BMJ Health & Care Informatics最新文献_第5页

Prediction of high-risk emergency department revisits from a machine-learning algorithm: a proof-of-concept study 通过机器学习算法预测高风险急诊科复诊：概念验证研究

IF 4.1 Q2 Computer Science

BMJ Health & Care Informatics

Pub Date : 2024-04-01 DOI: 10.1136/bmjhci-2023-100859

Chih-Wei Sung, Joshua Ho, Cheng-Yi Fan, Ching-Yu Chen, Chi-Hsin Chen, Shao-Yung Lin, Jia-How Chang, Jiun-Wei Chen, Edward Pei-Chuan Huang

Background High-risk emergency department (ED) revisit is considered an important quality indicator that may reflect an increase in complications and medical burden. However, because of its multidimensional and highly complex nature, this factor has not been comprehensively investigated. This study aimed to predict high-risk ED revisit with a machine-learning (ML) approach. Methods This 3-year retrospective cohort study assessed adult patients between January 2019 and December 2021 from National Taiwan University Hospital Hsin-Chu Branch with high-risk ED revisit, defined as hospital or intensive care unit admission after ED return within 72 hours. A total of 150 features were preliminarily screened, and 79 were used in the prediction model. Deep learning, random forest, extreme gradient boosting (XGBoost) and stacked ensemble algorithm were used. The stacked ensemble model combined multiple ML models and performed model stacking as a meta-level algorithm. Confusion matrix, accuracy, sensitivity, specificity and area under the receiver operating characteristic curve (AUROC) were used to evaluate performance. Results Analysis was performed for 6282 eligible adult patients: 5025 (80.0%) in the training set and 1257 (20.0%) in the testing set. High-risk ED revisit occurred for 971 (19.3%) of training set patients vs 252 (20.1%) in the testing set. Leading predictors of high-risk ED revisit were age, systolic blood pressure and heart rate. The stacked ensemble model showed more favourable prediction performance (AUROC 0.82) than the other models: deep learning (0.69), random forest (0.78) and XGBoost (0.79). Also, the stacked ensemble model achieved favourable accuracy and specificity. Conclusion The stacked ensemble algorithm exhibited better prediction performance in which the predictions were generated from different ML algorithms to optimally maximise the final set of results. Patients with older age and abnormal systolic blood pressure and heart rate at the index ED visit were vulnerable to high-risk ED revisit. Further studies should be conducted to externally validate the model. Data are available on reasonable request.

背景高风险急诊科（ED）再次就诊被认为是一项重要的质量指标，可能反映出并发症和医疗负担的增加。然而，由于其多维性和高度复杂性，这一因素尚未得到全面研究。本研究旨在通过机器学习（ML）方法预测高风险急诊室复诊率。方法这项为期 3 年的回顾性队列研究评估了 2019 年 1 月至 2021 年 12 月期间台大医院新竹分院的高风险 ED 再就诊成人患者。共初步筛选出 150 个特征，其中 79 个用于预测模型。使用了深度学习、随机森林、极梯度提升（XGBoost）和堆叠集合算法。堆叠集合模型结合了多个 ML 模型，作为元级算法进行模型堆叠。混淆矩阵、准确率、灵敏度、特异性和接收者工作特征曲线下面积（AUROC）用于评估性能。结果对 6282 名符合条件的成年患者进行了分析：其中 5025 人（80.0%）在训练集中，1257 人（20.0%）在测试集中。训练集患者中有 971 人（19.3%）再次到急诊室就诊，而测试集患者中有 252 人（20.1%）再次到急诊室就诊。高风险急诊室复诊的主要预测因素是年龄、收缩压和心率。与深度学习（0.69）、随机森林（0.78）和 XGBoost（0.79）等其他模型相比，堆叠集合模型的预测性能更佳（AUROC 0.82）。此外，堆叠集合模型的准确性和特异性也很高。结论叠加集合算法显示出更好的预测性能，其中的预测由不同的多重学习算法生成，以优化最大化最终结果集。在急诊室就诊时年龄较大、收缩压和心率异常的患者很容易在急诊室再次就诊。应开展进一步研究，从外部验证该模型。如有合理要求，可提供相关数据。

