Automatic Vertebral Body Segmentation Based on Deep Learning of Dixon Images for Bone Marrow Fat Fraction Quantification

Abstract

Background: Bone marrow fat (BMF) fraction quantification in vertebral bodies is used as a novel imaging biomarker to assess and characterize chronic lower back pain. However, manual segmentation of vertebral bodies is time consuming and laborious. Purpose: ( 1 ) Develop a deep learning pipeline for segmentation of vertebral bodies using quantitative water-fat MRI. ( 2 ) Compare BMF measurements between manual and automatic segmentation methods to assess performance. Materials and Methods: In this retrospective study, MR images using a 3D spoiled gradient-recalled echo (SPGR) sequence with Iterative Decomposition of water and fat with Echo Asymmetry and Least-squares estimation (IDEAL) reconstruction algorithm were obtained in 57 subjects (28 women, 29 men, mean age, 47.2 ± 12.6 years). An artificial network was trained for 100 epochs on a total of 165 lumbar vertebrae manually segmented from 31 subjects. Performance was assessed by analyzing the receiver operating characteristic curve, precision-recall, F1 scores, specificity, sensitivity, and similarity metrics. Bland-Altman analysis was used to assess performance of BMF fraction quantification using the predicted segmentations. Results: The deep learning segmentation method achieved an AUC of 0.92 (CI 95%: 0.9186, 0.9195) on a testing dataset ( n = 24 subjects) on classification of pixels as vertebrae. A sensitivity of 0.99 and specificity of 0.80 were achieved for a testing dataset, and a mean Dice similarity coefficient of 0.849 ± 0.091. Comparing manual and automatic segmentations on fat fraction maps of lumbar vertebrae ( n = 124 vertebral bodies) using Bland-Altman analysis resulted in a bias of only −0.605% (CI 95% = −0.847 to −0.363%) and agreement limits of −3.275% and +2.065%. Automatic segmentation was also feasible in 16 ± 1 s. Conclusion: Our results have demonstrated the feasibility of automated segmentation of vertebral bodies using deep learning models on water-fat MR (Dixon) images to define vertebral regions of interest with high specificity. These regions of interest can then be used to quantify BMF with comparable results as manual segmentation, providing a framework for completely automated investigation of vertebral changes in CLBP.

Fully automated radiomic screening pipeline for osteoporosis and abnormal bone density with a deep learning-based segmentation using a short lumbar mDixon sequence

“…Another similar results was from the study of automatic vertebral body segmentation that based on deep learning of Dixon images with an AUC of 0.92 and a mean Dice coefficient of 0.849±0.091 (10). These parameter values suggest that solely using BMFF average and demographic characteristics is not as effective as using radiomic models.…”

“…Rastegar et al (9) developed classification models to predict osteoporosis and osteopenia using radiomics on bone mineral densitometry images, achieving arear under the receiver operator characteristic curve (AUC) values from 0.50 to 0.78. Zhou et al found that deep learning can provide automated segmentation of vertebral bodies using water-fat MR (Dixon) images to qualify bone marrow fat recently (10). However, there is no study to for ABD and osteoporosis prediction, respectively.…”

Artificial Intelligence and Computer Vision in Low Back Pain: A Systematic Review

“…However, recent research has been exploring the use of deep learning-based AI systems which are able to perform multiple tasks at the basic and advanced level in a single model [1]. Vertebrae are by far the most investigated structure, with AI systems reaching > 90% DICE and > 90% accuracy in the majority of studies included in our review, both using DIP [28][29][30][31][32][33][34][35][36][37][38][39][40][41]43,44,[48][49][50]53,61,62] and deep learning models [67,69,77,[80][81][82][83][84]. In particular, a study from Lee et al [40] proposed a model to obtain an automated segmentation of lumbar pedicles from CT images in order to increase accuracy and safety during transpedicular screw placement.…”

“…Moreover, they used a deformed U-net [76] for the segmentation of paraspinal muscles on 120 MRI achieving an overall DICE greater than 91.3%. Zhou et al [77] utilized a U-net for vertebrae segmentation on MRI of 57 subjects, achieving a DICE of 84.9%.…”

The contributions of cartilage endplate composition and vertebral bone marrow fat to intervertebral disc degeneration in patients with chronic low back pain

“…One strength of our study is the deployment of two recently developed neural networks [15,16] to perform automated tissue segmentation. This enabled rapid and unbiased evaluation of the MRI biomarkers, demonstrating the relevance of neural networks to a diverse class of problems and addressing the inefficiency and subjectivity associated with manual tissue segmentation.…”