International Journal of Medical Informatics最新文献_第10页

Predicting whether patients in an acute medical unit are physiologically fit-for-discharge using machine learning: A proof-of-concept 利用机器学习预测急诊科病人的生理状况是否适合出院：概念验证

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-08-10 DOI: 10.1016/j.ijmedinf.2024.105586

S.H. Garssen , C.A. Vernooij , N. Kant , M.V. Koning , F.H. Bosch , C.J.M. Doggen , B.P. Veldkamp , W.F.J. Verhaegh , S.F. Oude Wesselink

Introduction

Delays in discharging patients from Acute Medical Units hamper patient flows throughout the hospital. The decision to discharge a patient is mainly based on the patients’ physiological condition, but may vary between physicians. An objective decision-support system based on patients’ physiological data may help minimizing unnecessary delays in discharge. The aim of this proof-of-concept study is to assess the feasibility of predicting whether patients in an Acute Medical Unit are physiologically fit-for-discharge using machine learning with commonly available hospital data. Furthermore, this study investigated how long before actual time of discharge from the Acute Medical Unit we could predict discharge fitness. Also, the predictive importance of features extracted from these data was assessed.

Methods

Electronic Medical Records of patients who participated in a Randomized Controlled Trial conducted in an Acute Medical Unit were used retrospectively (N = 199). Only commonly available hospital data were used. Logistic Regression and Random Forest models were applied to predict every hour whether patients were physiologically fit-for-discharge. Nested 5-fold cross-validation with 5 repeats was used to optimize the model hyperparameters and to estimate the predictive performances.

Results

Physiological discharge fitness was predictable with reasonable performance for Logistic Regression (mean AUROC: 0.67) and Random Forest (mean AUROC: 0.69). For an intuitively chosen classification threshold of 0.8, mean specificity was 93.3 % and sensitivity 14.1 %. Models could predict physiological discharge fitness more than 24 h earlier than actual time of discharge for most patients who were correctly predicted to be fit-for-discharge. Patient characteristics, vital signs and laboratory results were shown to be important predictors.

Conclusion

This proof-of-concept study showed that it is feasible to predict with machine learning whether patients in an Acute Medical Unit are physiologically fit-for-discharge using commonly available hospital data.

导言：急诊科病人延迟出院阻碍了整个医院的病人流动。病人出院的决定主要基于病人的生理状况，但不同医生的决定可能有所不同。基于病人生理数据的客观决策支持系统可能有助于减少不必要的出院延误。这项概念验证研究旨在评估利用机器学习和医院常用数据预测急诊科病人的生理状况是否适合出院的可行性。此外，本研究还调查了在急诊科实际出院时间之前多久，我们可以预测出院患者的健康状况。方法回顾性使用了在急诊科参与随机对照试验的患者的电子病历（N = 199）。仅使用医院常用数据。应用逻辑回归和随机森林模型预测患者每小时的生理状况是否适合出院。结果Logistic回归（平均AUROC：0.67）和随机森林（平均AUROC：0.69）都能以合理的性能预测患者是否适合出院。直观选择的分类阈值为 0.8，平均特异性为 93.3%，灵敏度为 14.1%。对于大多数被正确预测为适合出院的患者来说，模型可以比实际出院时间提前 24 小时以上预测出他们的出院健康状况。患者特征、生命体征和实验室结果均被证明是重要的预测因素。结论这项概念验证研究表明，利用常用的医院数据，通过机器学习预测急诊科患者的生理状况是否适合出院是可行的。

{"title":"Predicting whether patients in an acute medical unit are physiologically fit-for-discharge using machine learning: A proof-of-concept","authors":"S.H. Garssen , C.A. Vernooij , N. Kant , M.V. Koning , F.H. Bosch , C.J.M. Doggen , B.P. Veldkamp , W.F.J. Verhaegh , S.F. Oude Wesselink","doi":"10.1016/j.ijmedinf.2024.105586","DOIUrl":"10.1016/j.ijmedinf.2024.105586","url":null,"abstract":"<div><h3>Introduction</h3><p>Delays in discharging patients from Acute Medical Units hamper patient flows throughout the hospital. The decision to discharge a patient is mainly based on the patients’ physiological condition, but may vary between physicians. An objective decision-support system based on patients’ physiological data may help minimizing unnecessary delays in discharge. The aim of this proof-of-concept study is to assess the feasibility of predicting whether patients in an Acute Medical Unit are physiologically fit-for-discharge using machine learning with commonly available hospital data. Furthermore, this study investigated how long before actual time of discharge from the Acute Medical Unit we could predict discharge fitness. Also, the predictive importance of features extracted from these data was assessed.</p></div><div><h3>Methods</h3><p>Electronic Medical Records of patients who participated in a Randomized Controlled Trial conducted in an Acute Medical Unit were used retrospectively (N = 199). Only commonly available hospital data were used. Logistic Regression and Random Forest models were applied to predict every hour whether patients were physiologically fit-for-discharge. Nested 5-fold cross-validation with 5 repeats was used to optimize the model hyperparameters and to estimate the predictive performances.</p></div><div><h3>Results</h3><p>Physiological discharge fitness was predictable with reasonable performance for Logistic Regression (mean AUROC: 0.67) and Random Forest (mean AUROC: 0.69). For an intuitively chosen classification threshold of 0.8, mean specificity was 93.3 % and sensitivity 14.1 %. Models could predict physiological discharge fitness more than 24 h earlier than actual time of discharge for most patients who were correctly predicted to be fit-for-discharge. Patient characteristics, vital signs and laboratory results were shown to be important predictors.</p></div><div><h3>Conclusion</h3><p>This proof-of-concept study showed that it is feasible to predict with machine learning whether patients in an Acute Medical Unit are physiologically fit-for-discharge using commonly available hospital data.</p></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"191 ","pages":"Article 105586"},"PeriodicalIF":3.7,"publicationDate":"2024-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1386505624002491/pdfft?md5=8103cab1867353c839079580e16b5a16&pid=1-s2.0-S1386505624002491-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142012972","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

What attributes of digital devices are important to clinicians in rehabilitation? A cross-cultural best-worst scaling study 数字设备的哪些特性对康复临床医生很重要？跨文化最佳-最差比例研究

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-08-09 DOI: 10.1016/j.ijmedinf.2024.105589

Louise Michelle Nettleton Pearce , Martin Howell , Tiê Parma Yamato , Jéssica Maria Ribeiro Bacha , José Eduardo Pompeu , Kirsten Howard , Catherine Sherrington , Leanne Hassett

Background

Digital interventions are becoming increasingly popular in rehabilitation. Understanding of device features which impact clinician adoption and satisfaction is limited. Research in the field should be conducted across diverse settings to ensure digital interventions do not exacerbate healthcare inequities.

