Population aging presents a growing societal challenge and imposes a heavy burden on the healthcare system in many Asian countries. Given the limited availability of formal long-term care (LTC) facilities and personnel, family caregivers play a vital role in providing care for the increasing population of older adults. While awareness of the challenges faced by caregivers is rising, discussions often remain within academic circles, resulting in the lived experiences, well-being, and needs of family caregivers being frequently overlooked. In this review, we identify four key priority areas to advance research, practice, and policy related to family caregivers in Asia: (1) Emphasizing family caregivers as sociocultural navigators in the healthcare system; (2) addressing the mental and physical health needs of family caregivers; (3) recognizing the diverse caregiving experiences across different cultural backgrounds, socioeconomic status, and countries of residence; and (4) strengthening policy support for family caregivers. Our review also identifies deficiencies in institutional LTC and underscores the importance of providing training and empowerment to caregivers. Policymakers, practitioners, and researchers interested in supporting family caregivers should prioritize these key areas to tackle the challenge of population aging in Asian countries. Cross-country knowledge exchange and capacity development are crucial for better serving both the aging population and their caregivers.
{"title":"Caregiving in Asia: Priority areas for research, policy, and practice to support family caregivers","authors":"Nan Jiang, Bei Wu, Yan Li","doi":"10.1002/hcs2.124","DOIUrl":"10.1002/hcs2.124","url":null,"abstract":"<p>Population aging presents a growing societal challenge and imposes a heavy burden on the healthcare system in many Asian countries. Given the limited availability of formal long-term care (LTC) facilities and personnel, family caregivers play a vital role in providing care for the increasing population of older adults. While awareness of the challenges faced by caregivers is rising, discussions often remain within academic circles, resulting in the lived experiences, well-being, and needs of family caregivers being frequently overlooked. In this review, we identify four key priority areas to advance research, practice, and policy related to family caregivers in Asia: (1) Emphasizing family caregivers as sociocultural navigators in the healthcare system; (2) addressing the mental and physical health needs of family caregivers; (3) recognizing the diverse caregiving experiences across different cultural backgrounds, socioeconomic status, and countries of residence; and (4) strengthening policy support for family caregivers. Our review also identifies deficiencies in institutional LTC and underscores the importance of providing training and empowerment to caregivers. Policymakers, practitioners, and researchers interested in supporting family caregivers should prioritize these key areas to tackle the challenge of population aging in Asian countries. Cross-country knowledge exchange and capacity development are crucial for better serving both the aging population and their caregivers.</p>","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 6","pages":"374-382"},"PeriodicalIF":0.0,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11671212/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904704","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}
The COVID-19 pandemic presented unparalleled challenges to prompt and adaptive responses from nations worldwide. This review examines China's multifaceted approach to the crisis, focusing on five key areas of response: infrastructure and system design, medical care and treatment, disease prevention and control, economic and social resilience, and China's engagement in global health. This review demonstrates the effectiveness of a top-down command system at the national level, intersectoral coordination, a legal framework, and public social governance. This study also examines medical care and treatment strategies, highlighting the importance of rapid emergency response, evidence-based treatment, and well-planned vaccination rollout. Further discussion on disease prevention and control measures emphasizes the importance of adaptive measures, timely infection control, transmission interruption, population herd immunity, and technology applications. Socioeconomic impact was also assessed, detailing strategies for disease prevention, material supply, livelihood preservation, and social economy revival. Lastly, we examine China's contributions to the global health community, with a focus on knowledge-sharing, information exchange, and multilateral assistance. While it is true that each nation's response must be tailored to its own context, there are universal lessons to be drawn from China's approach. These insights are pivotal for enhancing global health security, especially as the world navigates evolving health crises.
