A Multi-Layer Perceptron Neural Network for Predicting the Diagnosis of Osteoporosis in Women Using Physical Activity Factors

다국어 초록 (Multilingual Abstract)

Background Osteoporosis (OP) is a bone disease caused by a decrease in bone mineral density (BMD). OP is common in women because BMD gradually decreases after age 35. OP due to decreased BMD is highly likely to cause fatal traumatic injuries such as hip fracture.
Purpose The purpose of this study was developed and evaluated a multi-layer perceptron neural network model that predicts OP using physical characteristics and activity factors of adult women over the age of 35 whose BMD begins to decline.
Study design Cross-sectional study.
Methods Data from KNHANES were used to develop a multi-layer perceptron model for predicting OP. Data preprocessing included variable selection and sample balancing, and LASSO was used for feature selection. The model used 5 hidden layers, dropout and batch normalization and was evaluated using evaluation scores such as accuracy and recall score calculated from a confusion matrix.
Results Models were trained and evaluated to predict OP using selected features including age, quality of life index, weight, grip strength and average working hours per week. The model achieved 76.8% accuracy, 74.5% precision, 80.5% recall, 77.4% F1 score, and 74.8% ROC AUC.
Conclusions A multi-layer perceptron neural network for predicting OP diagnosis using physical characteristics and activity factors in women aged 35 years or older showed relatively good performance. Since the selected variables can be easily measured through surveys, assessment tool, and digital hand dynamometer, this model will be useful for screening elderly women with OP or not in areas with poor medical facilities or difficult access.

참고문헌 (Reference)

