Automated Vertebral Bone Quality Determination from T1-Weighted Lumbar Spine MRI Data Using a Hybrid Convolutional Neural Network–Transformer Neural Network

Vertebral bone quality (VBQ) is a promising new method that can improve screening for osteoporosis. The drawback of the current method is that it requires manual determination of the regions of interest (ROIs) of vertebrae and cerebrospinal fluid (CSF) by a radiologist. In this work, an automatic me...

Full description

Saved in:
Bibliographic Details
Main Authors: Kristian Stojšić, Dina Miletić Rigo, Slaven Jurković
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/14/22/10343
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846154461739745280
author Kristian Stojšić
Dina Miletić Rigo
Slaven Jurković
author_facet Kristian Stojšić
Dina Miletić Rigo
Slaven Jurković
author_sort Kristian Stojšić
collection DOAJ
description Vertebral bone quality (VBQ) is a promising new method that can improve screening for osteoporosis. The drawback of the current method is that it requires manual determination of the regions of interest (ROIs) of vertebrae and cerebrospinal fluid (CSF) by a radiologist. In this work, an automatic method for determining the VBQ is proposed, in which the ROIs are obtained using a trained neural network model. A large, publicly available dataset of sagittal lumbar spine MRI images with ground truth segmentations was used to train a BRAU-Net++ hybrid CNN–transformer neural network. The performance of the trained model was evaluated using the dice similarity coefficient (DSC), accuracy, precision, recall and intersection-over-union (IoU) metrics. The trained model performed similarly to state-of-the-art lumbar spine segmentation models, with an average DSC value of 0.914 ± 0.007 for the vertebrae and 0.902 for the spinal canal. Four different methods of VBQ determination with automatic segmentation are presented and compared with one-way ANOVA. These methods use different algorithms for CSF extraction from the segmentation of the spinal canal using T1- and T2-weighted image data and applying erosion to the vertebral ROI to avoid a sharp change in SI at the edge of the vertebral body.
format Article
id doaj-art-5e99eb2dfd2d40bcb3b2aa1b487090a8
institution Kabale University
issn 2076-3417
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-5e99eb2dfd2d40bcb3b2aa1b487090a82024-11-26T17:48:23ZengMDPI AGApplied Sciences2076-34172024-11-0114221034310.3390/app142210343Automated Vertebral Bone Quality Determination from T1-Weighted Lumbar Spine MRI Data Using a Hybrid Convolutional Neural Network–Transformer Neural NetworkKristian Stojšić0Dina Miletić Rigo1Slaven Jurković2Medical Physics and Radiation Protection Department, University Hospital Rijeka, 51000 Rijeka, CroatiaClinical Department of Diagnostic and Interventional Radiology, University Hospital Rijeka, 51000 Rijeka, CroatiaMedical Physics and Radiation Protection Department, University Hospital Rijeka, 51000 Rijeka, CroatiaVertebral bone quality (VBQ) is a promising new method that can improve screening for osteoporosis. The drawback of the current method is that it requires manual determination of the regions of interest (ROIs) of vertebrae and cerebrospinal fluid (CSF) by a radiologist. In this work, an automatic method for determining the VBQ is proposed, in which the ROIs are obtained using a trained neural network model. A large, publicly available dataset of sagittal lumbar spine MRI images with ground truth segmentations was used to train a BRAU-Net++ hybrid CNN–transformer neural network. The performance of the trained model was evaluated using the dice similarity coefficient (DSC), accuracy, precision, recall and intersection-over-union (IoU) metrics. The trained model performed similarly to state-of-the-art lumbar spine segmentation models, with an average DSC value of 0.914 ± 0.007 for the vertebrae and 0.902 for the spinal canal. Four different methods of VBQ determination with automatic segmentation are presented and compared with one-way ANOVA. These methods use different algorithms for CSF extraction from the segmentation of the spinal canal using T1- and T2-weighted image data and applying erosion to the vertebral ROI to avoid a sharp change in SI at the edge of the vertebral body.https://www.mdpi.com/2076-3417/14/22/10343MRICNNtransformerspine imagingVBQMRI segmentation
spellingShingle Kristian Stojšić
Dina Miletić Rigo
Slaven Jurković
Automated Vertebral Bone Quality Determination from T1-Weighted Lumbar Spine MRI Data Using a Hybrid Convolutional Neural Network–Transformer Neural Network
Applied Sciences
MRI
CNN
transformer
spine imaging
VBQ
MRI segmentation
title Automated Vertebral Bone Quality Determination from T1-Weighted Lumbar Spine MRI Data Using a Hybrid Convolutional Neural Network–Transformer Neural Network
title_full Automated Vertebral Bone Quality Determination from T1-Weighted Lumbar Spine MRI Data Using a Hybrid Convolutional Neural Network–Transformer Neural Network
title_fullStr Automated Vertebral Bone Quality Determination from T1-Weighted Lumbar Spine MRI Data Using a Hybrid Convolutional Neural Network–Transformer Neural Network
title_full_unstemmed Automated Vertebral Bone Quality Determination from T1-Weighted Lumbar Spine MRI Data Using a Hybrid Convolutional Neural Network–Transformer Neural Network
title_short Automated Vertebral Bone Quality Determination from T1-Weighted Lumbar Spine MRI Data Using a Hybrid Convolutional Neural Network–Transformer Neural Network
title_sort automated vertebral bone quality determination from t1 weighted lumbar spine mri data using a hybrid convolutional neural network transformer neural network
topic MRI
CNN
transformer
spine imaging
VBQ
MRI segmentation
url https://www.mdpi.com/2076-3417/14/22/10343
work_keys_str_mv AT kristianstojsic automatedvertebralbonequalitydeterminationfromt1weightedlumbarspinemridatausingahybridconvolutionalneuralnetworktransformerneuralnetwork
AT dinamileticrigo automatedvertebralbonequalitydeterminationfromt1weightedlumbarspinemridatausingahybridconvolutionalneuralnetworktransformerneuralnetwork
AT slavenjurkovic automatedvertebralbonequalitydeterminationfromt1weightedlumbarspinemridatausingahybridconvolutionalneuralnetworktransformerneuralnetwork