Hasil untuk "Computer applications to medicine. Medical informatics"

Amisha, Paras Malik, Monika Pathania et al.

Background: Artificial intelligence (AI) is the term used to describe the use of computers and technology to simulate intelligent behavior and critical thinking comparable to a human being. John McCarthy first described the term AI in 1956 as the science and engineering of making intelligent machines. Objective: This descriptive article gives a broad overview of AI in medicine, dealing with the terms and concepts as well as the current and future applications of AI. It aims to develop knowledge and familiarity of AI among primary care physicians. Materials and Methods: PubMed and Google searches were performed using the key words 'artificial intelligence'. Further references were obtained by cross-referencing the key articles. Results: Recent advances in AI technology and its current applications in the field of medicine have been discussed in detail. Conclusions: AI promises to change the practice of medicine in hitherto unknown ways, but many of its practical applications are still in their infancy and need to be explored and developed better. Medical professionals also need to understand and acclimatize themselves with these advances for better healthcare delivery to the masses.

Xiaoyan Zhang, Dan Yang, Sihan Chen et al.

Abstract BackgroundSedentary behavior (SB) is a modifiable risk factor for complications in older adults with type 2 diabetes mellitus (T2DM). Despite widespread adoption of digital health platforms, theory-driven telehealth interventions specifically targeting SB reduction remain limited, particularly those incorporating cultural adaptation and behavioral change frameworks. ObjectiveThis study aims to develop and evaluate the feasibility of a theory-based personalized telehealth intervention to reduce SB in older adults with T2DM in China. MethodsThe intervention was developed over 14 months (January 2022-February 2023) following the intervention mapping and Behavior Change Wheel frameworks. A panel of 19 multidisciplinary experts (90.5% response rate) refined the program through a systematic iterative process. Subsequently, a 7-week quasi-experimental study (pre-post self-controlled design) was conducted to assess feasibility. We recruited 30 community-dwelling older adults with T2DM via WeChat-based convenience sampling. The primary outcome was SB measured by the Measure of Older Adults’ Sedentary Time for Type 2 Diabetes Mellitus questionnaire. Secondary outcomes included cardiovascular risk (blood pressure), glycemic control (fasting blood glucose), Diabetes-Specific Quality of Life, social isolation, BMI, and fall incidence. Pre-post changes from baseline to 7 weeks were statistically evaluated to assess the intervention’s feasibility and preliminary impact. ResultsThe intervention comprises 5 components: an eHealth education manual, a motion graphics library, an SMS text messaging library, a WeChat Q&A group, and a material incentive package. These components address “knowledge,” “social support,” and “intention” determinants through “education,” “enablement,” and “incentivisation” functions, respectively. All components used the “service provision” policy and various behavior change techniques. Preliminary feasibility testing (n=31) showed reduced sedentary time by 1.12 hours/day (PP ConclusionsThis study demonstrates the feasibility and potential impact of a systematically developed telehealth intervention for reducing SB in older adults with T2DM in China. The integration of intervention mapping with the Behavior Change Wheel provides a replicable framework for developing theory-driven digital health interventions. With significant reductions in sedentary time and improved social connectivity, this culturally adapted approach offers a scalable model for chronic disease self-management in aging populations. The systematic methodology and positive preliminary outcomes support further large-scale evaluation of evidence-based telehealth solutions for behavioral modification in diabetes care.

Saleh Alrashed, Nasro Min-Allah

Nana A Mensah, Michael J Fassett, Nehaa Khadka et al.

