OSTEOSHIELD: CONTINUOUS AI-BASED RISK EVALUATION SYSTEM FOR OSTEOPOROSIS AND FRACTURE PREVENTION USING WEARABLE SENSORS
OSTEOSHIELD: CONTINUOUS AI-BASED RISK EVALUATION SYSTEM FOR OSTEOPOROSIS AND FRACTURE PREVENTION USING WEARABLE SENSORS
Authors:
1.MATCHA VIJAY KUMAR , 2.MANNE PALLI AJAY KUMAR , 3.KARTHIKEYAN.V ,4. DR. F. ANTONY XAVIER BRONSON ,
5.Dr. M.ANAND .
1,2,3Students, Dept. of Computer Science and Engineering, 4,5Assistant Professor, Dept. of Computer Science and Engineering, 6 Guide
,Professor, Dept. of Computer Science & Mechanical Engineering, Dr. M. G. R Educational and Research Institute, Chennai, India matchavijaykumar123@gmail.com, mannepalliajaykumar84@gmail.com,
Abstract—Osteoshed introduces a sustained osteoporosis risk- lay-out process that assists in preventing fractures by integrating wearable sensors as well as statistical learning models. Osteoporosis presents itself in the form of progressive skeletal weakness in relation to loss of bone mineral density and integrity of bone microstructure, which often leads to debilitating fractures in geriatric populations. The use of conventional screening methods like Dual-Energy X-ray Absorptiometry is still clinically viable; nevertheless, due to its limited availability and high cost of operation, this technology cannot be used on a large scale.Proposed research paper includes an Arduino Uno platform connected with accelerator, temperature sensor, flex sensor, heartbeat sensor, and piezoelectric sensor to gather biomechanical and physiological data regarding the bone health. The processed parameters are arranged into table CSV data sets and analyzed with the help of Logistic Regression, which is a supervised machine learning model that is applicable to probabilistic classification and ability to interpret features. Experimental evidence shows that predictive modeling classifies subjects into the low-risk, moderate-risk and high-risk category with classification accuracy of 94.2 and better sensitivity in high risk classification conditions. The suggested output implementation model combines LCD-based real-time parameters display and buzzer- based alert systems to abnormal physiological deviations. Outcome provides a clinical support tool, an affordable, portable and data-driven mechanism of continuous surveillance and preventive healthcare delivery in environments with limited resources.
Keywords— Osteoporosis vulnerability, Wearable sensors, Machine Learning, Logistical Regression Classification, Prediction of bone density, Prediction of fracture risk, Physiological signal analysis, Real time health monitoring, Predictive analytics in healthcare, Clinical decision support system.