{"title":"Innovative public strategies in response to COVID-19: A review of practices from China","authors":"You Wu, Zijian Cao, Jing Yang, Xinran Bi, Weiqing Xiong, Xiaoru Feng, Yue Yan, Zeyu Zhang, Zongjiu Zhang","doi":"10.1002/hcs2.122","DOIUrl":"10.1002/hcs2.122","url":null,"abstract":"<p>The COVID-19 pandemic presented unparalleled challenges to prompt and adaptive responses from nations worldwide. This review examines China's multifaceted approach to the crisis, focusing on five key areas of response: infrastructure and system design, medical care and treatment, disease prevention and control, economic and social resilience, and China's engagement in global health. This review demonstrates the effectiveness of a top-down command system at the national level, intersectoral coordination, a legal framework, and public social governance. This study also examines medical care and treatment strategies, highlighting the importance of rapid emergency response, evidence-based treatment, and well-planned vaccination rollout. Further discussion on disease prevention and control measures emphasizes the importance of adaptive measures, timely infection control, transmission interruption, population herd immunity, and technology applications. Socioeconomic impact was also assessed, detailing strategies for disease prevention, material supply, livelihood preservation, and social economy revival. Lastly, we examine China's contributions to the global health community, with a focus on knowledge-sharing, information exchange, and multilateral assistance. While it is true that each nation's response must be tailored to its own context, there are universal lessons to be drawn from China's approach. These insights are pivotal for enhancing global health security, especially as the world navigates evolving health crises.</p>","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 6","pages":"383-408"},"PeriodicalIF":0.0,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11671218/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904714","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}
Haihong Zhang, You Wu, Haibo Wang, Weili Zhao, Yali Cong
<p>On October 19, 2024, the 75th General Assembly of the World Medical Association (WMA) in Finland adopted the latest version of the Declaration of Helsinki (DoH)—Ethical Principles for Medical Research Involving Human Participants [<span>1</span>] (hereafter referred to as “the Declaration”). This revision process took 30 months, with the working group comprising representatives from medical associations in 19 countries and regions. From April 2022 to September 2024, the working group held eight regional expert meetings and two times global consultations, gathering suggestions from both experts and the general public [<span>2</span>]. Besides, the working group developed regular online meeting working mechanisms.</p><p>Basically, the general ethical principles for respecting and protecting human participants are stable over time, while more tailored interpretations and justifications should be adapted in a timely manner. The revision focused on alignment with wide-accepted ethical guidelines at the international level, emphasizing coherence with other related documents within, and beyond the WMA. The Declaration emphasises the overarching principles and does not delve into many specifics; however, its core principles remain universally applicable.</p><p>While these revisions represent significant progress, some also reflect substantial compromises. Notably, to strengthen the protection of research participants’ rights and well-being, the Declaration reaffirms that “<i>These purposes can never take precedence over the rights and interests of individual research participants</i>” and requires all stakeholders, including individuals, teams, and organizations involved in medical research to adhere to ethical principles that respect for and protect of research participants [<span>7</span>]. Given the WMA's mandate as a global organization of physicians, the term “<i>medical research</i>” was retained rather than adopting broader terminology such as “health-related research.” However, the document refers to “<i>physicians</i>,” or “<i>physicians and other researchers</i>” in constituent paragraphs, acknowledging both the critical role of physicians in medical practices and the specialized division of roles in research involving human participants. Compared to the 2013 version, the new Declaration strengthens researchers' responsibilities and protections for research participants (e.g., Articles 9, 10, 12, 17, 21, 23, 32, and 34), increasing the instances of the word “must” from 46 to 58 and clarifying the distinction between “<i>should</i>” and “<i>must</i>.”</p><p>Renaming “<i>subjects</i>” as “<i>participants</i>” not only mandates respect for participants' rights and agency but also calls for a partnership between researchers and participants. Developing such partnership requires that physicians/researchers strictly fulfill their duties, with the best interests of patients, including those participating in research, as the priority, promoting and