BackgroundHemolytic disease of the fetus and newborn (HDFN) is a rare but serious condition caused by maternal-fetal red blood cell antigen incompatibility. In an affected pregnancy, maternal immunoglobulin G antibodies cross the placenta and target fetal or neonatal red blood cells, leading to hemolysis, hyperbilirubinemia, and anemia. Although routine screening and alloimmunization prevention programs have contributed to the decline in HDFN in the United States, further understanding of its epidemiology is still needed. ObjectiveThis protocol aims to provide an overview of the study design, methodology, and analytical approach used to investigate the epidemiology, treatment, and health care resource use of HDFN within a large integrated health care system. MethodsWe conducted a retrospective cohort study of pregnant women who received obstetric care in the Kaiser Permanente Southern California (KPSC) health care system from January 1, 2008, to June 30, 2022. To identify HDFN cases, we used a novel methodology developed by KPSC researchers combining structured data and detailed clinical information extracted from unstructured records via a natural language processing–assisted chart review process. Chi-square and Wilcoxon rank sum tests were used to compare the distributions of maternal and infant demographic characteristics, as well as medical and perinatal conditions, by HDFN status. We also evaluated the association between HDFN and adverse perinatal outcomes using logistic regression models. Planned analyses using this unique cohort will include describing the annual prevalence, health care resource use, and treatment patterns of mothers and infants by HDFN status. ResultsThe study population consisted of 464,711 pregnancies, of which 136 (0.03%) were HDFN cases confirmed by chart review, resulting in 138 (0.03%) births (n=137, 0.99% live births and n=1, 0.01% stillbirth). The mean age at pregnancy was 29.8 (SD 5.7) years, and the population was racially and ethnically diverse. ConclusionsWe present an overview of the methodology developed by KPSC clinicians and researchers on the epidemiology, treatment, and health care resource use of HDFN within a large and demographically diverse population of pregnant women. Our novel methodology, combining both structured and unstructured data and a natural language processing–assisted chart review process, ensures the successful identification of true cases to carry out pharmaco-epidemiological studies. International Registered Report Identifier (IRRID)DERR1-10.2196/77836

Catherine L. Jan, Sanil Joseph, Algis J. Vingrys et al.

Abstract There are no prospective clinical studies evaluating artificial intelligence implementation for glaucoma detection in real-world settings. We developed an automated retinal photography and AI-based screening system and prospectively assessed its accuracy, feasibility, and acceptability in Australian general practice (GP) clinics. Adults aged 50 years or older were recruited during routine GP visits, with retinal images captured using an automated fundus camera and analysed by the AI system for glaucoma risk classification. Of 414 participants, 277 (66.9%) had analysable images, with a total of 483 eyes included. The AI system achieved an AUROC of 0.80, sensitivity of 65.0%, and specificity of 94.6%. Among 161 previously undiagnosed patients, 18 (11.2%) were identified as referable glaucoma. Patient feedback was positive, and clinic staff supported AI-assisted screening to enhance glaucoma care. Despite challenges such as lower sensitivity and image acquisition limitations, the system shows promise for opportunistic screening in primary care settings.

Caner Özer, Patryk Rygiel, Bram de Wilde et al.

Artifacts pose a significant challenge in medical imaging, impacting diagnostic accuracy and downstream analysis. While image-based approaches for detecting artifacts can be effective, they often rely on preprocessing methods that can lead to information loss and high-memory-demand medical images, thereby limiting the scalability of classification models. In this work, we propose the use of implicit neural representations (INRs) for image quality assessment. INRs provide a compact and continuous representation of medical images, naturally handling variations in resolution and image size while reducing memory overhead. We develop deep weight space networks, graph neural networks, and relational attention transformers that operate on INRs to achieve image quality assessment. Our method is evaluated on the ACDC dataset with synthetically generated artifact patterns, demonstrating its effectiveness in assessing image quality while achieving similar performance with fewer parameters.

Van-Tinh Nguyen, Hoang-Duong Pham, Thanh-Hai To et al.

