Acceso usuarios
Jae Young Kim, Ji Hye Kim, Ea Wha Kang, Tae Ik Chang, Yong Kyu Lee, Kyung-Shin Park, Seok Young So, Seung Hyun Kim, Byung Jun Bae, Jeong Yeol Baek, Sug Kyun Shin, Mihyeon Kim, Dong-Ho Park
This study seeks to evaluate the ability of machine learning methods to predict the dry weight of chronic hemodialysis athletes. The researcher has reached out to kidney patients who have had to give up sports and athletic careers due to chronic hemodialysis. This paper explores the development of medical prediction algorithms that combine image analysis with numerical data, which is widely used in the field of medicine. This deep learning method is widely employed to enhance the treatment of athletes who have kidney conditions. Regular hemodialysis is crucial for maintaining the health of athletes who have kidney disease. Accurately predicting dry weight is a crucial step in the process of performing hemodialysis. In this context, dry weight refers to the optimal moisture level at which excess water is effectively eliminated from the patient (athletes) through ultrafiltration during hemodialysis. In order to accurately determine the optimal amount of hemodialysis, predicting the correct dry weight is crucial. However, this task is quite challenging and often yields inaccurate results due to the extensive data analysis required by experienced nephrologists. This paper presents a deep learning methodology utilising the Artificial Neural Network (ANN) approach to efficiently address these issues. The proposed method aims to predict dry weight rapidly by analysing image values and clinical data from X-ray images obtained during routine check-ups. The current study has several theoretical and practical implications. This study contributes to the existing literature on chronic hemodialysis and the dry weight of athletes, offering valuable insights to sports health organisations. By doing so, these organisations can effectively prepare to proactively evaluate the atypical health conditions of athletes.