{"title":"Sixty years of ethical evolution: The 2024 revision of the Declaration of Helsinki (DoH)","authors":"Haihong Zhang, You Wu, Haibo Wang, Weili Zhao, Yali Cong","doi":"10.1002/hcs2.126","DOIUrl":"10.1002/hcs2.126","url":null,"abstract":"<p>On October 19, 2024, the 75th General Assembly of the World Medical Association (WMA) in Finland adopted the latest version of the Declaration of Helsinki (DoH)—Ethical Principles for Medical Research Involving Human Participants [<span>1</span>] (hereafter referred to as “the Declaration”). This revision process took 30 months, with the working group comprising representatives from medical associations in 19 countries and regions. From April 2022 to September 2024, the working group held eight regional expert meetings and two times global consultations, gathering suggestions from both experts and the general public [<span>2</span>]. Besides, the working group developed regular online meeting working mechanisms.</p><p>Basically, the general ethical principles for respecting and protecting human participants are stable over time, while more tailored interpretations and justifications should be adapted in a timely manner. The revision focused on alignment with wide-accepted ethical guidelines at the international level, emphasizing coherence with other related documents within, and beyond the WMA. The Declaration emphasises the overarching principles and does not delve into many specifics; however, its core principles remain universally applicable.</p><p>While these revisions represent significant progress, some also reflect substantial compromises. Notably, to strengthen the protection of research participants’ rights and well-being, the Declaration reaffirms that “<i>These purposes can never take precedence over the rights and interests of individual research participants</i>” and requires all stakeholders, including individuals, teams, and organizations involved in medical research to adhere to ethical principles that respect for and protect of research participants [<span>7</span>]. Given the WMA's mandate as a global organization of physicians, the term “<i>medical research</i>” was retained rather than adopting broader terminology such as “health-related research.” However, the document refers to “<i>physicians</i>,” or “<i>physicians and other researchers</i>” in constituent paragraphs, acknowledging both the critical role of physicians in medical practices and the specialized division of roles in research involving human participants. Compared to the 2013 version, the new Declaration strengthens researchers' responsibilities and protections for research participants (e.g., Articles 9, 10, 12, 17, 21, 23, 32, and 34), increasing the instances of the word “must” from 46 to 58 and clarifying the distinction between “<i>should</i>” and “<i>must</i>.”</p><p>Renaming “<i>subjects</i>” as “<i>participants</i>” not only mandates respect for participants' rights and agency but also calls for a partnership between researchers and participants. Developing such partnership requires that physicians/researchers strictly fulfill their duties, with the best interests of patients, including those participating in research, as the priority, promoting and ","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 6","pages":"371-373"},"PeriodicalIF":0.0,"publicationDate":"2024-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11671210/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904672","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}
Han Yuan, Chuan Hong, Nguyen Tuan Anh Tran, Xinxing Xu, Nan Liu
Background
Pneumothorax is a medical emergency caused by the abnormal accumulation of air in the pleural space—the potential space between the lungs and chest wall. On 2D chest radiographs, pneumothorax occurs within the thoracic cavity and outside of the mediastinum, and we refer to this area as “lung + space.” While deep learning (DL) has increasingly been utilized to segment pneumothorax lesions in chest radiographs, many existing DL models employ an end-to-end approach. These models directly map chest radiographs to clinician-annotated lesion areas, often neglecting the vital domain knowledge that pneumothorax is inherently location-sensitive.
Methods
We propose a novel approach that incorporates the lung + space as a constraint during DL model training for pneumothorax segmentation on 2D chest radiographs. To circumvent the need for additional annotations and to prevent potential label leakage on the target task, our method utilizes external datasets and an auxiliary task of lung segmentation. This approach generates a specific constraint of lung + space for each chest radiograph. Furthermore, we have incorporated a discriminator to eliminate unreliable constraints caused by the domain shift between the auxiliary and target datasets.
Results
Our results demonstrated considerable improvements, with average performance gains of 4.6%, 3.6%, and 3.3% regarding intersection over union, dice similarity coefficient, and Hausdorff distance. These results were consistent across six baseline models built on three architectures (U-Net, LinkNet, or PSPNet) and two backbones (VGG-11 or MobileOne-S0). We further conducted an ablation study to evaluate the contribution of each component in the proposed method and undertook several robustness studies on hyper-parameter selection to validate the stability of our method.
Conclusions
The integration of domain knowledge in DL models for medical applications has often been underemphasized. Our research underscores the significance of incorporating medical domain knowledge about the location-specific nature of pneumothorax to enhance DL-based lesion segmentation and further bolster clinicians' trust in DL tools. Beyond pneumothorax, our approach is promising for other thoracic conditions that possess location-relevant characteristics.