Objective

This study aimed to understand rehabilitation clinicians’ preferences regarding device attributes and included a cross-cultural comparison.

Materials and Methods

Choice experiment methodology (best-worst scaling) was used to survey rehabilitation clinicians across Australia and Brazil. Participants completed 10 best-worst questions, choosing the most and least important device attributes from subsets of 31 attributes in a partially balanced block design. Results were analysed using multinomial models by country and latent class. Attribute preference scores (PS) were scaled to 0–100 (least to most important).

Results

A total of 122 clinicians from Brazil and 104 clinicians from Australia completed the survey. Most respondents were physiotherapists (83%) working with neurological populations (51%) in the private/self-employed sector (51%) who had experience using rehabilitation devices (87%). Despite preference heterogeneity across country and work sector (public/not-for-profit versus private/self-employed/other), clinicians consistently prioritised patient outcomes (PS 100.0, 95%CI: 86.2–100.0), patient engagement (PS 93.9, 95%CI: 80.6–94.2), usability (PS 81.3, 95%CI: 68.8–82.5), research evidence (PS 80.4, 95%CI: 68.1–81.7) and risk (PS 75.7, 95%CI: 63.8–77.3). In Australia, clinicians favoured device attributes which facilitate increased therapy dosage (PS 79.2, 95%CI: 62.6–81.1) and encourage patient independent practice (PS 66.8, 95%CI: 52.0–69.2). In Brazil, clinicians preferred attributes enabling device use for providing clinical data (PS 67.6, 95%CI: 51.8–70.9) and conducting clinical assessments (PS 65.6, 95%CI: 50.2–68.8).

Conclusion

Clinicians prioritise patients’ needs and practical application over technical aspects of digital rehabilitation devices. Contextual factors shape clinician preferences rather than individual clinician characteristics. Future device design and research should consider preferences and influences, involving diverse stakeholders to account for context-driven variations across cultures and healthcare settings.

数字干预在康复领域越来越受欢迎。人们对影响临床医生采用率和满意度的设备功能了解有限。该领域的研究应在不同的环境中进行，以确保数字化干预不会加剧医疗保健的不平等。本研究旨在了解康复临床医生对设备属性的偏好，并进行跨文化比较。研究采用选择实验方法（最佳-最差比例）对澳大利亚和巴西的康复临床医生进行了调查。参与者完成了 10 个 "最佳-最差 "问题，并在部分平衡块设计中从 31 个属性子集中选择了最重要和最不重要的设备属性。结果采用多项式模型按国家和潜在类别进行分析。属性偏好分数 (PS) 为 0-100（从最不重要到最重要）。共有 122 名巴西临床医生和 104 名澳大利亚临床医生完成了调查。大多数受访者是物理治疗师（83%），他们在私人/自营部门（51%）从事神经系统人群的工作，拥有使用康复设备的经验（87%）。尽管不同国家和不同工作部门（公共/非营利与私营/自营/其他）的偏好存在差异，但临床医生始终优先考虑（PS 100.0，95%CI：86.2-100.0）（PS 93.9，95%CI：80.6-94.2）（PS 81.3，95%CI：68.8-82.5）、（PS 80.4，95%CI：68.1-81.7）和（PS 75.7，95%CI：63.8-77.3）在澳大利亚，临床医生更青睐有利于增加（PS 79.2，95%CI：62.6-81.1）和鼓励患者（PS 66.8，95%CI：52.0-69.2）的设备属性。在巴西，临床医生更倾向于使用设备提供（PS 67.6，95%CI：51.8-70.9）和进行（PS 65.6，95%CI：50.2-68.8）。临床医生优先考虑的是患者的需求和实际应用，而不是数字康复设备的技术方面。影响临床医生偏好的是环境因素，而非临床医生的个人特征。未来的设备设计和研究应考虑偏好和影响因素，让不同的利益相关者参与其中，以考虑不同文化和医疗环境下的环境驱动差异。

{"title":"What attributes of digital devices are important to clinicians in rehabilitation? A cross-cultural best-worst scaling study","authors":"Louise Michelle Nettleton Pearce , Martin Howell , Tiê Parma Yamato , Jéssica Maria Ribeiro Bacha , José Eduardo Pompeu , Kirsten Howard , Catherine Sherrington , Leanne Hassett","doi":"10.1016/j.ijmedinf.2024.105589","DOIUrl":"10.1016/j.ijmedinf.2024.105589","url":null,"abstract":"<div><h3>Background</h3><p>Digital interventions are becoming increasingly popular in rehabilitation. Understanding of device features which impact clinician adoption and satisfaction is limited. Research in the field should be conducted across diverse settings to ensure digital interventions do not exacerbate healthcare inequities.</p></div><div><h3>Objective</h3><p>This study aimed to understand rehabilitation clinicians’ preferences regarding device attributes and included a cross-cultural comparison.</p></div><div><h3>Materials and Methods</h3><p>Choice experiment methodology (best-worst scaling) was used to survey rehabilitation clinicians across Australia and Brazil. Participants completed 10 best-worst questions, choosing the most and least important device attributes from subsets of 31 attributes in a partially balanced block design. Results were analysed using multinomial models by country and latent class. Attribute preference scores (PS) were scaled to 0–100 (least to most important).</p></div><div><h3>Results</h3><p>A total of 122 clinicians from Brazil and 104 clinicians from Australia completed the survey. Most respondents were physiotherapists (83%) working with neurological populations (51%) in the private/self-employed sector (51%) who had experience using rehabilitation devices (87%). Despite preference heterogeneity across country and work sector (public/not-for-profit versus private/self-employed/other), clinicians consistently prioritised <em>patient outcomes</em> (PS 100.0, 95%CI: 86.2–100.0)<em>, patient engagement</em> (PS 93.9, 95%CI: 80.6–94.2)<em>, usability</em> (PS 81.3, 95%CI: 68.8–82.5), <em>research evidence</em> (PS 80.4, 95%CI: 68.1–81.7) and <em>risk</em> (PS 75.7, 95%CI: 63.8–77.3)<em>.</em> In Australia, clinicians favoured device attributes which facilitate increased <em>therapy dosage</em> (PS 79.2, 95%CI: 62.6–81.1) and encourage patient <em>independent practice</em> (PS 66.8, 95%CI: 52.0–69.2). In Brazil, clinicians preferred attributes enabling device use for providing <em>clinical data</em> (PS 67.6, 95%CI: 51.8–70.9) and conducting <em>clinical assessments</em> (PS 65.6, 95%CI: 50.2–68.8).</p></div><div><h3>Conclusion</h3><p>Clinicians prioritise patients’ needs and practical application over technical aspects of digital rehabilitation devices. Contextual factors shape clinician preferences rather than individual clinician characteristics. Future device design and research should consider preferences and influences, involving diverse stakeholders to account for context-driven variations across cultures and healthcare settings.</p></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"191 ","pages":"Article 105589"},"PeriodicalIF":3.7,"publicationDate":"2024-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1386505624002521/pdfft?md5=c72a7884fc43eced2e4c1d6d7fbafb2f&pid=1-s2.0-S1386505624002521-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141935700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing communication and care coordination: A scoping review of encounter notification systems between emergency departments and primary care providers 加强沟通和护理协调：对急诊科和初级医疗服务提供者之间的会诊通知系统进行范围审查。