{"title":"Prediction of high-risk emergency department revisits from a machine-learning algorithm: a proof-of-concept study","authors":"Chih-Wei Sung, Joshua Ho, Cheng-Yi Fan, Ching-Yu Chen, Chi-Hsin Chen, Shao-Yung Lin, Jia-How Chang, Jiun-Wei Chen, Edward Pei-Chuan Huang","doi":"10.1136/bmjhci-2023-100859","DOIUrl":"https://doi.org/10.1136/bmjhci-2023-100859","url":null,"abstract":"Background High-risk emergency department (ED) revisit is considered an important quality indicator that may reflect an increase in complications and medical burden. However, because of its multidimensional and highly complex nature, this factor has not been comprehensively investigated. This study aimed to predict high-risk ED revisit with a machine-learning (ML) approach. Methods This 3-year retrospective cohort study assessed adult patients between January 2019 and December 2021 from National Taiwan University Hospital Hsin-Chu Branch with high-risk ED revisit, defined as hospital or intensive care unit admission after ED return within 72 hours. A total of 150 features were preliminarily screened, and 79 were used in the prediction model. Deep learning, random forest, extreme gradient boosting (XGBoost) and stacked ensemble algorithm were used. The stacked ensemble model combined multiple ML models and performed model stacking as a meta-level algorithm. Confusion matrix, accuracy, sensitivity, specificity and area under the receiver operating characteristic curve (AUROC) were used to evaluate performance. Results Analysis was performed for 6282 eligible adult patients: 5025 (80.0%) in the training set and 1257 (20.0%) in the testing set. High-risk ED revisit occurred for 971 (19.3%) of training set patients vs 252 (20.1%) in the testing set. Leading predictors of high-risk ED revisit were age, systolic blood pressure and heart rate. The stacked ensemble model showed more favourable prediction performance (AUROC 0.82) than the other models: deep learning (0.69), random forest (0.78) and XGBoost (0.79). Also, the stacked ensemble model achieved favourable accuracy and specificity. Conclusion The stacked ensemble algorithm exhibited better prediction performance in which the predictions were generated from different ML algorithms to optimally maximise the final set of results. Patients with older age and abnormal systolic blood pressure and heart rate at the index ED visit were vulnerable to high-risk ED revisit. Further studies should be conducted to externally validate the model. Data are available on reasonable request.","PeriodicalId":9050,"journal":{"name":"BMJ Health & Care Informatics","volume":"8 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140636847","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}

引用次数: 0

‘If you build it, they will come…to the wrong door: evaluating patient and caregiver-initiated ethics consultations via a patient portal’ 如果你建造了它，他们就会来......走错门：通过患者门户网站评估患者和护理人员发起的伦理咨询

IF 4.1 Q2 Computer Science

BMJ Health & Care Informatics

Pub Date : 2024-04-01 DOI: 10.1136/bmjhci-2023-100988

Liz Blackler, Amy E Scharf, Konstantina Matsoukas, Michelle Colletti, Louis P Voigt

Objectives Memorial Sloan Kettering Cancer Center (MSK) sought to empower patients and caregivers to be more proactive in requesting ethics consultations. Methods Functionality was developed on MSK’s electronic patient portal that allowed patients and/or caregivers to request ethics consultations. The Ethics Consultation Service (ECS) responded to all requests, which were documented and analysed. Results Of the 74 requests made through the portal, only one fell under the purview of the ECS. The others were primarily requests for assistance with coordinating clinical care, hospital resources or frustrations with the hospital or clinical team. Discussion To better empower patients and caregivers to engage Ethics, healthcare organisations and ECSs must first provide them with accessible, understandable and iterative educational resources. Conclusion After 19.5 months, the ‘Request Ethics Consultation’ functionality on the patient portal was suspended. Developing resources on the role of Ethics for our patients and caregivers remains a priority. All data relevant to the study are included in the article or uploaded as supplementary information.

目的纪念斯隆-凯特琳癌症中心（MSK）希望增强患者和护理人员的能力，使他们能够更加积极主动地请求伦理咨询。方法在 MSK 的电子患者门户网站上开发了允许患者和/或护理人员请求伦理会诊的功能。伦理会诊服务（ECS）对所有请求做出了回应，并对这些请求进行了记录和分析。结果在通过门户网站提出的 74 项请求中，只有一项属于伦理咨询服务的范围。其他请求主要是要求协助协调临床护理、医院资源或对医院或临床团队的不满。讨论为了更好地增强患者和护理人员参与伦理的能力，医疗机构和 ECS 必须首先为他们提供方便、易懂和可重复的教育资源。结论在 19.5 个月后，患者门户网站上的 "请求伦理咨询 "功能被暂停。为患者和护理人员开发有关伦理作用的资源仍是当务之急。所有与研究相关的数据均已包含在文章中或作为补充信息上传。

{"title":"‘If you build it, they will come…to the wrong door: evaluating patient and caregiver-initiated ethics consultations via a patient portal’","authors":"Liz Blackler, Amy E Scharf, Konstantina Matsoukas, Michelle Colletti, Louis P Voigt","doi":"10.1136/bmjhci-2023-100988","DOIUrl":"https://doi.org/10.1136/bmjhci-2023-100988","url":null,"abstract":"Objectives Memorial Sloan Kettering Cancer Center (MSK) sought to empower patients and caregivers to be more proactive in requesting ethics consultations. Methods Functionality was developed on MSK’s electronic patient portal that allowed patients and/or caregivers to request ethics consultations. The Ethics Consultation Service (ECS) responded to all requests, which were documented and analysed. Results Of the 74 requests made through the portal, only one fell under the purview of the ECS. The others were primarily requests for assistance with coordinating clinical care, hospital resources or frustrations with the hospital or clinical team. Discussion To better empower patients and caregivers to engage Ethics, healthcare organisations and ECSs must first provide them with accessible, understandable and iterative educational resources. Conclusion After 19.5 months, the ‘Request Ethics Consultation’ functionality on the patient portal was suspended. Developing resources on the role of Ethics for our patients and caregivers remains a priority. All data relevant to the study are included in the article or uploaded as supplementary information.","PeriodicalId":9050,"journal":{"name":"BMJ Health & Care Informatics","volume":"34 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140806563","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}

引用次数: 0

Codesign of health technology interventions to support best-practice perioperative care and surgical waitlist management. 对医疗技术干预措施进行代码设计，以支持最佳实践围手术期护理和手术候诊名单管理。

IF 4.1 Q2 Computer Science

BMJ Health & Care Informatics

Pub Date : 2024-03-12 DOI: 10.1136/bmjhci-2023-100928

Sarah Joy Aitken, Sophie James, Amy Lawrence, Anthony Glover, Henry Pleass, Janani Thillianadesan, Sue Monaro, Kerry Hitos, Vasi Naganathan

Objectives: This project aimed to determine where health technology can support best-practice perioperative care for patients waiting for surgery.