{"title":"Leveraging anatomical constraints with uncertainty for pneumothorax segmentation","authors":"Han Yuan, Chuan Hong, Nguyen Tuan Anh Tran, Xinxing Xu, Nan Liu","doi":"10.1002/hcs2.119","DOIUrl":"10.1002/hcs2.119","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Pneumothorax is a medical emergency caused by the abnormal accumulation of air in the pleural space—the potential space between the lungs and chest wall. On 2D chest radiographs, pneumothorax occurs within the thoracic cavity and outside of the mediastinum, and we refer to this area as “lung + space.” While deep learning (DL) has increasingly been utilized to segment pneumothorax lesions in chest radiographs, many existing DL models employ an end-to-end approach. These models directly map chest radiographs to clinician-annotated lesion areas, often neglecting the vital domain knowledge that pneumothorax is inherently location-sensitive.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>We propose a novel approach that incorporates the lung + space as a constraint during DL model training for pneumothorax segmentation on 2D chest radiographs. To circumvent the need for additional annotations and to prevent potential label leakage on the target task, our method utilizes external datasets and an auxiliary task of lung segmentation. This approach generates a specific constraint of lung + space for each chest radiograph. Furthermore, we have incorporated a discriminator to eliminate unreliable constraints caused by the domain shift between the auxiliary and target datasets.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>Our results demonstrated considerable improvements, with average performance gains of 4.6%, 3.6%, and 3.3% regarding intersection over union, dice similarity coefficient, and Hausdorff distance. These results were consistent across six baseline models built on three architectures (U-Net, LinkNet, or PSPNet) and two backbones (VGG-11 or MobileOne-S0). We further conducted an ablation study to evaluate the contribution of each component in the proposed method and undertook several robustness studies on hyper-parameter selection to validate the stability of our method.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>The integration of domain knowledge in DL models for medical applications has often been underemphasized. Our research underscores the significance of incorporating medical domain knowledge about the location-specific nature of pneumothorax to enhance DL-based lesion segmentation and further bolster clinicians' trust in DL tools. Beyond pneumothorax, our approach is promising for other thoracic conditions that possess location-relevant characteristics.</p>\u0000 </section>\u0000 </div>","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 6","pages":"456-474"},"PeriodicalIF":0.0,"publicationDate":"2024-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11671217/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904715","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}
Somit Jain, Shobhit Agrawal, Eshaan Mohapatra, Kathiravan Srinivasan
Background
The global impact of the highly contagious COVID-19 virus has created unprecedented challenges, significantly impacting public health and economies worldwide. This research article conducts a time series analysis of COVID-19 data across various countries, including India, Brazil, Russia, and the United States, with a particular emphasis on total confirmed cases.
Methods
The proposed approach combines auto-regressive integrated moving average (ARIMA)'s ability to capture linear trends and seasonality with long short-term memory (LSTM) networks, which are designed to learn complex nonlinear dependencies in the data. This hybrid approach surpasses both individual models and existing ARIMA-artificial neural network (ANN) hybrids, which often struggle with highly nonlinear time series like COVID-19 data. By integrating ARIMA and LSTM, the model aims to achieve superior forecasting accuracy compared to baseline models, including ARIMA, Gated Recurrent Unit (GRU), LSTM, and Prophet.
Results
The hybrid ARIMA-LSTM model outperformed the benchmark models, achieving a mean absolute percentage error (MAPE) score of 2.4%. Among the benchmark models, GRU performed the best with a MAPE score of 2.9%, followed by LSTM with a score of 3.6%.
Conclusions
The proposed ARIMA-LSTM hybrid model outperforms ARIMA, GRU, LSTM, Prophet, and the ARIMA-ANN hybrid model when evaluating using metrics like MAPE, symmetric mean absolute percentage error, and median absolute percentage error across all countries analyzed. These findings have the potential to significantly improve preparedness and response efforts by public health authorities, allowing for more efficient resource allocation and targeted interventions.