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-08-06 DOI: 10.1016/j.ijmedinf.2024.105579

Staria Joseph , Rebecca Tomaschek , Balthasar L. Hug , Patrick E. Beeler

Objective

This scoping review aims to explore the current state of encounter notification systems (ENS) between emergency departments (EDs) and primary care providers (PCPs), focusing on their mechanisms, effectiveness, impacts, and challenges in healthcare settings.

Methods

A systematic search was conducted using PubMed/MEDLINE and Google Scholar to identify relevant literature on ENS between EDs and PCPs. Eligible studies were selected based on predefined criteria, and data were synthesized narratively.

Results

The initial search yielded 1,396 articles, with 29 included in the review. Studies highlighted the significance of encounter notifications in improving communication and care coordination between EDs and PCPs, leading to enhanced patient outcomes. However, challenges such as technological barriers, privacy concerns, and variations in healthcare settings were identified.

Conclusion

ENS play a crucial role in enhancing communication and care coordination between EDs and PCPs. Despite challenges, these systems offer substantial benefits and opportunities for improving patient care in the ED-primary care continuum. Future research should focus on addressing implementation barriers and evaluating long-term impacts to optimize the effectiveness of ENS in this context.

目的：本范围综述旨在探讨急诊科（ED）与初级保健提供者（PCP）之间的会诊通知系统（ENS）的现状，重点关注其在医疗机构中的机制、有效性、影响和挑战：使用 PubMed/MEDLINE 和 Google Scholar 进行了系统检索，以确定急诊科与初级保健提供者之间的 ENS 相关文献。根据预先确定的标准筛选出符合条件的研究，并对数据进行叙述性综合：结果：初步搜索共获得 1,396 篇文章，其中 29 篇被纳入综述。研究强调了会诊通知在改善急诊室和初级保健医生之间的沟通和护理协调方面的重要性，从而提高了患者的治疗效果。然而，研究也发现了一些挑战，如技术障碍、隐私问题和医疗环境的差异：ENS 在加强急诊室与初级保健医生之间的沟通和护理协调方面发挥着至关重要的作用。尽管存在挑战，但这些系统仍为改善急诊室-初级保健连续性中的患者护理带来了巨大的益处和机遇。未来的研究应侧重于解决实施障碍和评估长期影响，以优化 ENS 在这方面的有效性。

{"title":"Enhancing communication and care coordination: A scoping review of encounter notification systems between emergency departments and primary care providers","authors":"Staria Joseph , Rebecca Tomaschek , Balthasar L. Hug , Patrick E. Beeler","doi":"10.1016/j.ijmedinf.2024.105579","DOIUrl":"10.1016/j.ijmedinf.2024.105579","url":null,"abstract":"<div><h3>Objective</h3><p>This scoping review aims to explore the current state of encounter notification systems (ENS) between emergency departments (EDs) and primary care providers (PCPs), focusing on their mechanisms, effectiveness, impacts, and challenges in healthcare settings.</p></div><div><h3>Methods</h3><p>A systematic search was conducted using PubMed/MEDLINE and Google Scholar to identify relevant literature on ENS between EDs and PCPs. Eligible studies were selected based on predefined criteria, and data were synthesized narratively.</p></div><div><h3>Results</h3><p>The initial search yielded 1,396 articles, with 29 included in the review. Studies highlighted the significance of encounter notifications in improving communication and care coordination between EDs and PCPs, leading to enhanced patient outcomes. However, challenges such as technological barriers, privacy concerns, and variations in healthcare settings were identified.</p></div><div><h3>Conclusion</h3><p>ENS play a crucial role in enhancing communication and care coordination between EDs and PCPs. Despite challenges, these systems offer substantial benefits and opportunities for improving patient care in the ED-primary care continuum. Future research should focus on addressing implementation barriers and evaluating long-term impacts to optimize the effectiveness of ENS in this context.</p></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"191 ","pages":"Article 105579"},"PeriodicalIF":3.7,"publicationDate":"2024-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1386505624002429/pdfft?md5=5a86a295cf35c6951c60814f0a9cbaca&pid=1-s2.0-S1386505624002429-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141914628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Artificial intelligence prediction of In-Hospital mortality in patients with dementia: A multi-center study 人工智能预测痴呆症患者的院内死亡率：一项多中心研究

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-08-06 DOI: 10.1016/j.ijmedinf.2024.105590

Ching-Chi Huang , Wan-Yin Kuo , Yu-Ting Shen , Chia-Jung Chen , Hung-Jung Lin , Chien-Chin Hsu , Chung-Feng Liu , Chien-Cheng Huang

Background

Prediction of mortality is very important for care planning in hospitalized patients with dementia and artificial intelligence has the potential to serve as a solution; however, this issue remains unclear. Thus, this study was conducted to elucidate this matter.