Methods: An exploratory codesign process used personas and journey mapping in three interprofessional workshops to identify key challenges in perioperative care across four health districts in Sydney, Australia. Through participatory methodology, the research inquiry directly involved perioperative clinicians. In three facilitated workshops, clinician and patient participants codesigned potential digital interventions to support perioperative pathways. Workshop output was coded and thematically analysed, using design principles.

Results: Codesign workshops, involving 51 participants, were conducted October to November 2022. Participants designed seven patient personas, with consumer representatives confirming acceptability and diversity. Interprofessional team members and consumers mapped key clinical moments, feelings and barriers for each persona during a hypothetical perioperative journey. Six key themes were identified: 'preventative care', 'personalised care', 'integrated communication', 'shared decision-making', 'care transitions' and 'partnership'. Twenty potential solutions were proposed, with top priorities a digital dashboard and virtual care coordination.

Discussion: Our findings emphasise the importance of interprofessional collaboration, patient and family engagement and supporting health technology infrastructure. Through user-based codesign, participants identified potential opportunities where health technology could improve system efficiencies and enhance care quality for patients waiting for surgical procedures. The codesign approach embedded users in the development of locally-driven, contextually oriented policies to address current perioperative service challenges, such as prolonged waiting times and care fragmentation.

Conclusion: Health technology innovation provides opportunities to improve perioperative care and integrate clinical information. Future research will prototype priority solutions for further implementation and evaluation.

目标该项目旨在确定医疗技术在哪些方面可以为等待手术的病人提供最佳的围手术期护理：方法：在三个跨专业研讨会上，采用角色和旅程映射进行了探索性的代码设计过程，以确定澳大利亚悉尼四个医疗区围手术期护理所面临的主要挑战。通过参与式方法，围手术期临床医生直接参与了研究调查。在三场研讨会上，临床医生和患者参与者对支持围手术期路径的潜在数字干预措施进行了编码。利用设计原则对研讨会成果进行编码和主题分析：编码设计研讨会于 2022 年 10 月至 11 月举行，共有 51 人参加。与会者设计了七个病人角色，消费者代表确认了角色的可接受性和多样性。跨专业团队成员和消费者绘制了每个角色在假设围手术期过程中的关键临床时刻、感受和障碍。确定了六个关键主题：预防性护理"、"个性化护理"、"综合沟通"、"共同决策"、"护理过渡 "和 "伙伴关系"。提出了 20 个潜在解决方案，其中最优先的是数字仪表板和虚拟护理协调：讨论：我们的研究结果强调了跨专业合作、患者和家庭参与以及支持医疗技术基础设施的重要性。通过以用户为基础的代码设计，参与者发现了医疗技术可以提高系统效率、改善等待手术的患者护理质量的潜在机会。代码设计方法让用户参与制定以当地情况为导向的政策，以应对当前围手术期服务面临的挑战，如等待时间过长和护理分散等：结论：医疗技术创新为改善围手术期护理和整合临床信息提供了机遇。未来的研究将为进一步实施和评估优先解决方案提供原型。

{"title":"Codesign of health technology interventions to support best-practice perioperative care and surgical waitlist management.","authors":"Sarah Joy Aitken, Sophie James, Amy Lawrence, Anthony Glover, Henry Pleass, Janani Thillianadesan, Sue Monaro, Kerry Hitos, Vasi Naganathan","doi":"10.1136/bmjhci-2023-100928","DOIUrl":"10.1136/bmjhci-2023-100928","url":null,"abstract":"Objectives: This project aimed to determine where health technology can support best-practice perioperative care for patients waiting for surgery.Methods: An exploratory codesign process used personas and journey mapping in three interprofessional workshops to identify key challenges in perioperative care across four health districts in Sydney, Australia. Through participatory methodology, the research inquiry directly involved perioperative clinicians. In three facilitated workshops, clinician and patient participants codesigned potential digital interventions to support perioperative pathways. Workshop output was coded and thematically analysed, using design principles.Results: Codesign workshops, involving 51 participants, were conducted October to November 2022. Participants designed seven patient personas, with consumer representatives confirming acceptability and diversity. Interprofessional team members and consumers mapped key clinical moments, feelings and barriers for each persona during a hypothetical perioperative journey. Six key themes were identified: 'preventative care', 'personalised care', 'integrated communication', 'shared decision-making', 'care transitions' and 'partnership'. Twenty potential solutions were proposed, with top priorities a digital dashboard and virtual care coordination.Discussion: Our findings emphasise the importance of interprofessional collaboration, patient and family engagement and supporting health technology infrastructure. Through user-based codesign, participants identified potential opportunities where health technology could improve system efficiencies and enhance care quality for patients waiting for surgical procedures. The codesign approach embedded users in the development of locally-driven, contextually oriented policies to address current perioperative service challenges, such as prolonged waiting times and care fragmentation.Conclusion: Health technology innovation provides opportunities to improve perioperative care and integrate clinical information. Future research will prototype priority solutions for further implementation and evaluation.","PeriodicalId":9050,"journal":{"name":"BMJ Health & Care Informatics","volume":"31 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10936498/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140109053","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}

