Share：
Share this content in WeChat
X
[Chinese] [PDF] 36 1
• Review •
Application progress of artificial intelligence in imaging of vertebral fractures
ZHANG Lei  MENG Linghui 

DOI:10.12015/issn.1674-8034.2026.02.029.


[Abstract] Imaging plays a crucial role in the identification and diagnosis of spinal vertebral fractures (VF), and is particularly vital for the clinical formulation of treatment plans. In recent years, with the rapid development of artificial intelligence (AI) technology, it has shown great potential in the image segmentation, detection and diagnosis of VF. However, existing reviews have predominantly focused on isolated tasks like fracture detection or etiological classification, and have not provided a systematic overview of the overall progress and key challenges in the field. Therefore, this article provides a comprehensive review of the application of AI in the imaging of spinal VF, covering technical methods and research status in fracture segmentation, annotation, detection, and diagnosis. It points out current limitations such as restricted datasets, reliance on single-modal imaging, and insufficient etiological analysis. Furthermore, it prospects the development directions of expanding datasets, integrating multi-modal imaging, and strengthening cross-disciplinary research, in order to promote the transformation of related AI models into stable and reliable clinical auxiliary diagnostic tools. Ultimately, it provides theoretical references for improving the diagnosis and treatment efficiency and accuracy of VF.
[Keywords] vertebral fractures；artificial intelligence；deep learning；magnetic resonance imaging；computed tomography

ZHANG Lei1   MENG Linghui2*  

1 Department of Imaging, Hebei Medical University Third Hospital, Shijiazhuang 050051, China

2 Department of Imaging, Hebei Medical University Second Hospital, Shijiazhuang 050000, China

Corresponding author: MENG L H, E-mail: 37900418@hebmu.edu.cn

Conflicts of interest   None.

Received  2025-11-30
Accepted  2026-01-28
DOI: 10.12015/issn.1674-8034.2026.02.029
DOI:10.12015/issn.1674-8034.2026.02.029.

