Partition-based optimization model for generative anatomy modeling language (POM-GAML)

Abstract

BackgroundThis paper presents a novel approach for Generative Anatomy Modeling Language (GAML). This approach automatically detects the geometric partitions in 3D anatomy that in turn speeds up integrated non-linear optimization model in GAML for 3D anatomy modeling with constraints (e.g. joints). This integrated non-linear optimization model requires the exponential execution time. However, our approach effectively computes the solution for non-linear optimization model and reduces computation time from exponential to linear time. This is achieved by grouping the 3D geometric constraints into communities.MethodsVarious community detection algorithms (k-means clustering, Clauset Newman Moore, and Density-Based Spatial Clustering of Applications with Noise) were used to find communities and partition the non-linear optimization problem into sub-problems. GAML was used to create a case study for 3D shoulder model to benchmark our approach with up to 5000 constraints.ResultsOur results show that the computation time was reduced from exponential time to linear time and the error rate between the partitioned and non-partitioned approach decreases with the increasing number of constraints. For the largest constraint set (5000 constraints), speed up was over 2689-fold whereas error was computed as low as 2.2%.ConclusionThis study presents a novel approach to group anatomical constraints in 3D human shoulder model using community detection algorithms. A case study for 3D modeling for shoulder models developed for arthroscopic rotator cuff simulation was presented. Our results significantly reduced the computation time in conjunction with a decrease in error using constrained optimization by linear approximation, non-linear optimization solver.

Kernel Granulometric Texture Analysis and Light RES-ASPP-UNET Classification for Covid-19 Detection

“…The approach called "neuro-evolution" [14,15] is very interesting when it works based on the encoding of various network structures using heuristic genetic algorithm (GA) [16,17]. This optimal framework helps to enhance the network during the training process [18,19].…”

Evolving Deep Learning Convolutional Neural Networks for Early COVID-19 Detection in Chest X-ray Images

“…An alternative approach to train AI models that is currently well-received by the research community consists in solving the aforementioned optimisation problem with heuristic algorithms, often nature-inspired ones as those mentioned in Section 2.3, which have been successfully employed in training support vector machines [28], clustering algorithms [8,29,30], autoencoders [10,31], fuzzy and neural systems [32,33]. An interesting approach referred to as "neuroevolution" is based on encoding different network structures in the form of candidate solutions of a Genetic Algorithm (GA) heuristic [34], thus evolving the optimal architecture of the network while training it.…”

Proceedings of the 2018 MidSouth Computational Biology and Bioinformatics Society (MCBIOS) conference

“…Demriel et al [9] present a novel approach for Generative Anatomy Modeling Language (GAML) to automatically detect geometric partitions in 3-dimensional anatomy. The primary contribution of this approach is to speed-up prior non-linear optimization GAML models by reducing their exponential time complexity to a linear complexity.…”