引用次数: 0

Invitation to join the Healthcare AI Language Group: HeALgroup.AI Initiative. 邀请加入医疗保健人工智能语言组：HeALgroup.AI 计划。

IF 4.1 Q2 Computer Science

BMJ Health & Care Informatics

Pub Date : 2024-03-12 DOI: 10.1136/bmjhci-2023-100884

Sebastian Manuel Staubli, Basel Jobeir, Michael Spiro, Dimitri Aristotle Raptis

引用次数: 0

Bibliometric analysis of the 3-year trends (2018-2021) in literature on artificial intelligence in ophthalmology and vision sciences. 眼科学和视觉科学领域人工智能文献的 3 年趋势（2018-2021 年）文献计量分析。

IF 4.1 Q1 HEALTH CARE SCIENCES & SERVICES

BMJ Health & Care Informatics

Pub Date : 2024-02-28 DOI: 10.1136/bmjhci-2023-100780

Hayley Monson, Jeffrey Demaine, Adrianna Perryman, Tina Felfeli

Objectives: The objective of this analysis is to present a current view of the field of ophthalmology and vision research and artificial intelligence (AI) from topical and geographical perspectives. This will clarify the direction of the field in the future and aid clinicians in adapting to new technological developments.

Methods: A comprehensive search of four different databases was conducted. Statistical and bibliometric analysis were done to characterise the literature. Softwares used included the R Studio bibliometrix package, and VOSviewer.

Results: A total of 3939 articles were included in the final bibliometric analysis. Diabetic retinopathy (391, 6% of the top 100 keywords) was the most frequently occurring indexed keyword by a large margin. The highest impact literature was produced by the least populated countries and in those countries who collaborate internationally. This was confirmed via a hypothesis test where no correlation was found between gross number of published articles and average number of citations (p value=0.866, r=0.038), while graphing ratio of international collaboration against average citations produced a positive correlation (r=0.283). Majority of publications were found to be concentrated in journals specialising in vision and computer science, with this category of journals having the highest number of publications per journal (18.00 publications/journal), though they represented a small proportion of the total journals (<1%).

Conclusion: This study provides a unique characterisation of the literature at the intersection of AI and ophthalmology and presents correlations between article impact and geography, in addition to summarising popular research topics.

目标：本分析报告旨在从专题和地理角度介绍眼科和视觉研究以及人工智能（AI）领域的现状。这将明确该领域未来的发展方向，并帮助临床医生适应新的技术发展：方法：对四个不同的数据库进行了全面检索。方法：对四个不同的数据库进行了全面搜索，并进行了统计和文献计量分析，以确定文献的特点。使用的软件包括 R Studio bibliometrix 软件包和 VOSviewer：最终的文献计量分析共纳入了 3939 篇文章。糖尿病视网膜病变（391篇，占前100个关键词的6%）是出现频率最高的索引关键词。人口最少的国家和开展国际合作的国家发表的文献影响最大。通过假设检验证实了这一点，即发表文章总数与平均引用次数之间没有相关性（P 值=0.866，r=0.038），而国际合作比率与平均引用次数之间的曲线图则产生了正相关性（r=0.283）。大部分论文集中在视觉和计算机科学专业期刊上，这类期刊的论文数量最多（18.00 篇/期刊），但在期刊总数中所占比例较小（结论：这类期刊的论文数量最多，但在期刊总数中所占比例较小）：本研究对人工智能与眼科学交叉领域的文献进行了独特的描述，除了总结热门研究课题外，还介绍了文章影响力与地域之间的相关性。

{"title":"Bibliometric analysis of the 3-year trends (2018-2021) in literature on artificial intelligence in ophthalmology and vision sciences.","authors":"Hayley Monson, Jeffrey Demaine, Adrianna Perryman, Tina Felfeli","doi":"10.1136/bmjhci-2023-100780","DOIUrl":"10.1136/bmjhci-2023-100780","url":null,"abstract":"Objectives: The objective of this analysis is to present a current view of the field of ophthalmology and vision research and artificial intelligence (AI) from topical and geographical perspectives. This will clarify the direction of the field in the future and aid clinicians in adapting to new technological developments.Methods: A comprehensive search of four different databases was conducted. Statistical and bibliometric analysis were done to characterise the literature. Softwares used included the R Studio bibliometrix package, and VOSviewer.Results: A total of 3939 articles were included in the final bibliometric analysis. Diabetic retinopathy (391, 6% of the top 100 keywords) was the most frequently occurring indexed keyword by a large margin. The highest impact literature was produced by the least populated countries and in those countries who collaborate internationally. This was confirmed via a hypothesis test where no correlation was found between gross number of published articles and average number of citations (p value=0.866, r=0.038), while graphing ratio of international collaboration against average citations produced a positive correlation (r=0.283). Majority of publications were found to be concentrated in journals specialising in vision and computer science, with this category of journals having the highest number of publications per journal (18.00 publications/journal), though they represented a small proportion of the total journals (<1%).Conclusion: This study provides a unique characterisation of the literature at the intersection of AI and ophthalmology and presents correlations between article impact and geography, in addition to summarising popular research topics.","PeriodicalId":9050,"journal":{"name":"BMJ Health & Care Informatics","volume":"31 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10910687/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139989284","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}

