Challenges and opportunities of the implementation of machine learning in geriatric clinical care: A systematic review

Background and aims: Machine learning can play a key role in preventing, diagnosing andtreating the problems of elderly patients. However, the implementation of machine learning forgeriatric clinical care has associated challenges. The aim of this study was to assess the challengesand opportunities of the implementation of machine learning in geriatric clinical care.Method: This systematic review was carried out utilizing the Preferred Reporting Items for SystematicReviews and Meta-Analyses (PRISMA) guidelines. An extensive search was carriedin online databases including PubMed, Web of Science, Scopus, Google Scholar, and ProQuestwith the keywords such as “Machine Learning”, “Transfer Learning”, “Artificial Intelligence”,“Geriatric”, “Elderly”, “Care”, and “Clinical Care”, from the earliest records up to October ۲۰,۲۰۲۲. Also, all English-language studies related to the purpose of the present study were included.Letters to the editor, opinions, conference abstracts, intervention, reviews were excluded fromthis study. The appraisal tool for cross-sectional studies (AXIS tool) was used to assess the qualityof included studies. All stages of search and quality evaluation of articles were conducted by tworesearchers, independently.Results: A total of ۱۰ out of ۱۳۴ studies were included in the study. The challenges of the implementationof machine learning in geriatric clinical care were including racial biases (n=۷), lack ofprivacy (n=۷), inequality (n=۶), insecurity (n=۶), disruption of human communication and datamanagement (n=۶), cost of care (n=۶), and annotation problem (n=۵). There are many opportunitiesto implement machine learning to improve geriatric care in the clinical setting. These opportunitiesinclude the automation of clinical tasks (n=۶), optimization of decision-making (n=۵),clinical support in practice (n=۵), expansion of clinical capacity (n=۴), improvement of the safetylevel of elderly patients and increase in the quality of their care (n=۲).Conclusion: One of the strategies for improving the problems of elderly patients, reducing costsand increasing the health of patients is the use of machine learning. The most challenges of theimplementation of machine learning in geriatric clinical care were including racial biases, lack ofprivacy, inequality, insecurity, disruption of human communication and data management, cost ofcare, and annotation problem. In general, the implementation of machine learning to improve thegeriatric clinical care is a questionable hypothesis that requires additional evidence. Therefore,these challenges are still a major concern for the implementation of ML in geriatric clinical care.Hence, more research is needed to address the challenges of using machine learning for geriatricclinical care.

کلیدواژه ها:

machine learning ، artificial intelligence ، geriatric