Speaker:Mohammad Hashemi

Deep-Learning-Based Prediction of Bone Degradation Utilizing Novel VirtualDataset of Cellular Microstructures

Degradation of bone, especially for astronauts exposed to microgravity conditions, is crucial for space exploration missions since the lower applied external forces accelerate the diminution in bone stiffness and strength substantially. Even though existing computational models and simulations help us understand this phenomenon and possibly restrict its effect in the future, they are time-consuming to simulate the changes in the bones, not just the bone microstructures, of each individual in detail. In this study, a robust yet fast computational method to predict and visualize bone degradation has been developed. The deep-learning part of it, TransVNet, can take in different 3D voxelized images and predict their evolution throughout months utilizing a hybrid 3D-CNN-VisionTransformer autoencoder architecture. Because of the limited available experimental data and the challenges of obtaining new samples, a digital twin dataset of diverse and initial bone-like microstructures was generated to train our TransVNet on the evolution of the 3D images through our previously developed degradation model for microgravity. The preliminary results show its satisfactory and high performance on the bone degradation dataset. Nevertheless, the presented AI framework can be used and customized for accelerating and generalizing more sophisticated degradation models or similar 3D image sequence prediction tasks.

Committee: Ali Jannesari (major professor), Hongyang Gao, and Azadeh Sheidaei

Join on Zoom: https://iastate.zoom.us/j/96653396958