引用次数: 0

Explainable machine learning for breast cancer diagnosis from mammography and ultrasound images: a systematic review 从乳房 X 射线照相术和超声波图像诊断乳腺癌的可解释机器学习：系统综述

IF 4.1 Q2 Computer Science

BMJ Health & Care Informatics

Pub Date : 2024-02-01 DOI: 10.1136/bmjhci-2023-100954

Daraje kaba Gurmessa, Worku Jimma

Background Breast cancer is the most common disease in women. Recently, explainable artificial intelligence (XAI) approaches have been dedicated to investigate breast cancer. An overwhelming study has been done on XAI for breast cancer. Therefore, this study aims to review an XAI for breast cancer diagnosis from mammography and ultrasound (US) images. We investigated how XAI methods for breast cancer diagnosis have been evaluated, the existing ethical challenges, research gaps, the XAI used and the relation between the accuracy and explainability of algorithms. Methods In this work, Preferred Reporting Items for Systematic Reviews and Meta-Analyses checklist and diagram were used. Peer-reviewed articles and conference proceedings from PubMed, IEEE Explore, ScienceDirect, Scopus and Google Scholar databases were searched. There is no stated date limit to filter the papers. The papers were searched on 19 September 2023, using various combinations of the search terms ‘breast cancer’, ‘explainable’, ‘interpretable’, ‘machine learning’, ‘artificial intelligence’ and ‘XAI’. Rayyan online platform detected duplicates, inclusion and exclusion of papers. Results This study identified 14 primary studies employing XAI for breast cancer diagnosis from mammography and US images. Out of the selected 14 studies, only 1 research evaluated humans’ confidence in using the XAI system—additionally, 92.86% of identified papers identified dataset and dataset-related issues as research gaps and future direction. The result showed that further research and evaluation are needed to determine the most effective XAI method for breast cancer. Conclusion XAI is not conceded to increase users’ and doctors’ trust in the system. For the real-world application, effective and systematic evaluation of its trustworthiness in this scenario is lacking. PROSPERO registration number CRD42023458665. Data are available upon reasonable request.

背景乳腺癌是女性最常见的疾病。最近，可解释人工智能（XAI）方法被用于研究乳腺癌。目前，针对乳腺癌的 XAI 研究还很少。因此，本研究旨在回顾一种用于从乳房 X 射线照相术和超声波（US）图像诊断乳腺癌的 XAI。我们调查了用于乳腺癌诊断的 XAI 方法是如何被评估的、现有的伦理挑战、研究差距、所使用的 XAI 以及算法的准确性和可解释性之间的关系。方法在这项工作中，使用了《系统综述和元分析首选报告项目》清单和图表。从 PubMed、IEEE Explore、ScienceDirect、Scopus 和 Google Scholar 数据库中搜索了同行评审文章和会议论文集。论文筛选没有明确的日期限制。论文搜索日期为 2023 年 9 月 19 日，使用了 "乳腺癌"、"可解释"、"可解释"、"机器学习"、"人工智能 "和 "XAI "等搜索词的不同组合。Rayyan 在线平台检测了重复、纳入和排除的论文。结果本研究共发现了 14 项利用 XAI 从乳房 X 射线照相术和 US 图像诊断乳腺癌的主要研究。在所选的 14 项研究中，只有 1 项研究对人类使用 XAI 系统的信心进行了评估，此外，92.86% 的已识别论文将数据集和数据集相关问题确定为研究差距和未来方向。结果表明，要确定最有效的乳腺癌 XAI 方法，还需要进一步的研究和评估。结论 XAI 并不能增加用户和医生对系统的信任。在实际应用中，还缺乏对其可信度的有效和系统评估。PROSPERO 注册号为 CRD42023458665。如有合理要求，可提供相关数据。

{"title":"Explainable machine learning for breast cancer diagnosis from mammography and ultrasound images: a systematic review","authors":"Daraje kaba Gurmessa, Worku Jimma","doi":"10.1136/bmjhci-2023-100954","DOIUrl":"https://doi.org/10.1136/bmjhci-2023-100954","url":null,"abstract":"Background Breast cancer is the most common disease in women. Recently, explainable artificial intelligence (XAI) approaches have been dedicated to investigate breast cancer. An overwhelming study has been done on XAI for breast cancer. Therefore, this study aims to review an XAI for breast cancer diagnosis from mammography and ultrasound (US) images. We investigated how XAI methods for breast cancer diagnosis have been evaluated, the existing ethical challenges, research gaps, the XAI used and the relation between the accuracy and explainability of algorithms. Methods In this work, Preferred Reporting Items for Systematic Reviews and Meta-Analyses checklist and diagram were used. Peer-reviewed articles and conference proceedings from PubMed, IEEE Explore, ScienceDirect, Scopus and Google Scholar databases were searched. There is no stated date limit to filter the papers. The papers were searched on 19 September 2023, using various combinations of the search terms ‘breast cancer’, ‘explainable’, ‘interpretable’, ‘machine learning’, ‘artificial intelligence’ and ‘XAI’. Rayyan online platform detected duplicates, inclusion and exclusion of papers. Results This study identified 14 primary studies employing XAI for breast cancer diagnosis from mammography and US images. Out of the selected 14 studies, only 1 research evaluated humans’ confidence in using the XAI system—additionally, 92.86% of identified papers identified dataset and dataset-related issues as research gaps and future direction. The result showed that further research and evaluation are needed to determine the most effective XAI method for breast cancer. Conclusion XAI is not conceded to increase users’ and doctors’ trust in the system. For the real-world application, effective and systematic evaluation of its trustworthiness in this scenario is lacking. PROSPERO registration number CRD42023458665. Data are available upon reasonable request.","PeriodicalId":9050,"journal":{"name":"BMJ Health & Care Informatics","volume":"25 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139669953","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}

