Predicting pediatric spinal deformity (PSD) from X-ray images collected on the patient's initial visit is a challenging task. This work builds on our previous method and provides a novel bio-informed framework based on a mechanistic machine learning technique with dynamic patient-specific parameters to predict PSD. We provide a geometry-based bone growth model that can be utilized in a range of applications to enhance the bio-informed mechanistic machine learning framework. The proposed technique is utilized to examine and predict spine curvature in PSD cases such as adolescent idiopathic scoliosis. The best fit of a segmented 3D volumetric geometry of the human spine acquired from 2D X-ray images is employed. Using an active contour model based on gradient vector flow snakes, the anteroposterior and lateral views of the X-ray images are segmented to derive the 2D contours surrounding each vertebra. Using minimal user input, the snake parameters are calibrated and automatically computed over the dataset, resulting in fast image segmentation and data collection. The 2D segmented outlines of each vertebra are transformed into a 3D image segmentation result. The Iterative Closest Point mesh registration technique is then used to establish a mesh morphing approach and creates a 3D atlas spine model. Using the comprehensive 3D volumetric model, one can automatically extract spinal geometry data as inputs to the mechanistic machine learning network. Moreover, the proposed bio-informed deep learning network with the modified bone growth model achieves competitive or even superior performance against other state-of-the-art learning-based methods.

The rapid growth of online social media usage in our daily lives has increased the importance of analyzing the dynamics of online social networks. However, the dynamic data of existing online social media platforms are not readily accessible. Hence, there is a necessity to synthesize networks emulating those of online social media for further study. In this work, we propose an epidemiology-inspired and community-based, time-evolving online social network generation algorithm (EpiCNet), to generate a time-evolving sequence of random networks that closely mirror the characteristics of real-world online social networks. Variants of the algorithm can produce both undirected and directed networks to accommodate different user interaction paradigms. EpiCNet utilizes compartmental models inspired by mathematical epidemiology to simulate the flow of individuals into and out of the online social network. It also employs an overlapping community structure to enable more realistic connections between individuals in the network. Furthermore, EpiCNet evolves the community structure and connections in the simulated online social network as a function of time and with an emphasis on the behavior of individuals. EpiCNet is capable of simulating a variety of online social networks by adjusting a set of tunable parameters that specify the individual behavior and the evolution of communities over time. The experimental results show that the network properties of the synthetic time-evolving online social network generated by EpiCNet, such as clustering coefficient, node degree, and diameter, match those of typical real-world online social networks such as Facebook and Twitter.