[1]
BAE Y, KIM M, JEONG W, et al. Epidemiology and future burden of vertebral fractures: insights from the global burden of disease 1990-2021[J/OL]. Healthcare, 2025, 13(15): 1774 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/40805807/. DOI: 10.3390/healthcare13151774.
[2]
LIN S J, CAI X X, CHENG Q, et al. Association between bone turnover markers, BMD and height loss of cemented vertebrae after percutaneous vertebroplasty in patients with osteoporotic vertebral compression fractures[J/OL]. J Orthop Surg Res, 2022, 17(1): 202 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/35379274/. DOI: 10.1186/s13018-022-03087-4.
[3]
DONG Y M, PENG R P, KANG H L, et al. Global incidence, prevalence, and disability of vertebral fractures: a systematic analysis of the global burden of disease study 2019[J]. Spine J, 2022, 22(5): 857-868. DOI: 10.1016/j.spinee.2021.12.007.
[4]
ENACHE A V, TOADER C, ONCIUL R, et al. Surgical stabilization of the spine: a clinical review of spinal fractures, spondylolisthesis, and instrumentation methods[J/OL]. J Clin Med, 2025, 14(4): 1124 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/40004655/. DOI: 10.3390/jcm14041124.
[5]
LEITE G S L, DE OLIVEIRA D O, DE SOUSA PONTES M D, et al. Early vs. late surgery in thoracic spine fractures: impact on clinical outcomes in polytrauma patients[J/OL]. Eur J Trauma Emerg Surg, 2025, 51(1): 334 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/41239017/. DOI: 10.1007/s00068-025-03003-5.
[6]
CASTIGLIONI I, RUNDO L, CODARI M, et al. AI applications to medical images: From machine learning to deep learning[J/OL]. Phys Med, 2021, 83: 9-24 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/33662856/. DOI: 10.1016/j.ejmp.2021.02.006.
[7]
LV Q, ZHANG S Z, WANG Y C. Deep learning model of image classification using machine learning[J/OL]. Adv Multimed, 2022, 2022: 3351256 [2025-11-12]. https://onlinelibrary.wiley.com/doi/10.1155/2022/3351256. DOI: 10.1155/2022/3351256.
[8]
LIAO J, GUO L P, JIANG L, et al. A machine learning-based feature extraction method for image classification using ResNet architecture[J/OL]. Digit Signal Process, 2025, 160: 105036 [2025-11-12]. https://www.sciencedirect.com/science/article/abs/pii/S1051200425000582via%3Dihub. DOI: 10.1016/j.dsp.2025.105036.
[9]
KUTBI M. Artificial intelligence-based applications for bone fracture detection using medical images: a systematic review[J/OL]. Diagnostics, 2024, 14(17): 1879 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/39272664/. DOI: 10.3390/diagnostics14171879.
[10]
MAKI S, FURUYA T, INOUE M, et al. Machine learning and deep learning in spinal injury: a narrative review of algorithms in diagnosis and prognosis[J/OL]. J Clin Med, 2024, 13(3): 705 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/38337399/. DOI: 10.3390/jcm13030705.
[11]
HOLMES J H, SACCHI L, BELLAZZI R, et al. Artificial intelligence in medicine AIME 2015[J/OL]. Artif Intell Med, 2017, 81: 1-2 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/28733119/. DOI: 10.1016/j.artmed.2017.06.011.
[12]
MA J C, YANG H, CHOU Y, et al. Generalizability of lesion detection and segmentation when ScaleNAS is trained on a large multi-organ dataset and validated in the liver[J]. Med Phys, 2025, 52(2): 1005-1018. DOI: 10.1002/mp.17504.
[13]
XIA T Y, ZHAO B, LI B R, et al. MRI-based radiomics and deep learning in biological characteristics and prognosis of hepatocellular carcinoma: opportunities and challenges[J]. J Magn Reson Imaging, 2024, 59(3): 767-783. DOI: 10.1002/jmri.28982.
[14]
GAO Y X, JIANG Y, PENG Y H, et al. Medical image segmentation: a comprehensive review of deep learning-based methods[J/OL]. Tomography, 2025, 11(5): 52 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/40423254/. DOI: 10.3390/tomography11050052.
[15]
REN J S, WANG Y. Overview of object detection algorithms using convolutional neural networks[J]. J Comput Commun, 2022, 10(1): 115-132. DOI: 10.4236/jcc.2022.101006
[16]
WANG J T, RUHAIYEM N I R, FU P P. A comprehensive review of U-Net and its variants: advances and applications in medical image segmentation[J/OL]. IET Image Process, 2025, 19(1): e70019 [2025-11-12]. https://ietresearch.onlinelibrary.wiley.com/doi/10.1049/ipr2.70019. DOI: 10.1049/ipr2.70019.
[17]
ZHANG H M, SBOEV A, RYBKA R, et al. Combining aggregated attention and transformer architecture for accurate and efficient performance of Spiking Neural Networks[J/OL]. Neural Netw, 2025, 191: 107789 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/40633294/. DOI: 10.1016/j.neunet.2025.107789.
[18]
NAGUIB S M, HAMZA H M, HOSNY K M, et al. Classification of cervical spine fracture and dislocation using refined pre-trained deep model and saliency map[J/OL]. Diagnostics, 2023, 13(7): 1273 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/37046491/. DOI: 10.3390/diagnostics13071273.
[19]
