MRI radiomics-based machine learning classification of atypical cartilaginous tumour and grade II chondrosarcoma of long bones

Abstract

Summary Background Atypical cartilaginous tumour (ACT) and grade II chondrosarcoma (CS2) of long bones are respectively managed with watchful waiting or curettage and wide resection. Preoperatively, imaging diagnosis can be challenging due to interobserver variability and biopsy suffers from sample errors. The aim of this study is to determine diagnostic performance of MRI radiomics-based machine learning in differentiating ACT from CS2 of long bones. Methods One-hundred-fifty-eight patients with surgically treated and histology-proven cartilaginous bone tumours were retrospectively included at two tertiary bone tumour centres. The training cohort consisted of 93 MRI scans from centre 1 (n=74 ACT; n=19 CS2). The external test cohort consisted of 65 MRI scans from centre 2 (n=45 ACT; n=20 CS2). Bidimensional segmentation was performed on T1-weighted MRI. Radiomic features were extracted. After dimensionality reduction and class balancing in centre 1, a machine-learning classifier (Extra Trees Classifier) was tuned on the training cohort using 10-fold cross-validation and tested on the external test cohort. In centre 2, its performance was compared with an experienced musculoskeletal oncology radiologist using McNemar's test. Findings After tuning on the training cohort (AUC=0.88), the machine-learning classifier had 92% accuracy (60/65, AUC=0.94) in identifying the lesions in the external test cohort. Its accuracies in correctly classifying ACT and CS2 were 98% (44/45) and 80% (16/20), respectively. The radiologist had 98% accuracy (64/65) with no difference compared to the classifier (p=0.134). Interpretation Machine learning showed high accuracy in classifying ACT and CS2 of long bones based on MRI radiomic features. Funding ESSR Young Researchers Grant.

Deep learning models for predicting the survival of patients with chondrosarcoma based on a surveillance, epidemiology, and end results analysis

“…Accurate prediction for chondrosarcoma survival is crucial for the counseling, follow-up, and treatment planning of patients. Previous studies have revealed various prognostic factors influencing the survival times of patients with chondrosarcoma, including patient age, tumor size, histological type, tumor grade, and metastasis ( 6 , 19 – 21 ).. At the same time, increasing amounts of imaging ( 22 , 23 ) and genetic data ( 2 , 24 ) are being mined for survival analysis of chondrosarcoma patients. In the face of high-dimensional data, the limitations of the linear relationship between variables assumed by the classical CoxPH model are evident ( 11 ).…”

An Online Calculator for the Prediction of 5-year Survival in Patients with Non-Metastatic Chondrosarcoma Using Machine Learning

“…[6, 7, 24] However, according to our previous study, the application of machine learning techniques in survival and prognosis of primary bone tumor is still rare and mainly developed to improve diagnostic e cacy through imaging perspective. [25][26][27] By analyzing the clinicopathological data of 1267 patients in the SEER database and performing external validation, Huang et al established a nomogram and successfully predicted the 3-year and 5-year survival probability for patients with non-metastatic chondrosarcoma.…”

Radiomics of Musculoskeletal Sarcomas: A Narrative Review

“…The CT radiomics-based machine-learning classifier achieved 75% accuracy overall, 81% accuracy in identifying atypical cartilaginous tumors, and 70% accuracy in identifying higher-grade chondrosarcomas, and still there was no difference in comparison with an experienced radiologist (p = 0.75). Additionally, Gitto et al recently obtained 92% accuracy in differentiating atypical cartilaginous tumor from grade 2 chondrosarcoma of long bones using T1WI MRI radiomics-based machine learning, with no difference compared to an experienced musculoskeletal oncology radiologist (p = 0.134) [63].…”