Integrating Manual Annotation with Deep Transfer Learning and Radiomics for Vertebral Fracture Analysis

Overview

Journal BMC Med Imaging

Publisher Biomed Central

Date 2025 Feb 6

PMID 39915711

Authors

Jing Wang

Zhirui Dong

Huanxin He

Zhiyang Gao

Yukai Huang

Guangcheng Yuan

Libo Jiang

Mingdong Zhao

Affiliations

Soon will be listed here.

Abstract

Background: Vertebral compression fractures (VCFs) are prevalent in the elderly, often caused by osteoporosis or trauma. Differentiating acute from chronic VCFs is vital for treatment planning, but MRI, the gold standard, is inaccessible for some. However, CT, a more accessible alternative, lacks precision. This study aimed to enhance CT's diagnostic accuracy for VCFs using deep transfer learning (DTL) and radiomics.

Methods: We retrospectively analyzed 218 VCF patients scanned with CT and MRI within 3 days from Oct 2022 to Feb 2024. MRI categorized VCFs. 3D regions of interest (ROIs) from CT scans underwent feature extraction and DTL modeling. Receiver operating characteristic (ROC) analysis evaluated models, with the best fused with radiomic features via LASSO. AUCs compared via Delong test, and clinical utility assessed by decision curve analysis (DCA).

Results: Patients were split into training (175) and test (43) sets. Traditional radiomics with LR yielded AUCs of 0.973 (training) and 0.869 (test). Optimal DTL modeling improved to 0.992 (training) and 0.941 (test). Feature fusion further boosted AUCs to 1.000 (training) and 0.964 (test). DCA validated its clinical significance.

Conclusion: The feature fusion model enhances the differential diagnosis of acute and chronic VCFs, outperforming single-model approaches and offering a valuable decision-support tool for patients unable to undergo spinal MRI.

References

Duan S, Hua Y, Cao G, Hu J, Cui W, Zhang D . Differential diagnosis of benign and malignant vertebral compression fractures: Comparison and correlation of radiomics and deep learning frameworks based on spinal CT and clinical characteristics. Eur J Radiol. 2023; 165:110899. DOI: 10.1016/j.ejrad.2023.110899. View

Al Zorgani M, Ugail H, Pors K, Dauda A . Deep Transfer Learning-Based Approach for Glucose Transporter-1 (GLUT1) Expression Assessment. J Digit Imaging. 2023; 36(6):2367-2381. PMC: 10584776. DOI: 10.1007/s10278-023-00859-0. View

Abramson Z, Sheyn A, Goode C, Choudhri A . Perimandibular "Fat Stranding" Sign: A Diagnostic Aid for Subtle Mandibular Fractures. AJR Am J Roentgenol. 2021; 218(5):917-918. DOI: 10.2214/AJR.21.27141. View

Zhang J, Liu J, Liang Z, Xia L, Zhang W, Xing Y . Differentiation of acute and chronic vertebral compression fractures using conventional CT based on deep transfer learning features and hand-crafted radiomics features. BMC Musculoskelet Disord. 2023; 24(1):165. PMC: 9987077. DOI: 10.1186/s12891-023-06281-5. View

Wang X, Xie T, Luo J, Zhou Z, Yu X, Guo X . Radiomics predicts the prognosis of patients with locally advanced breast cancer by reflecting the heterogeneity of tumor cells and the tumor microenvironment. Breast Cancer Res. 2022; 24(1):20. PMC: 8922933. DOI: 10.1186/s13058-022-01516-0. View

Frellesen C, Azadegan M, Martin S, Otani K, DAngelo T, Booz C . Dual-Energy Computed Tomography-Based Display of Bone Marrow Edema in Incidental Vertebral Compression Fractures: Diagnostic Accuracy and Characterization in Oncological Patients Undergoing Routine Staging Computed Tomography. Invest Radiol. 2018; 53(7):409-416. DOI: 10.1097/RLI.0000000000000458. View

Yaqoob S, Cafiso S, Morabito G, Pappalardo G . Deep transfer learning-based anomaly detection for cycling safety. J Safety Res. 2023; 87:122-131. DOI: 10.1016/j.jsr.2023.09.010. View

