Bacillus Calmette-Guérin (BCG) can reduce recurrence and delay progression among patients with high-risk non–muscle invasive bladder cancer (NMIBC), but is associated with a substantial emotional, physical, and social burden. # Objectives This study evaluated the adequacy of first-line intravesical BCG treatment among high-risk NMIBC patients in the United States, including the subgroup with carcinoma in situ (CIS) of the bladder. # Methods Adults with high-risk NMIBC treated with BCG were selected from de-identified MarketScan® Commercial, Medicare, and Medicaid Databases (1/1/2010-2/28/2021). Adequacy of BCG induction and maintenance was evaluated from the first BCG claim until the end of the patient’s observation, using a previously published claims-based algorithm (induction: ≥5 instillations within 70 days; induction and maintenance: ≥7 instillations within 274 days of first instillation) and a definition based on the landmark Southwest Oncology Group (SWOG) trial (induction: ≥5 instillations without gaps >7 days; followed by ≥2 instillations at month 3, 6, and every 6 months thereafter). Proportions of patients with adequate BCG induction and maintenance were reported overall and compared between those with and without CIS. # Results Of 5803 high-risk NMIBC patients treated with first-line BCG (mean age, 67.3 years; 20.6% female), 930 (16.0%) had CIS. After first-line BCG, 56.6% received another treatment. Although 86.9% had adequate BCG induction based on the claims-based algorithm (SWOG, 73.6%), only 41.5% had adequate BCG induction and maintenance (SWOG, 1.6%). Similar trends were observed for patients with and without CIS, with higher adherence to guidelines for patients with CIS (adequate induction using claims-based algorithm: 90.3% vs 86.2%; adequate induction and maintenance: 50.8% vs 39.7%, all _P_ < .001). A greater proportion of CIS patients than non-CIS patients had cystectomy (CIS, 14.4%, non-CIS, 8.5%; _P_ < .001) after first-line BCG. # Discussion Among patients with NMIBC treated with first-line intravesical BCG, most received adequate BCG induction but less than half had adequate BCG maintenance. BCG treatment was also inadequate for patients with CIS, with only half of patients receiving adequate BCG maintenance and a higher proportion undergoing cystectomy following first-line BCG. # Conclusions Results emphasize the need for additional treatment options for patients with NMIBC.

Harshana Liyanage, S. Liaw, Jitendra Jonnagaddala et al.

Summary Background: Artificial intelligence (AI) is heralded as an approach that might augment or substitute for the limited processing power of the human brain of primary health care (PHC) professionals. However, there are concerns that AI-mediated decisions may be hard to validate and challenge, or may result in rogue decisions. Objective: To form consensus about perceptions, issues, and challenges of AI in primary care. Method: A three-round Delphi study was conducted. Round 1 explored experts’ viewpoints on AI in PHC (n=20). Round 2 rated the appropriateness of statements arising from round one (n=12). The third round was an online panel discussion of findings (n=8) with the members of both the International Medical Informatics Association and the European Federation of Medical Informatics Primary Health Care Informatics Working Groups. Results: PHC and informatics experts reported AI has potential to improve managerial and clinical decisions and processes, and this would be facilitated by common data standards. The respondents did not agree that AI applications should learn and adapt to clinician preferences or behaviour and they did not agree on the extent of AI potential for harm to patients. It was more difficult to assess the impact of AI-based applications on continuity and coordination of care. Conclusion: While the use of AI in medicine should enhance healthcare delivery, we need to ensure meticulous design and evaluation of AI applications. The primary care informatics community needs to be proactive and to guide the ethical and rigorous development of AI applications so that they will be safe and effective.

Nephi Walton, R. Nagarajan, Chen Wang et al.

Abstract Objective Given the importance AI in genomics and its potential impact on human health, the American Medical Informatics Association—Genomics and Translational Biomedical Informatics (GenTBI) Workgroup developed this assessment of factors that can further enable the clinical application of AI in this space. Process A list of relevant factors was developed through GenTBI workgroup discussions in multiple in-person and online meetings, along with review of pertinent publications. This list was then summarized and reviewed to achieve consensus among the group members. Conclusions Substantial informatics research and development are needed to fully realize the clinical potential of such technologies. The development of larger datasets is crucial to emulating the success AI is achieving in other domains. It is important that AI methods do not exacerbate existing socio-economic, racial, and ethnic disparities. Genomic data standards are critical to effectively scale such technologies across institutions. With so much uncertainty, complexity and novelty in genomics and medicine, and with an evolving regulatory environment, the current focus should be on using these technologies in an interface with clinicians that emphasizes the value each brings to clinical decision-making.