Methods

We identified 10,573 hospitalized patients aged ≥ 45 years with dementia from three hospitals between 2010 and 2020 for this study. Utilizing 44 feature variables extracted from electronic medical records, an artificial intelligence (AI) model was constructed to predict death during hospitalization. The data was randomly separated into 70 % training set and 30 % testing set. We compared predictive accuracy among six algorithms including logistic regression, random forest, extreme gradient boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), multilayer perceptron (MLP), and support vector machine (SVM). Additionally, another set of data collected in 2021 was used as the validation set to assess the performance of six algorithms.

Results

The average age was 79.8 years, with females constituting 54.5 % of the sample. The in-hospital mortality rate was 6.7 %. LightGBM exhibited the highest area under the curve (0.991) for predicting mortality compared to other algorithms (XGBoost: 0.987, random forest: 0.985, logistic regression: 0.918, MLP: 0.898, SVM: 0.897). The accuracy, sensitivity, positive predictive value, and negative predictive value of LightGBM were 0.943, 0.944, 0.943, 0.542, and 0.996, respectively. Among the features in LightGBM, the three most important variables were the Glasgow Coma Scale, respiratory rate, and blood urea nitrogen. In the validation set, the area under the curve of LightGBM reached 0.753.

Conclusions

The AI prediction model demonstrates strong accuracy in predicting in-hospital mortality among patients with dementia, suggesting its potential implementation to enhance future care quality.

背景预测死亡率对住院痴呆症患者的护理计划非常重要，人工智能有可能成为一种解决方案；然而，这一问题仍不清楚。因此，我们开展了这项研究来阐明这一问题。方法我们在 2010 年至 2020 年间从三家医院确定了 10,573 名年龄≥45 岁的住院痴呆症患者作为研究对象。利用从电子病历中提取的 44 个特征变量，构建了一个人工智能（AI）模型来预测住院期间的死亡。数据被随机分为 70% 的训练集和 30% 的测试集。我们比较了六种算法的预测准确性，包括逻辑回归、随机森林、极梯度提升（XGBoost）、轻梯度提升机（LightGBM）、多层感知器（MLP）和支持向量机（SVM）。此外，2021 年收集的另一组数据被用作验证集，以评估六种算法的性能。结果平均年龄为 79.8 岁，女性占样本的 54.5%。院内死亡率为 6.7%。与其他算法（XGBoost：0.987；随机森林：0.985；逻辑回归：0.991）相比，LightGBM 预测死亡率的曲线下面积（0.991）最高：0.985、逻辑回归：0.918、MLP：0.898、SVM：0.897）。LightGBM 的准确度、灵敏度、阳性预测值和阴性预测值分别为 0.943、0.944、0.943、0.542 和 0.996。在 LightGBM 的特征中，最重要的三个变量是格拉斯哥昏迷量表、呼吸频率和血尿素氮。在验证集中，LightGBM 的曲线下面积达到了 0.753。结论人工智能预测模型在预测痴呆症患者的院内死亡率方面表现出了很高的准确性，这表明该模型的应用有望提高未来的护理质量。

{"title":"Artificial intelligence prediction of In-Hospital mortality in patients with dementia: A multi-center study","authors":"Ching-Chi Huang , Wan-Yin Kuo , Yu-Ting Shen , Chia-Jung Chen , Hung-Jung Lin , Chien-Chin Hsu , Chung-Feng Liu , Chien-Cheng Huang","doi":"10.1016/j.ijmedinf.2024.105590","DOIUrl":"10.1016/j.ijmedinf.2024.105590","url":null,"abstract":"<div><h3>Background</h3><p>Prediction of mortality is very important for care planning in hospitalized patients with dementia and artificial intelligence has the potential to serve as a solution; however, this issue remains unclear. Thus, this study was conducted to elucidate this matter.</p></div><div><h3>Methods</h3><p>We identified 10,573 hospitalized patients aged ≥ 45 years with dementia from three hospitals between 2010 and 2020 for this study. Utilizing 44 feature variables extracted from electronic medical records, an artificial intelligence (AI) model was constructed to predict death during hospitalization. The data was randomly separated into 70 % training set and 30 % testing set. We compared predictive accuracy among six algorithms including logistic regression, random forest, extreme gradient boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), multilayer perceptron (MLP), and support vector machine (SVM). Additionally, another set of data collected in 2021 was used as the validation set to assess the performance of six algorithms.</p></div><div><h3>Results</h3><p>The average age was 79.8 years, with females constituting 54.5 % of the sample. The in-hospital mortality rate was 6.7 %. LightGBM exhibited the highest area under the curve (0.991) for predicting mortality compared to other algorithms (XGBoost: 0.987, random forest: 0.985, logistic regression: 0.918, MLP: 0.898, SVM: 0.897). The accuracy, sensitivity, positive predictive value, and negative predictive value of LightGBM were 0.943, 0.944, 0.943, 0.542, and 0.996, respectively. Among the features in LightGBM, the three most important variables were the Glasgow Coma Scale, respiratory rate, and blood urea nitrogen. In the validation set, the area under the curve of LightGBM reached 0.753.</p></div><div><h3>Conclusions</h3><p>The AI prediction model demonstrates strong accuracy in predicting in-hospital mortality among patients with dementia, suggesting its potential implementation to enhance future care quality.</p></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"191 ","pages":"Article 105590"},"PeriodicalIF":3.7,"publicationDate":"2024-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141978206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Utilizing online reviews for analyzing digital healthcare consultation services: Examining perspectives of both healthcare customers and healthcare professionals 利用在线评论分析数字医疗咨询服务：研究医疗保健客户和医疗保健专业人员的观点。

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-08-05 DOI: 10.1016/j.ijmedinf.2024.105587

Sreevatsa Bellary, Pradip Kumar Bala, Shibashish Chakraborty

Introduction

Digital healthcare consultation services, also known as telemedicine, have seen a surge in their usage, especially after the COVID-19 pandemic. The purpose of this study is to investigate the satisfaction determinants of healthcare customers (patients) and healthcare professionals (doctors), providing digital healthcare consultation services.

Methods

The analysis involved scraping online reviews of 11 telemedicine apps meant for patients and 7 telemedicine apps meant for doctors, yielding a total of 44,440 patient reviews and 4748 doctor reviews. A structural topic modeling analysis followed by regression, dominance, correspondence, and emotion analysis was conducted to derive insights.