Understanding medical texts presents significant challenges due to complex terminology and context-specific language. This paper introduces Medalyze, an AI-powered application designed to enhance the comprehension of medical texts using three specialized FLAN-T5-Large models. These models are fine-tuned for (1) summarizing medical reports, (2) extracting health issues from patient-doctor conversations, and (3) identifying the key question in a passage. Medalyze is deployed across a web and mobile platform with real-time inference, leveraging scalable API and YugabyteDB. Experimental evaluations demonstrate the system's superior summarization performance over GPT-4 in domain-specific tasks, based on metrics like BLEU, ROUGE-L, BERTScore, and SpaCy Similarity. Medalyze provides a practical, privacy-preserving, and lightweight solution for improving information accessibility in healthcare.

Katharina Eckstein, Constantin Ulrich, Michael Baumgartner et al.

Large-scale pre-training holds the promise to advance 3D medical object detection, a crucial component of accurate computer-aided diagnosis. Yet, it remains underexplored compared to segmentation, where pre-training has already demonstrated significant benefits. Existing pre-training approaches for 3D object detection rely on 2D medical data or natural image pre-training, failing to fully leverage 3D volumetric information. In this work, we present the first systematic study of how existing pre-training methods can be integrated into state-of-the-art detection architectures, covering both CNNs and Transformers. Our results show that pre-training consistently improves detection performance across various tasks and datasets. Notably, reconstruction-based self-supervised pre-training outperforms supervised pre-training, while contrastive pre-training provides no clear benefit for 3D medical object detection. Our code is publicly available at: https://github.com/MIC-DKFZ/nnDetection-finetuning.

Jiaqi Liu, Tong Wang, Su Liu et al.

The research evaluates lightweight medical abstract classification methods to establish their maximum performance capabilities under financial budget restrictions. On the public medical abstracts corpus, we finetune BERT base and Distil BERT with three objectives cross entropy (CE), class weighted CE, and focal loss under identical tokenization, sequence length, optimizer, and schedule. DistilBERT with plain CE gives the strongest raw argmax trade off, while a post hoc operating point selection (validation calibrated, classwise thresholds) sub stantially improves deployed performance; under this tuned regime, focal benefits most. We report Accuracy, Macro F1, and WeightedF1, release evaluation artifacts, and include confusion analyses to clarify error structure. The practical takeaway is to start with a compact encoder and CE, then add lightweight calibration or thresholding when deployment requires higher macro balance.

Dawei Li, Zijian Gu, Peng Wang et al.

Multimodal large language models (MLLMs) have shown strong potential for medical image reasoning, yet fairness across demographic groups remains a major concern. Existing debiasing methods often rely on large labeled datasets or fine-tuning, which are impractical for foundation-scale models. We explore In-Context Learning (ICL) as a lightweight, tuning-free alternative for improving fairness. Through systematic analysis, we find that conventional demonstration selection (DS) strategies fail to ensure fairness due to demographic imbalance in selected exemplars. To address this, we propose Fairness-Aware Demonstration Selection (FADS), which builds demographically balanced and semantically relevant demonstrations via clustering-based sampling. Experiments on multiple medical imaging benchmarks show that FADS consistently reduces gender-, race-, and ethnicity-related disparities while maintaining strong accuracy, offering an efficient and scalable path toward fair medical image reasoning. These results highlight the potential of fairness-aware in-context learning as a scalable and data-efficient solution for equitable medical image reasoning.