Carli D, Venmans A, Lodder P, Donga E, van Oudheusden T, Boukrab I . Vertebroplasty versus Active Control Intervention for Chronic Osteoporotic Vertebral Compression Fractures: The VERTOS V Randomized Controlled Trial. Radiology. 2023; 308(1):e222535. DOI: 10.1148/radiol.222535. View

Khan M, Jennings J, Baker J, Smolock A, Shah L, Pinchot J . ACR Appropriateness Criteria® Management of Vertebral Compression Fractures: 2022 Update. J Am Coll Radiol. 2023; 20(5S):S102-S124. DOI: 10.1016/j.jacr.2023.02.015. View

10.

Jiang Y, Zhou K, Sun Z, Wang H, Xie J, Zhang T . Non-invasive tumor microenvironment evaluation and treatment response prediction in gastric cancer using deep learning radiomics. Cell Rep Med. 2023; 4(8):101146. PMC: 10439253. DOI: 10.1016/j.xcrm.2023.101146. View

11.

Parreira P, Maher C, Megale R, March L, Ferreira M . An overview of clinical guidelines for the management of vertebral compression fracture: a systematic review. Spine J. 2017; 17(12):1932-1938. DOI: 10.1016/j.spinee.2017.07.174. View

12.

Wang C, Tsai J, Chuang M, Wang M, Huang K, Lin R . Bone marrow edema in vertebral compression fractures: detection with dual-energy CT. Radiology. 2013; 269(2):525-33. DOI: 10.1148/radiology.13122577. View

13.

Hedgire S, Baliyan V, Zucker E, Bittner D, Staziaki P, Takx R . Perivascular Epicardial Fat Stranding at Coronary CT Angiography: A Marker of Acute Plaque Rupture and Spontaneous Coronary Artery Dissection. Radiology. 2018; 287(3):808-815. PMC: 8382040. DOI: 10.1148/radiol.2017171568. View

14.

Lin H, Tsai H, Chao C, Lin S . Periappendiceal fat-stranding models for discriminating between complicated and uncomplicated acute appendicitis: a diagnostic and validation study. World J Emerg Surg. 2021; 16(1):52. PMC: 8511616. DOI: 10.1186/s13017-021-00398-5. View

15.

Zhang J, Xia L, Tang J, Xia J, Liu Y, Zhang W . Constructing a Deep Learning Radiomics Model Based on X-ray Images and Clinical Data for Predicting and Distinguishing Acute and Chronic Osteoporotic Vertebral Fractures: A Multicenter Study. Acad Radiol. 2023; 31(5):2011-2026. DOI: 10.1016/j.acra.2023.10.061. View

16.

Kim A, Yoon M, Ham S, Cho Y, Ko Y, Park B . Prediction of the Acuity of Vertebral Compression Fractures on CT Using Radiologic and Radiomic Features. Acad Radiol. 2022; 29(10):1512-1520. DOI: 10.1016/j.acra.2021.12.008. View

17.

Fritz B, Yi P, Kijowski R, Fritz J . Radiomics and Deep Learning for Disease Detection in Musculoskeletal Radiology: An Overview of Novel MRI- and CT-Based Approaches. Invest Radiol. 2022; 58(1):3-13. DOI: 10.1097/RLI.0000000000000907. View

18.

Ding J, Chen S, Sosa M, Cattell R, Lei L, Sun J . Optimizing the Peritumoral Region Size in Radiomics Analysis for Sentinel Lymph Node Status Prediction in Breast Cancer. Acad Radiol. 2020; 29 Suppl 1:S223-S228. PMC: 9583077. DOI: 10.1016/j.acra.2020.10.015. View

19.

Court-Brown C, McQueen M . Global Forum: Fractures in the Elderly. J Bone Joint Surg Am. 2016; 98(9):e36. DOI: 10.2106/JBJS.15.00793. View

20.

Han Y, Wang Z, Chen A, Ali I, Cai J, Ye S . A deep transfer learning-based protocol accelerates full quantum mechanics calculation of protein. Brief Bioinform. 2022; 24(1). DOI: 10.1093/bib/bbac532. View