{"title":"A novel ensemble ARIMA-LSTM approach for evaluating COVID-19 cases and future outbreak preparedness","authors":"Somit Jain, Shobhit Agrawal, Eshaan Mohapatra, Kathiravan Srinivasan","doi":"10.1002/hcs2.123","DOIUrl":"10.1002/hcs2.123","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>The global impact of the highly contagious COVID-19 virus has created unprecedented challenges, significantly impacting public health and economies worldwide. This research article conducts a time series analysis of COVID-19 data across various countries, including India, Brazil, Russia, and the United States, with a particular emphasis on total confirmed cases.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>The proposed approach combines auto-regressive integrated moving average (ARIMA)'s ability to capture linear trends and seasonality with long short-term memory (LSTM) networks, which are designed to learn complex nonlinear dependencies in the data. This hybrid approach surpasses both individual models and existing ARIMA-artificial neural network (ANN) hybrids, which often struggle with highly nonlinear time series like COVID-19 data. By integrating ARIMA and LSTM, the model aims to achieve superior forecasting accuracy compared to baseline models, including ARIMA, Gated Recurrent Unit (GRU), LSTM, and Prophet.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>The hybrid ARIMA-LSTM model outperformed the benchmark models, achieving a mean absolute percentage error (MAPE) score of 2.4%. Among the benchmark models, GRU performed the best with a MAPE score of 2.9%, followed by LSTM with a score of 3.6%.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>The proposed ARIMA-LSTM hybrid model outperforms ARIMA, GRU, LSTM, Prophet, and the ARIMA-ANN hybrid model when evaluating using metrics like MAPE, symmetric mean absolute percentage error, and median absolute percentage error across all countries analyzed. These findings have the potential to significantly improve preparedness and response efforts by public health authorities, allowing for more efficient resource allocation and targeted interventions.</p>\u0000 </section>\u0000 </div>","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 6","pages":"409-425"},"PeriodicalIF":0.0,"publicationDate":"2024-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11671211/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904696","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}
Chenlin Du, Zeyu Zhang, Baoqin Liu, Zijian Cao, Nan Jiang, Zongjiu Zhang
Background
Frailty in older adults is linked to increased risks and lower quality of life. Pre-frailty, a condition preceding frailty, is intervenable, but its determinants and assessment are challenging. This study aims to develop and validate an explainable machine learning model for pre-frailty risk assessment among community-dwelling older adults.
Methods
The study included 3141 adults aged 60 or above from the China Health and Retirement Longitudinal Study. Pre-frailty was characterized by one or two criteria from the physical frailty phenotype scale. We extracted 80 distinct features across seven dimensions to evaluate pre-frailty risk. A model was constructed using recursive feature elimination and a stacking-CatBoost distillation module on 80% of the sample and validated on a separate 20% holdout data set.
Results
The study used data from 2508 community-dwelling older adults (mean age, 67.24 years [range, 60–96]; 1215 [48.44%] females) to develop a pre-frailty risk assessment model. We selected 57 predictive features and built a distilled CatBoost model, which achieved the highest discrimination (AUROC: 0.7560 [95% CI: 0.7169, 0.7928]) on the 20% holdout data set. The living city, BMI, and peak expiratory flow (PEF) were the three most significant contributors to pre-frailty risk. Physical and environmental factors were the top 2 impactful feature dimensions.
Conclusions
An accurate and interpretable pre-frailty risk assessment framework using state-of-the-art machine learning techniques and explanation methods has been developed. Our framework incorporates a wide range of features and determinants, allowing for a comprehensive and nuanced understanding of pre-frailty risk.
{"title":"Explainable machine learning model for pre-frailty risk assessment in community-dwelling older adults","authors":"Chenlin Du, Zeyu Zhang, Baoqin Liu, Zijian Cao, Nan Jiang, Zongjiu Zhang","doi":"10.1002/hcs2.120","DOIUrl":"10.1002/hcs2.120","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Frailty in older adults is linked to increased risks and lower quality of life. Pre-frailty, a condition preceding frailty, is intervenable, but its determinants and assessment are challenging. This study aims to develop and validate an explainable machine learning model for pre-frailty risk assessment among community-dwelling older adults.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>The study included 3141 adults aged 60 or above from the China Health and Retirement Longitudinal Study. Pre-frailty was characterized by one or two criteria from the physical frailty phenotype scale. We extracted 80 distinct features across seven dimensions to evaluate pre-frailty risk. A model was constructed using recursive feature elimination and a stacking-CatBoost distillation module on 80% of the sample and validated on a separate 20% holdout data set.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>The study used data from 2508 community-dwelling older adults (mean age, 67.24 years [range, 60–96]; 1215 [48.44%] females) to develop a pre-frailty risk assessment model. We selected 57 predictive features and built a distilled CatBoost model, which achieved the highest discrimination (AUROC: 0.7560 [95% CI: 0.7169, 0.7928]) on the 20% holdout data set. The living city, BMI, and peak expiratory flow (PEF) were the three most significant contributors to pre-frailty risk. Physical and environmental factors were the top 2 impactful feature dimensions.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>An accurate and interpretable pre-frailty risk assessment framework using state-of-the-art machine learning techniques and explanation methods has been developed. Our framework incorporates a wide range of features and determinants, allowing for a comprehensive and nuanced understanding of pre-frailty risk.</p>\u0000 </section>\u0000 </div>","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 6","pages":"426-437"},"PeriodicalIF":0.0,"publicationDate":"2024-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11671213/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904707","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}
Skin cancer poses a significant global health threat, with early detection being essential for successful treatment. While deep learning algorithms have greatly enhanced the categorization of skin lesions, the black-box nature of many models limits interpretability, posing challenges for dermatologists.