Results

The study identified ten determinants of satisfaction from patients’ and eight from doctors’ perspectives. For patients, ’service variety and quality’ (β = 0.5527) was the top positive determinant, while ’payment disputes’ (β = -0.1173) and ’in-app membership’ (β = -0.031) negatively impacted satisfaction. For doctors, ’patient consultation management’ (β = 0.2009) was the leading positive determinant, with ’profile management’ (β = -0.1843), ’subscription’ (β = -0.183), and ’customer care support’ (β = -0.0908) being the negative ones. The most influential negative emotion for patients, anger, was closely associated with ’customer care service’ and ’in-app memberships,’ while joy was tied to ’service variety and quality’ and ’offers and discounts.’ For doctors, anger was associated with ’cost-effectiveness,’ and joy with ’app responsiveness.’

Conclusion

This study offers new insights by examining patient and doctor determinants at a granular level which can be used by telemedicine app developers and managers to build customer-centric services.

引言数字医疗保健咨询服务（也称为远程医疗）的使用率激增，尤其是在 COVID-19 大流行之后。本研究旨在调查医疗保健客户（患者）和医疗保健专业人员（医生）对提供数字医疗保健咨询服务的满意度的决定因素：分析涉及对 11 款面向患者的远程医疗应用程序和 7 款面向医生的远程医疗应用程序的在线评论进行搜索，共获得 44,440 条患者评论和 4748 条医生评论。在进行了结构主题建模分析后，又进行了回归分析、优势分析、对应分析和情感分析，以得出深入的见解：研究从患者和医生的角度分别确定了十项和八项满意度决定因素。对患者而言，"服务种类和质量"（β = 0.5527）是最大的积极决定因素，而 "付款纠纷"（β = -0.1173）和 "应用程序内会员"（β = -0.031）对满意度产生了负面影响。对医生而言，"患者咨询管理"（β = 0.2009）是最主要的积极决定因素，而 "档案管理"（β = -0.1843）、"订阅"（β = -0.183）和 "客户服务支持"（β = -0.0908）则是消极决定因素。对患者影响最大的负面情绪--愤怒与 "客户服务 "和 "应用程序内会员 "密切相关，而喜悦则与 "服务种类和质量 "和 "优惠和折扣 "相关。对于医生来说，愤怒与 "成本效益 "相关，而喜悦与 "应用程序响应速度 "相关：本研究通过对患者和医生决定因素的细化研究提供了新的见解，可供远程医疗应用程序开发者和管理者用于构建以客户为中心的服务。

{"title":"Utilizing online reviews for analyzing digital healthcare consultation services: Examining perspectives of both healthcare customers and healthcare professionals","authors":"Sreevatsa Bellary, Pradip Kumar Bala, Shibashish Chakraborty","doi":"10.1016/j.ijmedinf.2024.105587","DOIUrl":"10.1016/j.ijmedinf.2024.105587","url":null,"abstract":"<div><h3>Introduction</h3><p>Digital healthcare consultation services, also known as telemedicine, have seen a surge in their usage, especially after the COVID-19 pandemic. The purpose of this study is to investigate the satisfaction determinants of healthcare customers (patients) and healthcare professionals (doctors), providing digital healthcare consultation services.</p></div><div><h3>Methods</h3><p>The analysis involved scraping online reviews of 11 telemedicine apps meant for patients and 7 telemedicine apps meant for doctors, yielding a total of 44,440 patient reviews and 4748 doctor reviews. A structural topic modeling analysis followed by regression, dominance, correspondence, and emotion analysis was conducted to derive insights.</p></div><div><h3>Results</h3><p>The study identified ten determinants of satisfaction from patients’ and eight from doctors’ perspectives. For patients, ’service variety and quality’ (β = 0.5527) was the top positive determinant, while ’payment disputes’ (β = -0.1173) and ’in-app membership’ (β = -0.031) negatively impacted satisfaction. For doctors, ’patient consultation management’ (β = 0.2009) was the leading positive determinant, with ’profile management’ (β = -0.1843), ’subscription’ (β = -0.183), and ’customer care support’ (β = -0.0908) being the negative ones. The most influential negative emotion for patients, anger, was closely associated with ’customer care service’ and ’in-app memberships,’ while joy was tied to ’service variety and quality’ and ’offers and discounts.’ For doctors, anger was associated with ’cost-effectiveness,’ and joy with ’app responsiveness.’</p></div><div><h3>Conclusion</h3><p>This study offers new insights by examining patient and doctor determinants at a granular level which can be used by telemedicine app developers and managers to build customer-centric services.</p></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"191 ","pages":"Article 105587"},"PeriodicalIF":3.7,"publicationDate":"2024-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141908411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Development and validation of a Multi-Causal investigation and discovery framework for knowledge harmonization (MINDMerge): A case study with acute kidney injury risk factor discovery using electronic medical records 开发和验证用于知识协调的多因果调查和发现框架（MINDMerge）：利用电子病历发现急性肾损伤风险因素的案例研究。

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-08-05 DOI: 10.1016/j.ijmedinf.2024.105588

Mingyang Zhang , Xiangzhou Zhang , Mingyang Dai , Lijuan Wu , Kang Liu , Hongnian Wang , Weiqi Chen , Mei Liu , Yong Hu

Objective

Accurate diagnoses and personalized treatments in medicine rely on identifying causality. However, existing causal discovery algorithms often yield inconsistent results due to distinct learning mechanisms. To address this challenge, we introduce MINDMerge, a multi-causal investigation and discovery framework designed to synthesize causal graphs from various algorithms.

Methods

MINDMerge integrates five causal models to reconcile inconsistencies arising from different algorithms. Employing credibility weighting and a novel cycle-breaking mechanism in causal networks, we initially developed and tested MINDMerge using three synthetic networks. Subsequently, we validated its effectiveness in discovering risk factors and predicting acute kidney injury (AKI) using two electronic medical records (EMR) datasets, eICU Collaborative Research Database and MIMIC-III Database. Causal reasoning was employed to analyze the relationships between risk factors and AKI. The identified causal risk factors of AKI were used in building a prediction model, and the prediction model was evaluated using the area under the receiver operating characteristics curve (AUC) and recall.

