Researchers have developed a new deep learning algorithm that has shown to outperform existing computer-based methods for predicting osteoporosis risk. This new algorithm could potentially lead to earlier diagnoses and better outcomes for patients at risk of osteoporosis, a bone disease that can lead to fractures.
The study, published in the journal Frontiers in Artificial Intelligence, tested the deep neural network (DNN) model against four conventional machine learning algorithms and a traditional regression model using data from over 8,000 participants aged 40 and older in the Louisiana Osteoporosis Study. The DNN model achieved the best overall predictive performance, accurately identifying true positives and avoiding mistakes.
Lead author Chuan Qiu, a research assistant professor at the Tulane School of Medicine Center, emphasized the importance of early detection of osteoporosis risk for preventative measures. The researchers also identified the 10 most important factors for predicting osteoporosis risk, including weight, age, gender, grip strength, height, and lifestyle factors like alcohol consumption and smoking.
The ultimate goal of the researchers is to develop a tool that allows individuals to input their information and receive highly accurate osteoporosis risk scores, empowering them to seek treatment to strengthen their bones and reduce further damage.
Source link
Source link: https://www.bizzbuzz.news/amp/technology/newly-developed-deep-learning-model-shows-promise-in-detecting-osteoporosis-study-1324952