Methods
To address these limitations, SkinSage XAI utilizes advanced explainable artificial intelligence (XAI) techniques for skin lesion categorization. A data set of around 50,000 images from the Customized HAM10000, selected for diversity, serves as the foundation. The Inception v3 model is used for classification, supported by gradient-weighted class activation mapping and local interpretable model-agnostic explanations algorithms, which provide clear visual explanations for model outputs.
Results
SkinSage XAI demonstrated high performance, accurately categorizing seven types of skin lesions—dermatofibroma, benign keratosis, melanocytic nevus, vascular lesion, actinic keratosis, basal cell carcinoma, and melanoma. It achieved an accuracy of 96%, with precision at 96.42%, recall at 96.28%, F1 score at 96.14%, and an area under the curve of 99.83%.
Conclusions
SkinSage XAI represents a significant advancement in dermatology and artificial intelligence by bridging gaps in accuracy and explainability. The system provides transparent, accurate diagnoses, improving decision-making for dermatologists and potentially enhancing patient outcomes.
{"title":"SkinSage XAI: An explainable deep learning solution for skin lesion diagnosis","authors":"Geetika Munjal, Paarth Bhardwaj, Vaibhav Bhargava, Shivendra Singh, Nimish Nagpal","doi":"10.1002/hcs2.121","DOIUrl":"10.1002/hcs2.121","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Skin cancer poses a significant global health threat, with early detection being essential for successful treatment. While deep learning algorithms have greatly enhanced the categorization of skin lesions, the black-box nature of many models limits interpretability, posing challenges for dermatologists.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>To address these limitations, SkinSage XAI utilizes advanced explainable artificial intelligence (XAI) techniques for skin lesion categorization. A data set of around 50,000 images from the Customized HAM10000, selected for diversity, serves as the foundation. The Inception v3 model is used for classification, supported by gradient-weighted class activation mapping and local interpretable model-agnostic explanations algorithms, which provide clear visual explanations for model outputs.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>SkinSage XAI demonstrated high performance, accurately categorizing seven types of skin lesions—dermatofibroma, benign keratosis, melanocytic nevus, vascular lesion, actinic keratosis, basal cell carcinoma, and melanoma. It achieved an accuracy of 96%, with precision at 96.42%, recall at 96.28%, <span><i>F</i><sub>1</sub></span> score at 96.14%, and an area under the curve of 99.83%.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>SkinSage XAI represents a significant advancement in dermatology and artificial intelligence by bridging gaps in accuracy and explainability. The system provides transparent, accurate diagnoses, improving decision-making for dermatologists and potentially enhancing patient outcomes.</p>\u0000 </section>\u0000 </div>","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 6","pages":"438-455"},"PeriodicalIF":0.0,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11671215/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904685","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}
This study investigates the geographical distribution of private hospitals in Australian capital cities in relation to the Index of Relative Socioeconomic Disadvantage.
Methods
Using Geographic Information System analysis, the study examined how private hospitals are distributed across different socioeconomic quartiles, providing a comprehensive visualisation of health care accessibility.
Results
The results indicate an unequal distribution with a substantial concentration of private hospitals within the vicinity of communities classified in the highest socioeconomic classification. This raises significant concerns about health care equity, particularly in light of the increased strain on health care systems before, during and after the COVID-19 pandemic.