Results

Synthetic data experiments demonstrated that our model outperformed significantly in capturing ground-truth network structure compared to other causal models. Application of MINDMerge on real-world data revealed direct connections of pulmonary disease, hypertension, diabetes, x-ray assessment, and BUN with AKI. With the identified variables, AKI risk can be inferred at the individual level based on established BNs and prior information. Compared against existing benchmark models, MINDMerge maintained a higher AUC for AKI prediction in both internal (AUC: 0.832) and external network validations (AUC: 0.861).

Conclusion

MINDMerge can identify causal risk factors of AKI, serving as a valuable diagnostic tool for clinical decision-making and facilitating effective intervention.

目的：医学中的精确诊断和个性化治疗依赖于因果关系的识别。然而，由于不同的学习机制，现有的因果发现算法往往产生不一致的结果。为了应对这一挑战，我们引入了 MINDMerge，这是一个多因果调查和发现框架，旨在综合各种算法的因果图：MINDMerge整合了五个因果模型，以调和不同算法产生的不一致性。通过在因果网络中采用可信度加权和新颖的循环打破机制，我们利用三个合成网络初步开发并测试了 MINDMerge。随后，我们利用两个电子病历（EMR）数据集，即 eICU 合作研究数据库和 MIMIC-III 数据库，验证了 MINDMerge 在发现风险因素和预测急性肾损伤（AKI）方面的有效性。利用因果推理分析了风险因素与 AKI 之间的关系。确定的 AKI 因果风险因素被用于建立预测模型，并使用接收者操作特征曲线下面积（AUC）和召回率对预测模型进行评估：合成数据实验表明，与其他因果模型相比，我们的模型在捕捉地面实况网络结构方面表现出色。在真实世界数据中应用 MINDMerge 发现了肺部疾病、高血压、糖尿病、X 光评估和血尿素氮与 AKI 的直接联系。有了确定的变量，就可以根据已建立的 BN 和先验信息推断出个体水平的 AKI 风险。与现有的基准模型相比，MINDMerge 在内部（AUC：0.832）和外部网络验证（AUC：0.861）中都保持了较高的 AKI 预测 AUC：结论：MINDMerge 可识别 AKI 的成因风险因素，是临床决策的重要诊断工具，有助于采取有效的干预措施。

{"title":"Development and validation of a Multi-Causal investigation and discovery framework for knowledge harmonization (MINDMerge): A case study with acute kidney injury risk factor discovery using electronic medical records","authors":"Mingyang Zhang , Xiangzhou Zhang , Mingyang Dai , Lijuan Wu , Kang Liu , Hongnian Wang , Weiqi Chen , Mei Liu , Yong Hu","doi":"10.1016/j.ijmedinf.2024.105588","DOIUrl":"10.1016/j.ijmedinf.2024.105588","url":null,"abstract":"<div><h3>Objective</h3><p>Accurate diagnoses and personalized treatments in medicine rely on identifying causality. However, existing causal discovery algorithms often yield inconsistent results due to distinct learning mechanisms. To address this challenge, we introduce MINDMerge, a multi-causal investigation and discovery framework designed to synthesize causal graphs from various algorithms.</p></div><div><h3>Methods</h3><p>MINDMerge integrates five causal models to reconcile inconsistencies arising from different algorithms. Employing credibility weighting and a novel cycle-breaking mechanism in causal networks, we initially developed and tested MINDMerge using three synthetic networks. Subsequently, we validated its effectiveness in discovering risk factors and predicting acute kidney injury (AKI) using two electronic medical records (EMR) datasets, eICU Collaborative Research Database and MIMIC-III Database. Causal reasoning was employed to analyze the relationships between risk factors and AKI. The identified causal risk factors of AKI were used in building a prediction model, and the prediction model was evaluated using the area under the receiver operating characteristics curve (AUC) and recall.</p></div><div><h3>Results</h3><p>Synthetic data experiments demonstrated that our model outperformed significantly in capturing ground-truth network structure compared to other causal models. Application of MINDMerge on real-world data revealed direct connections of pulmonary disease, hypertension, diabetes, x-ray assessment, and BUN with AKI. With the identified variables, AKI risk can be inferred at the individual level based on established BNs and prior information. Compared against existing benchmark models, MINDMerge maintained a higher AUC for AKI prediction in both internal (AUC: 0.832) and external network validations (AUC: 0.861).</p></div><div><h3>Conclusion</h3><p>MINDMerge can identify causal risk factors of AKI, serving as a valuable diagnostic tool for clinical decision-making and facilitating effective intervention.</p></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"191 ","pages":"Article 105588"},"PeriodicalIF":3.7,"publicationDate":"2024-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141918135","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A pre-trained language model for emergency department intervention prediction using routine physiological data and clinical narratives 利用常规生理数据和临床叙述预训练急诊科干预预测语言模型。

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-08-03 DOI: 10.1016/j.ijmedinf.2024.105564

Ting-Yun Huang , Chee-Fah Chong , Heng-Yu Lin , Tzu-Ying Chen , Yung-Chun Chang , Ming-Chin Lin

Introduction

The urgency and complexity of emergency room (ER) settings require precise and swift decision-making processes for patient care. Ensuring the timely execution of critical examinations and interventions is vital for reducing diagnostic errors, but the literature highlights a need for innovative approaches to optimize diagnostic accuracy and patient outcomes. In response, our study endeavors to create predictive models for timely examinations and interventions by leveraging the patient’s symptoms and vital signs recorded during triage, and in so doing, augment traditional diagnostic methodologies.

Methods

Focusing on four key areas—medication dispensing, vital interventions, laboratory testing, and emergency radiology exams, the study employed Natural Language Processing (NLP) and seven advanced machine learning techniques. The research was centered around the innovative use of BioClinicalBERT, a state-of-the-art NLP framework.

Results

BioClinicalBERT emerged as the superior model, outperforming others in predictive accuracy. The integration of physiological data with patient narrative symptoms demonstrated greater effectiveness compared to models based solely on textual data. The robustness of our approach was confirmed by an Area Under the Receiver Operating Characteristic curve (AUROC) score of 0.9.