Eileen Marie Hanna, Ghadi El Hasbani, Danielle Azar

Abstract Background High-throughput experimental technologies can provide deeper insights into pathway perturbations in biomedical studies. Accordingly, their usage is central to the identification of molecular targets and the subsequent development of suitable treatments for various diseases. Classical interpretations of generated data, such as differential gene expression and pathway analyses, disregard interconnections between studied genes when looking for gene-disease associations. Given that these interconnections are central to cellular processes, there has been a recent interest in incorporating them in such studies. The latter allows the detection of gene modules that underlie complex phenotypes in gene interaction networks. Existing methods either impose radius-based restrictions or freely grow modules at the expense of a statistical bias towards large modules. We propose a heuristic method, inspired by Ant Colony Optimization, to apply gene-level scoring and module identification with distance-based search constraints and penalties, rather than radius-based constraints. Results We test and compare our results to other approaches using three datasets of different neurodegenerative diseases, namely Alzheimer’s, Parkinson’s, and Huntington’s, over three independent experiments. We report the outcomes of enrichment analyses and concordance of gene-level scores for each disease. Results indicate that the proposed approach generally shows superior stability in comparison to existing methods. It produces stable and meaningful enrichment results in all three datasets which have different case to control proportions and sample sizes. Conclusion The presented network-based gene expression analysis approach successfully identifies dysregulated gene modules associated with a certain disease. Using a heuristic based on Ant Colony Optimization, we perform a distance-based search with no radius constraints. Experimental results support the effectiveness and stability of our method in prioritizing modules of high relevance. Our tool is publicly available at github.com/GhadiElHasbani/ACOxGS.git.

Adam Tupper, Christian Gagné

Data augmentation is one of the most effective techniques to improve the generalization performance of deep neural networks. Yet, despite often facing limited data availability in medical image analysis, it is frequently underutilized. This appears to be due to a gap in our collective understanding of the efficacy of different augmentation techniques across medical imaging tasks and modalities. One domain where this is especially true is breast ultrasound images. This work addresses this issue by analyzing the effectiveness of different augmentation techniques for the classification of breast lesions in ultrasound images. We assess the generalizability of our findings across several datasets, demonstrate that certain augmentations are far more effective than others, and show that their usage leads to significant performance gains.

Julia Jentsch, Ali Burak Ünal, Şeyma Selcan Mağara et al.

Handling missing data is crucial in machine learning, but many datasets contain gaps due to errors or non-response. Unlike traditional methods such as listwise deletion, which are simple but inadequate, the literature offers more sophisticated and effective methods, thereby improving sample size and accuracy. However, these methods require accessing the whole dataset, which contradicts the privacy regulations when the data is distributed among multiple sources. Especially in the medical and healthcare domain, such access reveals sensitive information about patients. This study addresses privacy-preserving imputation methods for sensitive data using secure multi-party computation, enabling secure computations without revealing any party's sensitive information. In this study, we realized the mean, median, regression, and kNN imputation methods in a privacy-preserving way. We specifically target the medical and healthcare domains considering the significance of protection of the patient data, showcasing our methods on a diabetes dataset. Experiments on the diabetes dataset validated the correctness of our privacy-preserving imputation methods, yielding the largest error around $3 \times 10^{-3}$, closely matching plaintext methods. We also analyzed the scalability of our methods to varying numbers of samples, showing their applicability to real-world healthcare problems. Our analysis demonstrated that all our methods scale linearly with the number of samples. Except for kNN, the runtime of all our methods indicates that they can be utilized for large datasets.

Zelong Liu, Andrew Tieu, Nikhil Patel et al.

Artificial Intelligence (AI) has the potential to revolutionize diagnosis and segmentation in medical imaging. However, development and clinical implementation face multiple challenges including limited data availability, lack of generalizability, and the necessity to incorporate multi-modal data effectively. A foundation model, which is a large-scale pre-trained AI model, offers a versatile base that can be adapted to a variety of specific tasks and contexts. Here, we present VIsualization and Segmentation Masked AutoEncoder (VIS-MAE), novel model weights specifically designed for medical imaging. Specifically, VIS-MAE is trained on a dataset of 2.5 million unlabeled images from various modalities (CT, MR, PET,X-rays, and ultrasound), using self-supervised learning techniques. It is then adapted to classification and segmentation tasks using explicit labels. VIS-MAE has high label efficiency, outperforming several benchmark models in both in-domain and out-of-domain applications. In addition, VIS-MAE has improved label efficiency as it can achieve similar performance to other models with a reduced amount of labeled training data (50% or 80%) compared to other pre-trained weights. VIS-MAE represents a significant advancement in medical imaging AI, offering a generalizable and robust solution for improving segmentation and classification tasks while reducing the data annotation workload. The source code of this work is available at https://github.com/lzl199704/VIS-MAE.