• 1 Bruno AG, "Vertebral size, bone density, and strength in men and women matched for age and areal spine BMD" 29 (29): 562-569, 2014
• 2 Mueller MJ, "Tissue adaptation to physical stress : a proposed"Physical Stress Theory"to guide physical therapist practice, education, and research" 82 (82): 383-403, 2002
• 3 Schmitt NM, "The role of physical activity in the prevention of osteoporosis in postmen opausal women an update" 63 (63): 3438-, 2009
• 4 Sinaki M, "The role of exercise in the treatment of osteoporosis" 8 : 138-144, 2010
• 5 Blake GM, "The role of DXA bone density scans in the diagnosis and treatment of osteoporosis" 83 (83): 509-517, 2007
• 6 Herbert AJ, "The interactions of physical activity, exercise and genetics and their associations with bone mineral density : implications for injury risk in elite athletes" 119 : 29-47, 2019
• 7 Felson DT, "The effect of postmenopausal estrogen therapy on bone density in elderly women" 329 (329): 1141-1146, 1993
• 8 Khosla S, "Relationship of serum sex steroid levels and bone turnover markers with bone mineral density in men and women : a key role for bioavailable estrogen" 83 (83): 22662274-, 1998
• 9 김상준 ; 박은철 ; 양원규 ; 조은, "Relationship between Weight, Body Mass Index and Bone Mineral Density of Lumbar Spine in Women" 대한골대사학회 19 (19): 95-102, 2012
• 10 Seeman E, "Reduced bone density in women with fractures : contribution of low peak bone density and rapid bone loss" 4 : 15-25, 1994
• 11 Yasaka K, "Prediction of bone mineral density from computed to mography : application of deep learning with a convolu tional neural network" 30 : 3549-3557, 2020
• 12 Wang Y, "Prediction model for the risk of osteoporosis incorporating factors of disease history and living habits in physical examination of population in Chongqing, Southwest China : based on artificial neural network" 21 (21): 1-10, 2021
• 13 Lorentzon M, "Osteoporosis : the evolu tion of a diagnosis" 277 (277): 650-661, 2015
• 14 Bukowska Damska A, "Night shift work and osteoporosis : evidence and hypothesis" 36 (36): 171-180, 2019
• 15 Guzon Illescas O, "Mortality after osteoporotic hip fracture : inci dence, trends, and associated factors" 14 (14): 1-9, 2019
• 16 Solou K, "Morbidity and mortality after second hip fracture with and without nursing care program" 14 (14): 2022
• 17 Marshall D, "Meta analysis of how well measures of bone mineral density predict occur rence of osteoporotic fractures" 312 (312): 1254-1259, 1996
• 18 Muthukrishnan R, "LASSO: a feature selection technique in predictive modeling for machine learning" 18-20, 2016
• 19 Rutherford OM, "Is there a role for exercise in the prevention of osteoporotic fractures?" 33 (33): 378-386, 1999
• 20 Pickhardt PJ, "Improved CT based osteoporosis assessment with a fully automated deep learning tool" 4 (4): e220042-, 2022
• 21 Omsland TK, "Hip fractures in Norway 1999-2008 : time trends in total incidence and second hip fracture rates A NOREPOS study" 27 : 807-814, 2012
• 22 Clarke BL, "Female reproductive system and bone" 503 (503): 118-128, 2010
• 23 Voss LA, "Exercise induced loss of bone density in athletes" 6 (6): 349-357, 1998
• 24 Jordan KM, "Epidemiology of osteoporosis" 16 (16): 795-806, 2002
• 25 Cadore EL, "Effects of the physical activity on the bone mineral density and bone remodelation" 11 : 373-379, 2005
• 26 Cadore EL, "Effects of the physical activity on the bone mineral density and bone remodelation" 11 : 373-379, 2005
• 27 Lochmüller EM, "Effect of fixation, soft tissues, and scan projection on bone mineral measure ments with dual energy X ray absorptiometry(DXA)" 68 (68): 2001
• 28 Devlin NJ, "EQ 5D and the EuroQol group : past, present and future" 15 (15): 127-137, 2017
• 29 Zhang B, "Deep learning of lumbar spine X ray for osteopenia and osteoporosis screening : a multicenter retrospective cohort study" 140 : 115561-, 2020
• 30 Yamamoto N, "Deep learning for osteoporosis classification using hip radio graphs and patient clinical covariates" 10 (10): 1534-, 2020
• 31 Sato Y, "Deep learning for bone mineral density and T score prediction from chest X rays: a multicenter study" 10 (10): 2323-, 2022
• 32 Lu Y, "Classification of osteoporosis based on bone mineral densities" 16 (16): 901-910, 2001
• 33 Düppe H, "Bone mineral density, muscle strength and physical activity : a population based study of 332 subjects aged 15-42 years" 68 (68): 97-103, 1997
• 34 Düppe H, "Bone mineral density, muscle strength and physical activity : a population based study of 332 subjects aged 15-42 years" 68 (68): 97-103, 1997
• 35 Morel J, "Bone mineral density of 704 amateur sportsmen involved in different physical activities" 12 : 152-157, 2001
• 36 Morel J, "Bone mineral density of 704 amateur sportsmen involved in different physical activities" 12 : 152-157, 2001
• 37 Popat VB, "Bone mineral density in estrogen deficient young women" 94 (94): 2277-2283, 2009
• 38 Nguyen TV, "Bone loss, physical activity, and weight change in elderly women : the Dubbo Osteoporosis Epidemiology Study" 13 (13): 1458-1467, 1998
• 39 Manolagas SC, "Birth and death of bone cells : basic regulatory mechanisms and implications for the path ogenesis and treatment of osteoporosis" 21 (21): 115-137, 2000
• 40 Luo Y, "Association between grip strength and bone mineral density in general US population of NHANES 2013-2014" 15 : 1-9, 2020
• 41 Luo Y, "Association between grip strength and bone mineral density in general US popula tion of NHANES 2013-2014" 15 : 1-9, 2020
• 42 Ho CS, "Application of deep learning neural network in predicting bone mineral density from plain X ray radiography" 16 : 1-12, 2021
• 43 Lee EY, "Age related bone mineral density patterns in Koreans (KNHANES IV)" 97 (97): 3310-3318, 2012
• 44 Alarcón T, "Activities of daily living after hip fracture : profile and rate of recovery during 2 years of follow up" 22 : 1609-1613, 2011
• 45 Wolman RL, "ABC of sports medicine : osteoporosis and exercise" 309 (309): 400-, 1994