Deep negative volume segmentation

Abstract

Clinical examination of three-dimensional image data of compound anatomical objects, such as complex joints, remains a tedious process, demanding the time and the expertise of physicians. For instance, automation of the segmentation task of the TMJ (temporomandibular joint) has been hindered by its compound three-dimensional shape, multiple overlaid textures, an abundance of surrounding irregularities in the skull, and a virtually omnidirectional range of the jaw’s motion—all of which extend the manual annotation process to more than an hour per patient. To address the challenge, we invent a new workflow for the 3D segmentation task: namely, we propose to segment empty spaces between all the tissues surrounding the object—the so-called negative volume segmentation. Our approach is an end-to-end pipeline that comprises a V-Net for bone segmentation, a 3D volume construction by inflation of the reconstructed bone head in all directions along the normal vector to its mesh faces. Eventually confined within the skull bones, the inflated surface occupies the entire “negative” space in the joint, effectively providing a geometrical/topological metric of the joint’s health. We validate the idea on the CT scans in a 50-patient dataset, annotated by experts in maxillofacial medicine, quantitatively compare the asymmetry given the left and the right negative volumes, and automate the entire framework for clinical adoption.

morphoHeart: a novel quantitative tool to perform integrated 3D morphometric analyses of heart and ECM morphology during embryonic development

“…In addition to the comprehensive morphometric quantifications performed by morphoHeart , it also provides the first negative space segmentation of the cardiac ECM. A previous study used automated constrained mesh inflation and subtraction to define the negative space surrounding joints 77 ; however, our contour-directed negative space segmentation allows a more detailed analysis of regional expansions of the segmented matrix. While transgenic lines providing fluorescent readouts of Hyaluronic Acid have been described 6,41 which could be used for analysis of the cardiac ECM in morphoHeart , these lines are not specific to the heart, making clean segmentation challenging.…”

Medical Image Captioning via Generative Pretrained Transformers

“…Radiology is the medical discipline that uses medical imaging to diagnose and treat diseases. Today, radiology actively implements new artificial intelligence approaches [8][9][10] . There are three types of radiologists -diagnostic radiologists, interventional radiologists and radiation oncologists.…”

Generation of microbial colonies dataset with deep learning style transfer

“…DL-based automatic extraction of information from images has led to unprecedented progress in many fields. Neural networks were applied in numerous medical tasks, most extensively in radiology and pathology specialties 1 – 8 . Moreover, their performance is frequently comparable, or even better than in the case of human experts.…”