Detection of Onset and Progression of Osteoporosis Using Machine Learning

• Published in: Machine Learning for Healthcare Applications pp. 137 - 149
• Main Authors: Kerketta, Shilpi Ruchi, Ghosh, Debalina
• Format: Book Chapter
• Language: English
• Published: United States John Wiley & Sons, Incorporated 2021; John Wiley & Sons, Inc
• Subjects: accuracy; classifier; decision tree classifier; KNN classifier; machine learning (ML) algorithm; microwave setup; Osteoporosis; random forest classifier
• ISBN: 9781119791812; 1119791812
• DOI: 10.1002/9781119792611.ch9

Osteoporosis, an immedicable and progressive decline of bone strength caused by reduction in bone mineral density, is widely prevalent among post‐menopausal women and aging population. A timely and precise diagnosis of osteoporosis before a major breakage occurs is a challenging task. The prevalent techniques used for diagnosis of osteoporosis such as dual‐energy X‐ray and quantitative computed tomography, utilize ionizing signals and are time‐consuming, expensive and non‐portable. In recent times, biomedical estimation through microwave techniques aided by machine learning algorithms has produced some interesting outcomes. This article demonstrates the successful incorporation of microwave measurements along with machine learning algorithms to identify the various stages of onset and progression of osteoporosis. With the help of machine learning, it is possible to calculate the degree of bone mineral loss which can map the progression of osteoporosis. This will greatly assist the medical fraternity in treating the condition. The purpose of this article is to make a well structured and accurate machine learning technique for the early diagnosis of osteoporosis.