引用次数: 0

Seamless EMR data access: Integrated governance, digital health and the OMOP-CDM 无缝的 EMR 数据访问：综合治理、数字医疗和 OMOP-CDM

IF 4.1 Q2 Computer Science

BMJ Health & Care Informatics

Pub Date : 2024-02-01 DOI: 10.1136/bmjhci-2023-100953

Christine Mary Hallinan, Roger Ward, Graeme K Hart, Clair Sullivan, Nicole Pratt, Ashley P Ng, Daniel Capurro, Anton Van Der Vegt, Siaw-Teng Liaw, Oliver Daly, Blanca Gallego Luxan, David Bunker, Douglas Boyle

Objectives In this overview, we describe theObservational Medical Outcomes Partnership Common Data Model (OMOP-CDM), the established governance processes employed in EMR data repositories, and demonstrate how OMOP transformed data provides a lever for more efficient and secure access to electronic medical record (EMR) data by health service providers and researchers. Methods Through pseudonymisation and common data quality assessments, the OMOP-CDM provides a robust framework for converting complex EMR data into a standardised format. This allows for the creation of shared end-to-end analysis packages without the need for direct data exchange, thereby enhancing data security and privacy. By securely sharing de-identified and aggregated data and conducting analyses across multiple OMOP-converted databases, patient-level data is securely firewalled within its respective local site. Results By simplifying data management processes and governance, and through the promotion of interoperability, the OMOP-CDM supports a wide range of clinical, epidemiological, and translational research projects, as well as health service operational reporting. Discussion Adoption of the OMOP-CDM internationally and locally enables conversion of vast amounts of complex, and heterogeneous EMR data into a standardised structured data model, simplifies governance processes, and facilitates rapid repeatable cross-institution analysis through shared end-to-end analysis packages, without the sharing of data. Conclusion The adoption of the OMOP-CDM has the potential to transform health data analytics by providing a common platform for analysing EMR data across diverse healthcare settings. Data sharing not applicable as no datasets generated.

目的在本综述中，我们将介绍观察性医疗结果合作组织通用数据模型（OMOP-CDM）、EMR 数据存储库所采用的既定管理流程，并展示 OMOP 转换后的数据如何为医疗服务提供者和研究人员更高效、更安全地访问电子病历（EMR）数据提供杠杆作用。方法通过化名和通用数据质量评估，OMOP-CDM 为将复杂的 EMR 数据转换为标准化格式提供了一个强大的框架。这样就可以创建共享的端到端分析包，而无需直接交换数据，从而提高了数据的安全性和隐私性。通过安全共享去标识化和汇总数据，并在多个 OMOP 转换数据库中进行分析，患者级数据在各自的本地站点内被安全防火墙隔离。结果通过简化数据管理流程和治理，并通过促进互操作性，OMOP-CDM 支持了广泛的临床、流行病学和转化研究项目，以及医疗服务运营报告。讨论在国际和本地采用 OMOP-CDM 能够将大量复杂、异构的 EMR 数据转换为标准化的结构化数据模型，简化管理流程，并通过共享端到端分析包，在不共享数据的情况下，促进快速、可重复的跨机构分析。结论采用 OMOP-CDM 有可能改变健康数据分析，为分析不同医疗机构的 EMR 数据提供一个通用平台。由于未生成数据集，数据共享不适用。

