Fully automatic classification of lumbar disc degeneration based on deep learning
Journal Title: Chinese Journal of Clinical Research - Year 2024, Vol 37, Issue 5
Abstract
<b>Objective</b> To investigate the feasibility of a deep learning model for the fully automatic classification of disc degeneration based on lumbar structures on sagittal T2WI images. <b>Methods</b> The lumbar T2WI image data of 94 patients who underwent lumbar spine MRI examination in the Third Affiliated Hospital of Anhui Medical University from August 2020 to June 2022 were retrospectively selected，and 466 discs were obtained. The lumbar intervertebral disc were manually annotated by 2 radiologists on sagittal T2WI images．The data were randomly divided into train set (n=300)，validation set (n=72)，and test set (n=94). Firstly, a U-Net network was used to train the disc segmentation model. The evaluation indexes of the model included Dice coefficient and IoU score. Then, SpineNet network was used to train the classification model, and the evaluation indexes of the model included accuracy, sensitivity, specificity, F1 score, and ROC curves. <b>Results</b> In the test set, the dice coefficient and IoU values of U-Net model for lumbar disc segmentation were 0.920 and 0.853, respectively. The accuracy, specificity and sensitivity value of SpineNet classification models for lumbar disc degeneration were 0.913, 0.912 and 0.916, respectively. The ROC curve analysis showed that the AUC values for distinguishing mild to moderate, mild to serious, and moderate to serious lumbar disc degeneration were 0.89, 0.95, and 0.90, respectively. <b>Conclusion</b> It is feasible to realize the fully automatic classification of disc degeneration based on deep learning network.
Authors and Affiliations
DING Zhaoming*, LI Hongyan, CHEN Liang, CHEN Bing, SUN Hui, HOU Wentao, XIA Chunhua
Keywords
Related Articles
- EP ID EP736332
- DOI 10.13429/j.cnki.cjcr.2024.05.039
- Views 44
- Downloads 0
Correlation of Hsa circ0016599 expression level with the efficacy of neoadjuvant chemotherapy in patients with triple-negative breast cancer
"Objective To analyze the expression of circular ribonucleic acid (circRNA) Hsa circ0016599 in patients with triple-negative breast cancer (TNBC) and its effect and predictive value for the efficacy of neoadjuvant chemot...
Summary of best evidence for rehabilitation timing in patients undergoing arthroscopic rotator cuff repair surgery
"<b>Objective</b> To search，organize,and summarize the best evidence of rehabilitation timing for patients after arthroscopic rotator cuff repair surgery, providing reference for clinical practice. <b>Methods</b> Da...
Establishment of a predictive model for somatosensory evoked potentials during decompression laminectomy
<b>Objective</b> To explore the influencing factors affecting somatosensory evoked potential (SEP) during decompression laminectomy and to establish a prediction model. b>Methods</b> The SEP monitoring data from 120 pa...
Predictive value of serum actinin-4, TFF1, and TGFBI for the prognosis of patients with primary hepatocellular carcinoma after transarterial chemoembolization
Objective To investigate the predictive value of serum actinin-4, trefoil factor 1 (TFF1), and transforming growth factor-β-induced protein (TGFBI) for the prognosis of patients with primary hepatocellular carcinoma (PHC...
Efficacy of bronchoscopic ablation combined with local injection of endostatin in advanced non-small cell lung cancer patients with intratracheal metastasis
Objective To explore the efficacy and safety of bronchoscopic ablation combined with local injection of endostatin in patients with intratracheal metastasis of advanced non-small cell lung cancer(NSCLC). Methods A retr...