Conclusion

The findings of our study underscore the feasibility of establishing a decision support system for emergency patients, targeting timely interventions and examinations based on a nuanced analysis of symptoms. By using an advanced natural language processing technique, our approach shows promise for enhancing diagnostic accuracy. However, the current model is not yet fully mature for direct implementation into daily clinical practice. Recognizing the imperative nature of precision in the ER environment, future research endeavors must focus on refining and expanding predictive models to include detailed timely examinations and interventions. Although the progress achieved in this study represents an encouraging step towards a more innovative and technology-driven paradigm in emergency care, full clinical integration warrants further exploration and validation.

导言：急诊室（ER）环境的紧迫性和复杂性要求患者护理决策过程精确迅速。确保及时进行关键检查和干预对减少诊断错误至关重要，但文献强调需要创新方法来优化诊断准确性和患者预后。为此，我们的研究致力于利用分诊过程中记录的患者症状和生命体征，创建及时检查和干预的预测模型，从而增强传统的诊断方法：方法：这项研究重点关注四个关键领域--配药、生命体征干预、实验室检测和急诊放射检查，采用了自然语言处理（NLP）和七种先进的机器学习技术。研究以 BioClinicalBERT（一种最先进的 NLP 框架）的创新使用为中心：结果：BioClinicalBERT 成为最优秀的模型，其预测准确性优于其他模型。与仅基于文本数据的模型相比，生理数据与患者叙述症状的整合显示出更大的有效性。接收者工作特征曲线下面积 (AUROC) 得分为 0.9，这证实了我们方法的稳健性：我们的研究结果强调了为急诊病人建立决策支持系统的可行性，该系统可根据对症状的细致分析，有针对性地进行及时干预和检查。通过使用先进的自然语言处理技术，我们的方法有望提高诊断准确性。然而，目前的模式尚未完全成熟，无法直接应用于日常临床实践。认识到急诊室环境中精确性的必要性，未来的研究工作必须侧重于完善和扩展预测模型，以包括详细的及时检查和干预措施。尽管本研究取得的进展代表着急诊护理向更具创新性和技术驱动型模式迈出了令人鼓舞的一步，但全面的临床整合还需要进一步的探索和验证。

{"title":"A pre-trained language model for emergency department intervention prediction using routine physiological data and clinical narratives","authors":"Ting-Yun Huang , Chee-Fah Chong , Heng-Yu Lin , Tzu-Ying Chen , Yung-Chun Chang , Ming-Chin Lin","doi":"10.1016/j.ijmedinf.2024.105564","DOIUrl":"10.1016/j.ijmedinf.2024.105564","url":null,"abstract":"<div><h3>Introduction</h3><p>The urgency and complexity of emergency room (ER) settings require precise and swift decision-making processes for patient care. Ensuring the timely execution of critical examinations and interventions is vital for reducing diagnostic errors, but the literature highlights a need for innovative approaches to optimize diagnostic accuracy and patient outcomes. In response, our study endeavors to create predictive models for timely examinations and interventions by leveraging the patient’s symptoms and vital signs recorded during triage, and in so doing, augment traditional diagnostic methodologies.</p></div><div><h3>Methods</h3><p>Focusing on four key areas—medication dispensing, vital interventions, laboratory testing, and emergency radiology exams, the study employed Natural Language Processing (NLP) and seven advanced machine learning techniques. The research was centered around the innovative use of BioClinicalBERT, a state-of-the-art NLP framework.</p></div><div><h3>Results</h3><p>BioClinicalBERT emerged as the superior model, outperforming others in predictive accuracy. The integration of physiological data with patient narrative symptoms demonstrated greater effectiveness compared to models based solely on textual data. The robustness of our approach was confirmed by an Area Under the Receiver Operating Characteristic curve (AUROC) score of 0.9.</p></div><div><h3>Conclusion</h3><p>The findings of our study underscore the feasibility of establishing a decision support system for emergency patients, targeting timely interventions and examinations based on a nuanced analysis of symptoms. By using an advanced natural language processing technique, our approach shows promise for enhancing diagnostic accuracy. However, the current model is not yet fully mature for direct implementation into daily clinical practice. Recognizing the imperative nature of precision in the ER environment, future research endeavors must focus on refining and expanding predictive models to include detailed timely examinations and interventions. Although the progress achieved in this study represents an encouraging step towards a more innovative and technology-driven paradigm in emergency care, full clinical integration warrants further exploration and validation.</p></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"191 ","pages":"Article 105564"},"PeriodicalIF":3.7,"publicationDate":"2024-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141914627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel voice classification based on Gower distance for Parkinson disease detection 基于高尔距离的新型语音分类法用于帕金森病检测

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-08-02 DOI: 10.1016/j.ijmedinf.2024.105583

Mustafa Noaman Kadhim , Dhiah Al-Shammary , Fahim Sufi

Background

Traditional classifier for the classification of diseases, such as K-Nearest Neighbors (KNN), Linear Discriminant Analysis (LDA), Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM), often struggle with high-dimensional medical datasets.

Objective

This study presents a novel classifier to overcome the limitations of traditional classifiers in Parkinson’s disease (PD) detection based on Gower distance.

Methods

We present the Gower distance metric to handle diverse feature sets in voice recordings, which acts as a dissimilarity measure for all feature types, making the model adept at identifying subtle patterns indicative of PD. Additionally, the Cuckoo Search algorithm is employed for feature selection, reducing dimensionality by focusing on key features, thereby lessening the computational load associated with high-dimensional datasets.

Results

The proposed classifier based on Gower distance resulted in an accuracy rate of 98.3% with feature selection and achieved an accuracy of 94.92% without the feature selection method. It outperforms traditional classifiers and recent studies in PD detection from voice recordings.

Conclusions

This accuracy shows the capability of the approach in the correct classification of instances and points out the potential of the approach as a reliable diagnostic tool for the medical practitioner. The findings state that the proposed approach holds promise for improving the diagnosis and monitoring of PD, both within medical institutions and at homes for the elderly.

