Radiomic Machine Learning Classifiers in Spine Bone Tumors: A Multi-Software, Multi-Scanner Study

Abstract

Spinal lesion differential diagnosis remains challenging even in MRI. Radiomics and machine learning (ML) have proven useful even in absence of a standardized data mining pipeline. We aimed to assess ML diagnostic performance in spinal lesion differential diagnosis, employing radiomic data extracted by different software. Methods: Patients undergoing MRI for a vertebral lesion were retrospectively analyzed (n = 146, 67 males, 79 females; mean age 63 ± 16 years, range 8-89 years) and constituted the train (n = 100) and internal test cohorts (n = 46). Part of the latter had additional prior exams which constituted a multi-scanner, external test cohort (n = 35). Lesions were labeled as benign or malignant (2-label classification), and benign, primary malignant or metastases (3-label classification) for classification analyses. Features extracted via 3D Slicer heterogeneityCAD module (hCAD) and PyRadiomics were independently used to compare different combinations of feature selection methods and ML classifiers (n = 19). Results: In total, 90 and 1548 features were extracted by hCAD and PyRadiomics, respectively. The best feature selection method-ML algorithm combination was selected by 10 iterations of 10-fold cross-validation in the training data. For the 2-label classification ML obtained 94% accuracy in the internal test cohort, using hCAD data, and 86% in the external one. For the 3-label classification, PyRadiomics data allowed for 80% and 69% accuracy in the internal and external test sets, respectively. Conclusions: MRI radiomics combined with ML may be useful in spinal lesion assessment. More robust preprocessing led to better consistency despite scanner and protocol heterogeneity.

Sarcopenia: imaging assessment and clinical application

“…This is a group of different subfields but the most important areas in radiology are machine learning and deep learning [ 94 ]. With machine learning different algorithms are trained to recognize specific characteristics by learning patterns from datasets [ 95 , 96 ].…”

CT and MRI radiomics of bone and soft-tissue sarcomas: a systematic review of reproducibility and validation strategies

“…Proper validation of radiomic models is highly desirable to bridge the gap between concepts and clinical application [ 53 ]. Machine learning validation techniques are employed to avoid any information leak from the test to the training set during model development [ 60 ]. Resampling strategies can be extremely useful, especially with relatively limited samples of data, which may not be truly representative for the population of interest, with the aim of reducing overfitting and better estimating the performance of the radiomics-based predictive model on new data (i.e., the test set) [ 61 , 62 ].…”

CT radiomics-based machine learning classification of atypical cartilaginous tumours and appendicular chondrosarcomas

“…First, our study is retrospective, as this design allowed including relatively large numbers of patients with an uncommon disease, such as chondrosarcoma, and imaging data already available. Additionally, a prospective analysis is not strictly needed for radiomic studies [13]. Second, we performed bidimensional segmentation and chose the image showing the maximum lesion extension.…”

“…Due to its high-dimensional nature consisting of numerous radiomic features, radiomics benefits from powerful analytic tools and artificial intelligence with machine learning perfectly addresses this issue [12] . Machine learning algorithms can be trained using subsets of radiomic features creating classification models for the diagnosis of interest [13] , [14] , [15] .…”