Feature-aware Unsupervised Lesion Segmentation For Brain Tumor Images Using Fast Data Density Functional Transform

This article presents a novel approach to unsupervised recognition and segmentation of lesion images, using the framework of fast data density functional transform (fDDFT). By introducing the architecture of geometric deep learning and metrics of graph neural networks, gridized density functionals of the fDDFT establish an unsupervised feature-aware mechanism with global convolutional kernels to extract the most likely lesion boundaries and produce lesion segmentation. An AutoEncoder-assisted module reduces the computational complexity and inference time of each object in large three-dimensional datasets is 1.76 s on average. The experiment shows that the synergy of fDDFT and a naïve neural network improves the training and inference time by 58% and 51%, respectively, and the dice score raises to 0.9415.