Conclusions
This study underscores the need for targeted policy interventions to enhance the resilience and accessibility of the private health care sector, specifically targeting disadvantaged communities. It suggests that comprehensive, geographically-informed data is crucial for policymakers to make informed decisions that promote health equity in the postpandemic landscape.
{"title":"Are private hospital emergency departments in Australia distributed to serve the wealthy community?","authors":"Mazen Baazeem, Estie Kruger, Marc Tennant","doi":"10.1002/hcs2.112","DOIUrl":"10.1002/hcs2.112","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Objective</h3>\u0000 \u0000 <p>This study investigates the geographical distribution of private hospitals in Australian capital cities in relation to the Index of Relative Socioeconomic Disadvantage.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>Using Geographic Information System analysis, the study examined how private hospitals are distributed across different socioeconomic quartiles, providing a comprehensive visualisation of health care accessibility.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>The results indicate an unequal distribution with a substantial concentration of private hospitals within the vicinity of communities classified in the highest socioeconomic classification. This raises significant concerns about health care equity, particularly in light of the increased strain on health care systems before, during and after the COVID-19 pandemic.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>This study underscores the need for targeted policy interventions to enhance the resilience and accessibility of the private health care sector, specifically targeting disadvantaged communities. It suggests that comprehensive, geographically-informed data is crucial for policymakers to make informed decisions that promote health equity in the postpandemic landscape.</p>\u0000 </section>\u0000 </div>","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 5","pages":"287-297"},"PeriodicalIF":0.0,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11520243/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142549985","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}
Ge Wu, Mengchun Gong, You Wu, Li Liu, Boyang Shi, Zhirong Zeng
We summarized the unique challenges that China faced in digital health due to its large population, regional disparities, and uneven distribution of medical resources. Under the guidance of the Global Initiative on Digital Health (GIDH) released by WHO, we proposed corresponding solutions that address infrastructure, data, terminology, technology and security.
{"title":"Advancing digital health in China: Aligning challenges, opportunities, and solutions with the Global Initiative on Digital Health (GIDH)","authors":"Ge Wu, Mengchun Gong, You Wu, Li Liu, Boyang Shi, Zhirong Zeng","doi":"10.1002/hcs2.118","DOIUrl":"10.1002/hcs2.118","url":null,"abstract":"<p>We summarized the unique challenges that China faced in digital health due to its large population, regional disparities, and uneven distribution of medical resources. Under the guidance of the Global Initiative on Digital Health (GIDH) released by WHO, we proposed corresponding solutions that address infrastructure, data, terminology, technology and security.\u0000 <figure>\u0000 <div><picture>\u0000 <source></source></picture><p></p>\u0000 </div>\u0000 </figure></p>","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 5","pages":"365-369"},"PeriodicalIF":0.0,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11520242/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142549974","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}
In this commentary, we elucidate three indispensable evaluation steps toward the real-world deployment of machine learning within the healthcare sector and demonstrate referable examples for diagnostic, therapeutic, and prognostic tasks. We encourage researchers to move beyond retrospective and within-sample validation, and step into the practical implementation at the bedside rather than leaving developed machine learning models in the dust of archived literature.
{"title":"Toward real-world deployment of machine learning for health care: External validation, continual monitoring, and randomized clinical trials","authors":"Han Yuan","doi":"10.1002/hcs2.114","DOIUrl":"10.1002/hcs2.114","url":null,"abstract":"<p>In this commentary, we elucidate three indispensable evaluation steps toward the real-world deployment of machine learning within the healthcare sector and demonstrate referable examples for diagnostic, therapeutic, and prognostic tasks. We encourage researchers to move beyond retrospective and within-sample validation, and step into the practical implementation at the bedside rather than leaving developed machine learning models in the dust of archived literature.\u0000 <figure>\u0000 <div><picture>\u0000 <source></source></picture><p></p>\u0000 </div>\u0000 </figure></p>","PeriodicalId":100601,"journal":{"name":"Health Care Science","volume":"3 5","pages":"360-364"},"PeriodicalIF":0.0,"publicationDate":"2024-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11520244/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142549989","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号