CHEN Y, MO Y H, READIE A, et al. VertXNet: an ensemble method for vertebral body segmentation and identification from cervical and lumbar spinal X-rays[J/OL]. Sci Rep, 2024, 14(1): 3341 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/38336974/. DOI: 10.1038/s41598-023-49923-3.
[20]
KIM Y R, YOON Y S, CHA J G. Opportunistic screening for acute vertebral fractures on a routine abdominal or chest computed tomography scans using an automated deep learning model[J/OL]. Diagnostics, 2024, 14(7): 781 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/38611694/. DOI: 10.3390/diagnostics14070781.
[21]
ZHANG H, YUAN G J, ZHANG Z Y, et al. A multi-scene deep learning model for automated segmentation of acute vertebral compression fractures from radiographs: a multicenter cohort study[J/OL]. Insights Imaging, 2024, 15(1): 290 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/39621135/. DOI: 10.1186/s13244-024-01861-y.
[22]
LÖFFLER M T, SEKUBOYINA A, JACOB A, et al. A vertebral segmentation dataset with fracture grading[J/OL]. Radiol Artif Intell, 2020, 2(4): e190138 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/33937831/. DOI: 10.1148/ryai.2020190138.
[23]
PARK T, YOON M A, CHO Y C, et al. Automated segmentation of the fractured vertebrae on CT and its applicability in a radiomics model to predict fracture malignancy[J/OL]. Sci Rep, 2022, 12(1): 6735 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/35468985/. DOI: 10.1038/s41598-022-10807-7.
[24]
BODDEN J, DIECKMEYER M, SOLLMANN N, et al. Incidental vertebral fracture prediction using neuronal network-based automatic spine segmentation and volumetric bone mineral density extraction from routine clinical CT scans[J/OL]. Front Endocrinol, 2023, 14: 1207949 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/37529605/. DOI: 10.3389/fendo.2023.1207949.
[25]
REHMAN F, SHAH S I ALI, RIAZ M N, et al. A region-based deep level set formulation for vertebral bone segmentation of osteoporotic fractures[J]. J Digit Imag, 2020, 33(1): 191-203. DOI: 10.1007/s10278-019-00216-0.
[26]
SEKUBOYINA A, HUSSEINI M E, BAYAT A, et al. VerSe: a Vertebrae labelling and segmentation benchmark for multi-detector CT images[J/OL]. Med Image Anal, 2021, 73: 102166 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/34340104/. DOI: 10.1016/j.media.2021.102166.
[27]
SEKUBOYINA A, REMPFLER M, VALENTINITSCH A, et al. Labeling vertebrae with two-dimensional reformations of multidetector CT images: an adversarial approach for incorporating prior knowledge of spine anatomy[J/OL]. Radiol Artif Intell, 2020, 2(2): e190074 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/33937818/. DOI: 10.1148/ryai.2020190074.
[28]
TIAN J Y, WANG K X, WU P S, et al. Development of a deep learning model for detecting lumbar vertebral fractures on CT images: an external validation[J/OL]. Eur J Radiol, 2024, 180: 111685 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/39197270/. DOI: 10.1016/j.ejrad.2024.111685.
[29]
MOHANTY R, ALLABUN S, SOLANKI S S, et al. NAMSTCD: a novel augmented model for spinal cord segmentation and tumor classification using deep nets[J/OL]. Diagnostics, 2023, 13(8): 1417 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/37189520/. DOI: 10.3390/diagnostics13081417.
[30]
LI Y C, CHEN H H, HORNG-SHING LU H, et al. Can a deep-learning model for the automated detection of vertebral fractures approach the performance level of human subspecialists [J]. Clin Orthop Relat Res, 2021, 479(7): 1598-1612. DOI: 10.1097/CORR.0000000000001685.
[31]
CHEN W J, LIU X, LI K H, et al. A deep-learning model for identifying fresh vertebral compression fractures on digital radiography[J]. Eur Radiol, 2022, 32(3): 1496-1505. DOI: 10.1007/s00330-021-08247-4.
[32]
LIAWRUNGRUEANG W, CHO S T, KOTHEERANURAK V, et al. Osteoporotic vertebral compression fracture (OVCF) detection using artificial neural networks model based on the AO spine-DGOU osteoporotic fracture classification system[J/OL]. N Am Spine Soc J, 2024, 19: 100515 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/39188670/. DOI: 10.1016/j.xnsj.2024.100515.
[33]
OKEREKE I, MMEREM K, BALASUBRAMANIAN D. The management of cervical spine injuries-a literature review[J/OL]. Orthop Res Rev, 2021, 13: 151-162 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/34611449/. DOI: 10.2147/ORR.S324622.
[34]
SINGH M, TRIPATHI U, PATEL K K, et al. An efficient deep learning based approach for automated identification of cervical vertebrae fracture as a clinical support aid[J/OL]. Sci Rep, 2025, 15: 25651 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/40665115/. DOI: 10.1038/s41598-025-10448-6.
[35]
ER H, TÖREN M, ASAN B, et al. Deep learning-based differentiation of vertebral body lesions on magnetic resonance imaging[J/OL]. Diagnostics, 2025, 15(15): 1862 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/40804827/. DOI: 10.3390/diagnostics15151862.
[36]
