Pixel-Level Deep Segmentation: Artificial Intelligence Quantifies Muscle on Computed Tomography for Body Morphometric Analysis

• Type of Material: Article
• Authors: Hyunkwang Lee, Massachusetts General Hospital; Fabian M. Troschel, Massachusetts General Hospital; Shahein Tajmir, Emory University; Georg Fuchs, Charite; Julia Mario, Massachusetts General Hospital; Florian J. Fintelmann, Massachusetts General Hospital; Synho Do, Massachusetts General Hospital
• Language: English
• Date: 2017-08-01
• Publisher: SPRINGER
• Publication Version: Final Published Version
• Copyright Statement: © The Author(s) 2017
• License: Creative Commons Attribution 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.1007/s10278-017-9988-z
• Title of Journal or Parent Work: JOURNAL OF DIGITAL IMAGING
• Volume: 30
• Issue: 4
• Start Page: 487
• End Page: 498
• Keywords: MASS; CONVOLUTIONAL NEURAL-NETWORKS; Muscle segmentation; AUTOMATED SEGMENTATION; Convolutional neural networks; Science & Technology; IMAGES; Life Sciences & Biomedicine; Computed tomography; SKELETAL-MUSCLE; Computer-aided diagnosis (CAD); Radiology, Nuclear Medicine & Medical Imaging; Artificial intelligence; Deep learning; OBESITY; MORTALITY; SHAPE MODEL; BRAIN-TUMOR SEGMENTATION; SARCOPENIA
• Research Categories: Engineering, Biomedical; Health Sciences, Oncology; Artificial Intelligence

Abstract

Pretreatment risk stratification is key for personalized medicine. While many physicians rely on an “eyeball test” to assess whether patients will tolerate major surgery or chemotherapy, “eyeballing” is inherently subjective and difficult to quantify. The concept of morphometric age derived from cross-sectional imaging has been found to correlate well with outcomes such as length of stay, morbidity, and mortality. However, the determination of the morphometric age is time intensive and requires highly trained experts. In this study, we propose a fully automated deep learning system for the segmentation of skeletal muscle cross-sectional area (CSA) on an axial computed tomography image taken at the third lumbar vertebra. We utilized a fully automated deep segmentation model derived from an extended implementation of a fully convolutional network with weight initialization of an ImageNet pre-trained model, followed by post processing to eliminate intramuscular fat for a more accurate analysis. This experiment was conducted by varying window level (WL), window width (WW), and bit resolutions in order to better understand the effects of the parameters on the model performance. Our best model, fine-tuned on 250 training images and ground truth labels, achieves 0.93 ± 0.02 Dice similarity coefficient (DSC) and 3.68 ± 2.29% difference between predicted and ground truth muscle CSA on 150 held-out test cases. Ultimately, the fully automated segmentation system can be embedded into the clinical environment to accelerate the quantification of muscle and expanded to volume analysis of 3D datasets.