Dani Kiyasseh, Jasper Laca, Taseen F. Haque et al.

Abstract Artificial intelligence (AI) systems can now reliably assess surgeon skills through videos of intraoperative surgical activity. With such systems informing future high-stakes decisions such as whether to credential surgeons and grant them the privilege to operate on patients, it is critical that they treat all surgeons fairly. However, it remains an open question whether surgical AI systems exhibit bias against surgeon sub-cohorts, and, if so, whether such bias can be mitigated. Here, we examine and mitigate the bias exhibited by a family of surgical AI systems—SAIS—deployed on videos of robotic surgeries from three geographically-diverse hospitals (USA and EU). We show that SAIS exhibits an underskilling bias, erroneously downgrading surgical performance, and an overskilling bias, erroneously upgrading surgical performance, at different rates across surgeon sub-cohorts. To mitigate such bias, we leverage a strategy —TWIX—which teaches an AI system to provide a visual explanation for its skill assessment that otherwise would have been provided by human experts. We show that whereas baseline strategies inconsistently mitigate algorithmic bias, TWIX can effectively mitigate the underskilling and overskilling bias while simultaneously improving the performance of these AI systems across hospitals. We discovered that these findings carry over to the training environment where we assess medical students’ skills today. Our study is a critical prerequisite to the eventual implementation of AI-augmented global surgeon credentialing programs, ensuring that all surgeons are treated fairly.

Evi M. C. Huijben, Sina Amirrajab, Josien P. W. Pluim

Out-of-distribution (OOD) detection is crucial for the safety and reliability of artificial intelligence algorithms, especially in the medical domain. In the context of the Medical OOD (MOOD) detection challenge 2023, we propose a pipeline that combines a histogram-based method and a diffusion-based method. The histogram-based method is designed to accurately detect homogeneous anomalies in the toy examples of the challenge, such as blobs with constant intensity values. The diffusion-based method is based on one of the latest methods for unsupervised anomaly detection, called DDPM-OOD. We explore this method and propose extensive post-processing steps for pixel-level and sample-level anomaly detection on brain MRI and abdominal CT data provided by the challenge. Our results show that the proposed DDPM method is sensitive to blur and bias field samples, but faces challenges with anatomical deformation, black slice, and swapped patches. These findings suggest that further research is needed to improve the performance of DDPM for OOD detection in medical images.

Charlene Esteban Ronquillo First Co-Author, James Mitchell First Co-Author, Dari Alhuwail et al.

Summary Objectives : The objective of this paper is to draw attention to the currently underused potential of clinical documentation by nursing and allied health professions to improve the representation of social determinants of health (SDoH) and intersectionality data in electronic health records (EHRs), towards the development of equitable artificial intelligence (AI) technologies. Methods : A rapid review of the literature on the inclusion of nursing and allied health data and the nature of health equity information representation in the development and/or use of artificial intelligence approaches alongside expert perspectives from the International Medical Informatics Association (IMIA) Student and Emerging Professionals Working Group. Results : Consideration of social determinants of health and intersectionality data are limited in both the medical AI and nursing and allied health AI literature. As a concept being newly discussed in the context of AI, the lack of discussion of intersectionality in the literature was unsurprising. However, the limited consideration of social determinants of health was surprising, given its relatively longstanding recognition and the importance of representation of the features of diverse populations as a key requirement for equitable AI. Conclusions : Leveraging the rich contextual data collected by nursing and allied health professions has the potential to improve the capture and representation of social determinants of health and intersectionality. This will require addressing issues related to valuing AI goals (e.g., diagnostics versus supporting care delivery) and improved EHR infrastructure to facilitate documentation of data beyond medicine. Leveraging nursing and allied health data to support equitable AI development represents a current open question for further exploration and research.