Search

Simple Summary: One of the main applications of in vivo magnetic resonance spectroscopy (MRS) is in the non-invasive monitoring of the metabolic pattern of brain tumors. MRS comes in two basic modalities, single-voxel (SV), from which the signal is obtained, and multivoxel (MV), in which one or more contiguous grids of SVs are acquired. The purpose of our proof-of-concept study was to test whether it would be possible to train machine learning models using SV data at 1.5T, and test them with MV 3T data from independent patients, obtaining color-coded images of pathology (nosological images) to help radiologists in their preoperative evaluation of patients. With sequential forward feature selection followed by linear discriminant analysis, we obtained AUCs = 0.95 (meningioma), 0.89 (aggressive), 0.82 (low-grade glioma), and 0.82 (normal brain) in the MV test set. In vivo magnetic resonance spectroscopy (MRS) has two modalities, single-voxel (SV) and multivoxel (MV), in which one or more contiguous grids of SVs are acquired. Purpose: To test whether MV grids can be classified with models trained with SV. Methods: Retrospective study. Training dataset: Multicenter multiformat SV INTERPRET, 1.5T. Testing dataset: MV eTumour, 3T. Two classification tasks were completed: 3-class (meningioma vs. aggressive vs. normal) and 4-class (meningioma vs. low-grade glioma vs. aggressive vs. normal). Five different methods were tested for feature selection. The classification was implemented using linear discriminant analysis (LDA), random forest, and support vector machines. The evaluation was completed with balanced error rate (BER) and area under the curve (AUC) on both sets. The accuracy in class prediction was calculated by developing a solid tumor index (STI) and segmentation accuracy with the Dice score. Results: The best method was sequential forward feature selection combined with LDA, with AUCs = 0.95 (meningioma), 0.89 (aggressive), 0.82 (low-grade glioma), and 0.82 (normal). STI was 66% (4-class task) and 71% (3-class task) because two cases failed completely and two more had suboptimal STI as defined by us. Discussion: The reasons for failure in the classification of the MV test set were related to the presence of artifacts. [ABSTRACT FROM AUTHOR]