{"title":"Seamless EMR data access: Integrated governance, digital health and the OMOP-CDM","authors":"Christine Mary Hallinan, Roger Ward, Graeme K Hart, Clair Sullivan, Nicole Pratt, Ashley P Ng, Daniel Capurro, Anton Van Der Vegt, Siaw-Teng Liaw, Oliver Daly, Blanca Gallego Luxan, David Bunker, Douglas Boyle","doi":"10.1136/bmjhci-2023-100953","DOIUrl":"https://doi.org/10.1136/bmjhci-2023-100953","url":null,"abstract":"Objectives In this overview, we describe theObservational Medical Outcomes Partnership Common Data Model (OMOP-CDM), the established governance processes employed in EMR data repositories, and demonstrate how OMOP transformed data provides a lever for more efficient and secure access to electronic medical record (EMR) data by health service providers and researchers. Methods Through pseudonymisation and common data quality assessments, the OMOP-CDM provides a robust framework for converting complex EMR data into a standardised format. This allows for the creation of shared end-to-end analysis packages without the need for direct data exchange, thereby enhancing data security and privacy. By securely sharing de-identified and aggregated data and conducting analyses across multiple OMOP-converted databases, patient-level data is securely firewalled within its respective local site. Results By simplifying data management processes and governance, and through the promotion of interoperability, the OMOP-CDM supports a wide range of clinical, epidemiological, and translational research projects, as well as health service operational reporting. Discussion Adoption of the OMOP-CDM internationally and locally enables conversion of vast amounts of complex, and heterogeneous EMR data into a standardised structured data model, simplifies governance processes, and facilitates rapid repeatable cross-institution analysis through shared end-to-end analysis packages, without the sharing of data. Conclusion The adoption of the OMOP-CDM has the potential to transform health data analytics by providing a common platform for analysing EMR data across diverse healthcare settings. Data sharing not applicable as no datasets generated.","PeriodicalId":9050,"journal":{"name":"BMJ Health & Care Informatics","volume":"65 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139922526","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}

引用次数: 0

Performance of large language models on advocating the management of meningitis: a comparative qualitative stud 大语言模型在脑膜炎管理宣传方面的表现：一项定性比较研究

IF 4.1 Q2 Computer Science

BMJ Health & Care Informatics

Pub Date : 2024-02-01 DOI: 10.1136/bmjhci-2023-100978

Urs Fisch, Paulina Kliem, Pascale Grzonka, Raoul Sutter

Objectives We aimed to examine the adherence of large language models (LLMs) to bacterial meningitis guidelines using a hypothetical medical case, highlighting their utility and limitations in healthcare. Methods A simulated clinical scenario of a patient with bacterial meningitis secondary to mastoiditis was presented in three independent sessions to seven publicly accessible LLMs (Bard, Bing, Claude-2, GTP-3.5, GTP-4, Llama, PaLM). Responses were evaluated for adherence to good clinical practice and two international meningitis guidelines. Results A central nervous system infection was identified in 90% of LLM sessions. All recommended imaging, while 81% suggested lumbar puncture. Blood cultures and specific mastoiditis work-up were proposed in only 62% and 38% sessions, respectively. Only 38% of sessions provided the correct empirical antibiotic treatment, while antiviral treatment and dexamethasone were advised in 33% and 24%, respectively. Misleading statements were generated in 52%. No significant correlation was found between LLMs’ text length and performance (r=0.29, p=0.20). Among all LLMs, GTP-4 demonstrated the best performance. Discussion Latest LLMs provide valuable advice on differential diagnosis and diagnostic procedures but significantly vary in treatment-specific information for bacterial meningitis when introduced to a realistic clinical scenario. Misleading statements were common, with performance differences attributed to each LLM’s unique algorithm rather than output length. Conclusions Users must be aware of such limitations and performance variability when considering LLMs as a support tool for medical decision-making. Further research is needed to refine these models' comprehension of complex medical scenarios and their ability to provide reliable information. Data are available upon reasonable request.

目的我们旨在通过一个假设的医疗案例来检验大型语言模型（LLMs）对细菌性脑膜炎指南的遵从情况，从而突出其在医疗保健领域的实用性和局限性。方法将一个继发于乳突炎的细菌性脑膜炎患者的模拟临床情景分三次展示给七个可公开访问的大型语言模型（Bard、Bing、Claude-2、GTP-3.5、GTP-4、Llama、PaLM）。根据良好临床实践和两份国际脑膜炎指南，对回复进行了评估。结果 90% 的 LLM 会议确定了中枢神经系统感染。所有人都建议进行影像学检查，81%的人建议进行腰椎穿刺。分别只有 62% 和 38% 的会议建议进行血液培养和特定乳突炎检查。只有 38% 的会议提供了正确的经验性抗生素治疗，而分别有 33% 和 24% 的会议建议进行抗病毒治疗和地塞米松治疗。有 52% 的陈述具有误导性。结果表明，语言学习者的文字长度与学习成绩之间没有明显的相关性（r=0.29，p=0.20）。在所有 LLM 中，GTP-4 的性能最佳。讨论最新的 LLM 在鉴别诊断和诊断程序方面提供了有价值的建议，但在引入真实的临床场景时，在细菌性脑膜炎的治疗特异性信息方面存在显著差异。误导性陈述很常见，性能差异归因于每个 LLM 的独特算法而非输出长度。结论用户在考虑将 LLM 作为医疗决策支持工具时，必须意识到这些局限性和性能差异。还需要进一步的研究来完善这些模型对复杂医疗场景的理解能力以及提供可靠信息的能力。如有合理要求，可提供相关数据。

