Artificial Intelligence in Diagnosis management of Carpal Tunnel Syndrome: A Narrative Review

سال انتشار: 1401
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 191

نسخه کامل این مقاله ارائه نشده است و در دسترس نمی باشد

این مقاله در بخشهای موضوعی زیر دسته بندی شده است:

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

HUMS04_090

تاریخ نمایه سازی: 3 خرداد 1402

چکیده مقاله:

Introduction and Objective: In recent times using artificial intelligence (AI) and deep learning (DL) elements in diagnosing carpal tunnel syndrome (CTS) has been emerging. Our goal is to have a thorough review of studies of these fields and compare them in a wholeMethod: A thorough search was performed in the international database of Pubmed with the keywords identified as 'carpal tunnel syndrome', 'Artificial intelligence', and 'deep learning'. ۳۲ results were found during the search and ۷ of them were excluded from our search.Results:We found that recent research in this field is growing rapidly and proving that using AI in diagnosing CTS can be very beneficial for healthcare providers.In research that was done, they used different approaches to diagnose and classify CTS to degrees from mild to severe that affect on management of the disease.All of them used Ultrasound as the only or one of the major means of pinpointing CTS and assessing the severity of CTS based on different variables. Cross-sectional area(CSA), on the other hand, was used as the primary criterion in almost all of them.Studies also showed that using other criterion like flattening ratio and palmar bowing of Median nerve (MN), kinematic features of the hand, and clinical data about CTS like duration of symptoms, Numeric rating scale of pain, and thenar muscle weakness can improve the training of models that will be used for diagnosis of CTS.It also illustrated that there is a difference between the algorithms of training models.Conclusion: Current research aims to have a comprehensive review of AI and DL assisting healthcare providers in diagnosing CTS.The evidence in this study suggests that the usage of AI and DL in the diagnosis of CTS can improve the overall performance of radiologists and make their work more efficient.The limitations of the studies we reviewed were that their dataset was relatively limited and almost every time they came from one center. Also, these studies haven't included an expanded range of anatomical variants.A further study could assess more anatomical variants, also having bigger datasets and more criterion besides CSA can improve model training.

نویسندگان

Elias Mohammad Alegh

Student Research Committee, Babol University of Medical Sciences, Babol, Iran

Mahdi Shakeri

Student Research Committee, Babol University of Medical Sciences, Babol, Iran- USERN Office, Babol University of Medical Sciences, Babol, Iran

Ali Kor

Student Research Committee, Babol University of Medical Sciences, Babol, Iran

Javad Kamali

Student Research Committee, Babol University of Medical Sciences, Babol, Iran

Rashid Gharanjik

Student Research Committee, Babol University of Medical Sciences, Babol, Iran

Hamed Ahmadi

Student Research Committee, Babol University of Medical Sciences, Babol, Iran