Semi-Supervised Deep Learning for Multi-Tissue Segmentation from Multi-Contrast MRI
- PDF / 2,523,941 Bytes
- 14 Pages / 595.224 x 790.955 pts Page_size
- 13 Downloads / 149 Views
Semi-Supervised Deep Learning for Multi-Tissue Segmentation from Multi-Contrast MRI Syed Muhammad Anwar1 · Ismail Irmakci2 · Drew A. Torigian3 · Sachin Jambawalikar4 · Georgios Z. Papadakis5 · Can Akgun6 · Jutta Ellermann7 · Mehmet Akcakaya7 · Ulas Bagci1 Received: 3 March 2020 / Revised: 26 October 2020 / Accepted: 2 November 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Segmentation of thigh tissues (muscle, fat, inter-muscular adipose tissue (IMAT), bone, and bone marrow) from magnetic resonance imaging (MRI) scans is useful for clinical and research investigations in various conditions such as aging, diabetes mellitus, obesity, metabolic syndrome, and their associated comorbidities. Towards a fully automated, robust, and precise quantification of thigh tissues, herein we designed a novel semi-supervised segmentation algorithm based on deep network architectures. Built upon Tiramisu segmentation engine, our proposed deep networks use variational and specially designed targeted dropouts for faster and robust convergence, and utilize multi-contrast MRI scans as input data. In our experiments, we have used 150 scans from 50 distinct subjects from the Baltimore Longitudinal Study of Aging (BLSA). The proposed system made use of both labeled and unlabeled data with high efficacy for training, and outperformed the current state-ofthe-art methods. In particular, dice scores of 97.52%, 94.61%, 80.14%, 95.93%, and 96.83% are achieved for muscle, fat, IMAT, bone, and bone marrow segmentation, respectively. Our results indicate that the proposed system can be useful for clinical research studies where volumetric and distributional tissue quantification is pivotal and labeling is a significant issue. To the best of our knowledge, the proposed system is the first attempt at multi-tissue segmentation using a single end-to-end semi-supervised deep learning framework for multi-contrast thigh MRI scans. Keywords Semi-supervised learning · Tissue segmentation · IMAT
1 Introduction The body composition of tissues changes over time and human muscles tend to lose strength. This could be due to aging or clinical conditions such as diabetes milletus and metabolic syndrome [1]. The muscles and bones Ulas Bagci
[email protected] 1
Center for Research in Computer Vision, University of Central Florida, Orlando, FL, USA
2
Ege University, Izmir, Turkey
3
Department of Radiology, Hospital of the University of Pennsylvania, Philadelphia, PA, USA
4
Columbia University, New York, NY, USA
5
Foundation for Research and Technology Hellas, Crete, Greece
6
FlyWheel Inc., Boston, MA, USA
7
University of Minnesota, Minniapolis, MN, USA
in human body are mechano-responsive tissues, whose strength reduces with time due to the accumulation of adipose (fat) tissue. Fat accumulation in bone occurs in the marrow region. When it occurs within the muscle, it is called fat infiltration [2]. Fat infiltration in muscles acts as an indicator of various clinical outcomes [3], although, inter- and intra-muscular f
Data Loading...