{"title":"Performance of large language models on advocating the management of meningitis: a comparative qualitative stud","authors":"Urs Fisch, Paulina Kliem, Pascale Grzonka, Raoul Sutter","doi":"10.1136/bmjhci-2023-100978","DOIUrl":"https://doi.org/10.1136/bmjhci-2023-100978","url":null,"abstract":"Objectives We aimed to examine the adherence of large language models (LLMs) to bacterial meningitis guidelines using a hypothetical medical case, highlighting their utility and limitations in healthcare. Methods A simulated clinical scenario of a patient with bacterial meningitis secondary to mastoiditis was presented in three independent sessions to seven publicly accessible LLMs (Bard, Bing, Claude-2, GTP-3.5, GTP-4, Llama, PaLM). Responses were evaluated for adherence to good clinical practice and two international meningitis guidelines. Results A central nervous system infection was identified in 90% of LLM sessions. All recommended imaging, while 81% suggested lumbar puncture. Blood cultures and specific mastoiditis work-up were proposed in only 62% and 38% sessions, respectively. Only 38% of sessions provided the correct empirical antibiotic treatment, while antiviral treatment and dexamethasone were advised in 33% and 24%, respectively. Misleading statements were generated in 52%. No significant correlation was found between LLMs’ text length and performance (r=0.29, p=0.20). Among all LLMs, GTP-4 demonstrated the best performance. Discussion Latest LLMs provide valuable advice on differential diagnosis and diagnostic procedures but significantly vary in treatment-specific information for bacterial meningitis when introduced to a realistic clinical scenario. Misleading statements were common, with performance differences attributed to each LLM’s unique algorithm rather than output length. Conclusions Users must be aware of such limitations and performance variability when considering LLMs as a support tool for medical decision-making. Further research is needed to refine these models' comprehension of complex medical scenarios and their ability to provide reliable information. Data are available upon reasonable request.","PeriodicalId":9050,"journal":{"name":"BMJ Health & Care Informatics","volume":"6 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139667162","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}

引用次数: 0

Rapidly scalable and low-cost public health surveillance reporting system for COVID-19. 针对 COVID-19 的可快速扩展的低成本公共卫生监测报告系统。

IF 4.1 Q2 Computer Science

BMJ Health & Care Informatics

Pub Date : 2024-01-18 DOI: 10.1136/bmjhci-2023-100759

Vivek Jason Jayaraj, Chiu-Wan Ng, Victor Chee-Wai Hoe, Diane Woei-Quan Chong, Sanjay Rampal

Objective: Data-driven innovations are essential in strengthening disease control. We developed a low-cost, open-source system for robust epidemiological intelligence in response to the COVID-19 crisis, prioritising scalability, reproducibility and dynamic reporting.

Methods: A five-tiered workflow of data acquisition; processing; databasing, sharing, version control; visualisation; and monitoring was used. COVID-19 data were initially collated from press releases and then transitioned to official sources.

Results: Key COVID-19 indicators were tabulated and visualised, deployed using open-source hosting in October 2022. The system demonstrated high performance, handling extensive data volumes, with a 92.5% user conversion rate, evidencing its value and adaptability.

Conclusion: This cost-effective, scalable solution aids health specialists and authorities in tracking disease burden, particularly in low-resource settings. Such innovations are critical in health crises like COVID-19 and adaptable to diverse health scenarios.

目标：数据驱动的创新对加强疾病控制至关重要。为应对 COVID-19 危机，我们开发了一个低成本、开源的流行病学情报系统，优先考虑可扩展性、可重复性和动态报告：方法：采用五层工作流程：数据采集、处理、数据库、共享、版本控制、可视化和监测。COVID-19 数据最初从新闻稿中收集，然后过渡到官方来源：COVID-19 的关键指标已制成表格并实现可视化，于 2022 年 10 月使用开源主机进行部署。该系统性能卓越，可处理大量数据，用户转换率达 92.5%，证明了其价值和适应性：这一成本效益高、可扩展的解决方案有助于卫生专家和当局跟踪疾病负担，尤其是在资源匮乏的环境中。这种创新在 COVID-19 等健康危机中至关重要，并可适应各种健康情景。

{"title":"Rapidly scalable and low-cost public health surveillance reporting system for COVID-19.","authors":"Vivek Jason Jayaraj, Chiu-Wan Ng, Victor Chee-Wai Hoe, Diane Woei-Quan Chong, Sanjay Rampal","doi":"10.1136/bmjhci-2023-100759","DOIUrl":"10.1136/bmjhci-2023-100759","url":null,"abstract":"Objective: Data-driven innovations are essential in strengthening disease control. We developed a low-cost, open-source system for robust epidemiological intelligence in response to the COVID-19 crisis, prioritising scalability, reproducibility and dynamic reporting.Methods: A five-tiered workflow of data acquisition; processing; databasing, sharing, version control; visualisation; and monitoring was used. COVID-19 data were initially collated from press releases and then transitioned to official sources.Results: Key COVID-19 indicators were tabulated and visualised, deployed using open-source hosting in October 2022. The system demonstrated high performance, handling extensive data volumes, with a 92.5% user conversion rate, evidencing its value and adaptability.Conclusion: This cost-effective, scalable solution aids health specialists and authorities in tracking disease burden, particularly in low-resource settings. Such innovations are critical in health crises like COVID-19 and adaptable to diverse health scenarios.","PeriodicalId":9050,"journal":{"name":"BMJ Health & Care Informatics","volume":"31 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11077347/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139490782","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}

引用次数: 0

Regulating AI for health. 规范人工智能，促进健康。

IF 4.1 Q2 Computer Science

BMJ Health & Care Informatics

Pub Date : 2023-12-21 DOI: 10.1136/bmjhci-2023-100931

Ian Oppermann

引用次数: 0