WINDSOR R, JAMALUDIN A, KADIR T, et al. Automated detection, labelling and radiological grading of clinical spinal MRIs[J/OL]. Sci Rep, 2024, 14(1): 14993 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/38951574/. DOI: 10.1038/s41598-024-64580-w.
[37]
GOMEZ-VAQUERO C, MORA M, GONZÁLEZ-GIMÉNEZ X, et al. Improving fracture risk classification in Spain: a 10-year evaluation of a modified FRAX-based algorithm[J/OL]. Reumatol Clin, 2025, 21(8): 501950 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/41047196/. DOI: 10.1016/j.reumae.2025.501950.
[38]
VAN TOL F R, VERSTEEG A L, VERKOOIJEN H M, et al. Time to surgical treatment for metastatic spinal disease: identification of delay intervals[J]. Glob Spine J, 2023, 13(2): 316-323. DOI: 10.1177/2192568221994787.
[39]
HONG N, CHO S W, SHIN S, et al. Deep-learning-based detection of vertebral fracture and osteoporosis using lateral spine X-ray radiography[J]. J Bone Miner Res, 2023, 38(6): 887-895. DOI: 10.1002/jbmr.4814.
[40]
FOREMAN S C, SCHINZ D, EL HUSSEINI M, et al. Deep learning to differentiate benign and malignant vertebral fractures at multidetector CT[J/OL]. Radiology, 2024, 310(3): e231429 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/38530172/. DOI: 10.1148/radiol.231429.
[41]
LIU K, WANG Q Z, CHEN Y Y, et al. Identification of benign and malignant spinal fractures based on the ResNet50 deep learning model[J]. J Clin Radiol, 2021, 40(12): 2350-2355. DOI: 10.13437/j.cnki.jcr.2021.12.021.
[42]
YEOM J C, PARK S H, KIM Y J, et al. Performance comparison of machine learning using radiomic features and CNN-based deep learning in benign and malignant classification of vertebral compression fractures using CT scans[J/OL]. J Imaging Inform Med, 2025 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/40456998/. DOI: 10.1007/s10278-025-01553-z.
[43]
EWEJE F R, BAO B T, WU J, et al. Deep learning for classification of bone lesions on routine MRI[J/OL]. EBioMedicine, 2021, 68: 103402 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/34098339/. DOI: 10.1016/j.ebiom.2021.103402.
[44]
ZHANG H, YUAN G J, WANG C, et al. Differentiation of benign versus malignant indistinguishable vertebral compression fractures by different machine learning with MRI-based radiomic features[J]. Eur Radiol, 2023, 33(7): 5069-5076. DOI: 10.1007/s00330-023-09678-x.
[45]
LIANG W H, YU H, DUAN L S, et al. MRI-based 2.5D deep learning radiomics nomogram for the differentiation of benign versus malignant vertebral compression fractures[J/OL]. Front Oncol, 2025, 15: 1603672 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/40438697/. DOI: 10.3389/fonc.2025.1603672.
[46]
LIU B B, JIN Y C, FENG S X, et al. Benign vs malignant vertebral compression fractures with MRI: a comparison between automatic deep learning network and radiologist's assessment[J]. Eur Radiol, 2023, 33(7): 5060-5068. DOI: 10.1007/s00330-023-09713-x.
[47]
GENG W, ZHU J F, LI M, et al. Radiomics Based on Multimodal magnetic resonance imaging for the Differential Diagnosis of Benign and Malignant Vertebral Compression Fractures[J]. Orthop Surg, 2024, 16(10): 2464-2474. DOI: 10.1111/os.14148.
[48]
SUN Z Y, LIN M Q, ZHU Q Q, et al. A scoping review on multimodal deep learning in biomedical images and texts[J/OL]. J Biomed Inform, 2023, 146: 104482 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/37652343/. DOI: 10.1016/j.jbi.2023.104482.
[49]
SCHOUTEN D, NICOLETTI G, DILLE B, et al. Navigating the landscape of multimodal AI in medicine: a scoping review on technical challenges and clinical applications[J/OL]. Med Image Anal, 2025, 105: 103621 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/40482561/. DOI: 10.1016/j.media.2025.103621.
[50]
LI Z, GUAN M, SUN D, et al. A novel MRI- and CT-based scoring system to differentiate malignant from osteoporotic vertebral fractures in Chinese patients[J/OL]. BMC Musculoskelet Disord, 2018, 19(1): 406 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/30458738/. DOI: 10.1186/s12891-018-2331-0.
[51]
ZHAO X C, SONG L W, LI C P, et al. Application of deep learning model based on incomplete multimodal medical images in classifying primary bone tumors[J]. J Clin Radiol, 2025, 44(8): 1521-1528. DOI: 10.13437/j.cnki.jcr.2025.08.034.
[52]
ZHENG B, YU P F, ZHU Z Q, et al. Machine learning integration of multi-modal radiomics and clinical factors predicts refracture risk after percutaneous kyphoplasty in postmenopausal women[J/OL]. Sci Rep, 2025, 15(1): 38016 [2025-11-12]. https://pubmed.ncbi.nlm.nih.gov/41168252/. DOI: 10.1038/s41598-025-21894-7.
[53]
HONG N, WHITTIER D E, GLÜER C C, et al. The potential role for artificial intelligence in fracture risk prediction[J]. Lancet Diabetes Endocrinol, 2024, 12(8): 596-600. DOI: 10.1016/S2213-8587(24)00153-0.

PREV Research progress in quantitative susceptibility mapping MRI in spinal diseases
NEXT Research progress on the value of MRI imaging features in the differential diagnosis of hematogenous osteomyelitis and malignant bone tumors of the extremities
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn