Preeti
Osteoporosis, known as the “silent disease,” is difficult to detect in its early stages. Researchers at Tulane University have developed a new deep learning algorithm that outperformed existing methods for predicting osteoporosis risk. The deep neural network (DNN) model was tested against conventional machine learning algorithms and a regression model using data from over 8,000 participants in the Louisiana Osteoporosis Study. The DNN model showed the best predictive performance in identifying osteoporosis risk factors.
Lead author Chuan Qiu emphasized the importance of early detection for preventative measures and better outcomes for patients. The researchers identified the top 10 factors for predicting osteoporosis risk, including weight, age, gender, grip strength, and lifestyle factors like alcohol consumption and smoking. The simplified DNN model using these factors performed almost as well as the full model.
While there is still work to be done before implementing an AI platform for predicting osteoporosis risk, the researchers see the potential benefits of using deep learning models in healthcare. The ultimate goal is to provide individuals with accurate risk scores to empower them to seek treatment and strengthen their bones. This study represents a step towards utilizing AI for personalized healthcare interventions.
Source link
Source link: https://www.htworld.co.uk/news/ai/deep-learning-model-shows-promise-for-diagnosing-osteoporosis/?amp=1
GIPHY App Key not set. Please check settings