Artificial intelligence-deep learning: The objective of this work package is to employ artificial intelligence-deep learning (AI-DL) approaches to revolutionize current clinical approaches to bone disease. In recent years, artificial intelligence, particularly deep learning (DL) and convolutional neural networks (CNN), has made significant strides in the medical field. CNNs have demonstrated the ability to identify complex patterns in high-dimensional data and integrate information from multiple sources through numerous input channels into neural networks.

Human genetics stands to benefit greatly from DL, ushering in a new era of possibilities in Big Data analysis. Our AI-DL work has demonstrated the potential for unsupervised fracture prediction using a large set of “substandard” DXA images. We will utilize PGMRA as the initial AI-DL approach to integrate genetic profiles, imaging data from EOS, microindentation data from Osteoprobe, and clinical risk scores.