背景：用于疾病分类的传统分类器，如K-近邻（KNN）、线性判别分析（LDA）、随机森林（RF）、逻辑回归（LR）和支持向量机（SVM）等，在处理高维医学数据集时往往力不从心：本研究提出了一种基于高尔距离的新型分类器，以克服传统分类器在帕金森病（PD）检测中的局限性：我们采用高尔距离度量来处理语音记录中的各种特征集，该度量可作为所有特征类型的差异度量，从而使模型善于识别表明帕金森病的微妙模式。此外，该模型还采用了布谷鸟搜索算法进行特征选择，通过聚焦关键特征来降低维度，从而减轻与高维数据集相关的计算负荷：结果：基于高尔距离的分类器在进行特征选择后，准确率达到 98.3%，而在不使用特征选择方法的情况下，准确率为 94.92%。在从语音记录中检测腹泻方面，它优于传统分类器和最近的研究：该准确率显示了该方法在正确分类实例方面的能力，并指出了该方法作为医疗从业人员可靠诊断工具的潜力。研究结果表明，所提出的方法有望改善医疗机构和养老院对老年痴呆症的诊断和监测。

{"title":"A novel voice classification based on Gower distance for Parkinson disease detection","authors":"Mustafa Noaman Kadhim , Dhiah Al-Shammary , Fahim Sufi","doi":"10.1016/j.ijmedinf.2024.105583","DOIUrl":"10.1016/j.ijmedinf.2024.105583","url":null,"abstract":"<div><h3>Background</h3><p>Traditional classifier for the classification of diseases, such as K-Nearest Neighbors (KNN), Linear Discriminant Analysis (LDA), Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM), often struggle with high-dimensional medical datasets.</p></div><div><h3>Objective</h3><p>This study presents a novel classifier to overcome the limitations of traditional classifiers in Parkinson’s disease (PD) detection based on Gower distance.</p></div><div><h3>Methods</h3><p>We present the Gower distance metric to handle diverse feature sets in voice recordings, which acts as a dissimilarity measure for all feature types, making the model adept at identifying subtle patterns indicative of PD. Additionally, the Cuckoo Search algorithm is employed for feature selection, reducing dimensionality by focusing on key features, thereby lessening the computational load associated with high-dimensional datasets.</p></div><div><h3>Results</h3><p>The proposed classifier based on Gower distance resulted in an accuracy rate of 98.3% with feature selection and achieved an accuracy of 94.92% without the feature selection method. It outperforms traditional classifiers and recent studies in PD detection from voice recordings.</p></div><div><h3>Conclusions</h3><p>This accuracy shows the capability of the approach in the correct classification of instances and points out the potential of the approach as a reliable diagnostic tool for the medical practitioner. The findings state that the proposed approach holds promise for improving the diagnosis and monitoring of PD, both within medical institutions and at homes for the elderly.</p></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"191 ","pages":"Article 105583"},"PeriodicalIF":3.7,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1386505624002466/pdfft?md5=8e63bcbdafb3f27ecfbeea695e2e1a42&pid=1-s2.0-S1386505624002466-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141890988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Retraction notice to “Reliability and validity of the Chinese version of the mobile Agnew Relationship Measure (mARM-C)” [Int. J. Med. Inf. 189 (2024) 105482] 关于 "移动阿格纽关系测量（mARM-C）中文版的信度和效度 "的撤稿声明[Int. J. Med. Inf. 189 (2024) 105482]。

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-08-02 DOI: 10.1016/j.ijmedinf.2024.105569

Ran Mo, Shihong Hu

引用次数: 0

Implementing privacy preserving record linkage: Insights from Australian use cases 实施隐私保护记录链接：澳大利亚使用案例的启示。

IF 3.7 2区医学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

International Journal of Medical Informatics

Pub Date : 2024-07-31 DOI: 10.1016/j.ijmedinf.2024.105582

Sean Randall , Adrian Brown , Anna Ferrante , James Boyd , Suzanne Robinson

Objective

To describe the use of privacy preserving linkage methods operationally in Australia, and to present insights and key learnings from their implementation.

Methods

Privacy preserving record linkage (PPRL) utilising Bloom filters provides a unique practical mechanism that allows linkage to occur without the release of personally identifiable information (PII), while still ensuring high accuracy.

Results

The methodology has received wide uptake within Australia, with four state linkage units with privacy preserving capability. It has enabled access to general practice and private pathology data amongst other, both much sought after datasets previous inaccessible for linkage.

Conclusion

The Australian experience suggests privacy preserving linkage is a practical solution for improving data access for policy, planning and population health research. It is hoped interest in this methodology internationally continues to grow.

目的描述澳大利亚在实际操作中使用隐私保护链接方法的情况，并介绍从实施过程中获得的启示和主要经验：方法：利用布鲁姆过滤器的隐私保护记录关联（PPRL）提供了一种独特的实用机制，允许在不泄露个人身份信息（PII）的情况下进行关联，同时仍能确保高准确性：结果：该方法在澳大利亚得到了广泛应用，四个州的联网单位都具备了保护隐私的能力。结果：该方法在澳大利亚得到了广泛应用，有四个州的链接单位具备了保护隐私的能力，从而能够获取普通诊疗和私人病理数据，以及其他以前无法进行链接的、备受追捧的数据集：澳大利亚的经验表明，隐私保护链接是一种切实可行的解决方案，可改善政策、规划和人口健康研究方面的数据访问。希望国际社会对这一方法的兴趣继续增长。

{"title":"Implementing privacy preserving record linkage: Insights from Australian use cases","authors":"Sean Randall , Adrian Brown , Anna Ferrante , James Boyd , Suzanne Robinson","doi":"10.1016/j.ijmedinf.2024.105582","DOIUrl":"10.1016/j.ijmedinf.2024.105582","url":null,"abstract":"<div><h3>Objective</h3><p>To describe the use of privacy preserving linkage methods operationally in Australia, and to present insights and key learnings from their implementation.</p></div><div><h3>Methods</h3><p>Privacy preserving record linkage (PPRL) utilising Bloom filters provides a unique practical mechanism that allows linkage to occur without the release of personally identifiable information (PII), while still ensuring high accuracy.</p></div><div><h3>Results</h3><p>The methodology has received wide uptake within Australia, with four state linkage units with privacy preserving capability. It has enabled access to general practice and private pathology data amongst other, both much sought after datasets previous inaccessible for linkage.</p></div><div><h3>Conclusion</h3><p>The Australian experience suggests privacy preserving linkage is a practical solution for improving data access for policy, planning and population health research. It is hoped interest in this methodology internationally continues to grow.</p></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"191 ","pages":"Article 105582"},"PeriodicalIF":3.7,"publicationDate":"2024-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1386505624002454/pdfft?md5=026b0b8604d90c986fb1838e1fa8008d&pid=1-s2.0-S1386505624002454-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141890990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0