Abstract
Introduction: The integration of artificial intelligence (AI) into orthopedics has enhanced the diagnosis of various conditions; however, its use in diagnosing soft-tissue tumors remains limited owing to its complexity. This study aimed to develop and assess an AI-driven diagnostic support system for magnetic resonance imaging (MRI)-based soft-tissue tumor diagnosis, potentially improving accuracy and aiding radiologists and orthopedic surgeons. Methods: An experienced orthopedic oncologist and radiologist annotated 720 images from 77 cases (41 benign and 36 malignant soft-tissue tumors). Eleven tumor subtypes were identified and classified into benign and malignant groups based on histological diagnosis. Utilizing the standard machine learning classifier pipeline, we examined and down-selected imaging protocols and their predominant radiomic features within the tumor’s three-dimensional region to differentiate between benign and malignant tumors. Among the scan protocols, contrast-enhanced T1-weighted fat-suppressed images showed the most accurate classification based on radiomic features. We focused on the two-dimensional features from the largest tumor boundary surface and its neighboring slices, leveraging texture-based radiomic and deep convolutional neural network features from a pretrained VGG19 model. Results: The test data comprised 44 contrast-enhanced images (22 benign and 22 malignant soft-tissue tumors) containing six malignant and five benign subtypes distinct from the training data. We compared expert and nonexpert human performances against AI by assessing malignancy detection and the time required for classification. The AI model showed comparable accuracy (AUC 0.91) to that of radiologists (AUC 0.83) and orthopedic surgeons (AUC 0.73). Notably, the AI model processed data approximately 400 times faster than its human counterparts, showcasing its capacity to significantly boost diagnostic efficiency. Conclusion: We developed an AI-driven diagnostic support system for MRI-based soft-tissue tumor diagnosis. While additional refinement is necessary for clinical applications, our system has exhibited promising potential in differentiating between benign and malignant soft-tissue tumors based on MRI.
Plain Language Summary
MRI is invaluable for the qualitative diagnosis of soft-tissue tumors, but radiologists face challenges despite advancements in imaging technologies. This study aimed to develop and evaluate an AI-driven diagnostic support system for MRI-based soft-tissue tumor diagnosis. The study used 720 images from 77 cases. Using a standard machine learning classifier pipeline, we examined and down-selected imaging protocols and their predominant radiomic features within each tumor’s three-dimensional region to differentiate between benign and malignant tumors. Contrast-enhanced and T2-weighted images were identified as candidates for developing a machine learning model for malignancy classification. We focused on the two-dimensional features from the largest tumor boundary surface and its neighboring slices, leveraging texture-based radiomic and deep convolutional neural network features from a pretrained VGG19 model. A support vector machine classifier trained with contrast-enhanced images demonstrated the best performance in preliminary validation investigations. Validation was performed using test data comprising 44 contrast-enhanced images. The AI model exhibited comparable accuracy (AUC 0.91) to that of radiologists (AUC 0.83) and orthopedic surgeons (AUC 0.73). The AI model processed data approximately 400 times faster than its human counterparts, demonstrating its capacity to significantly enhance diagnostic efficiency. While additional refinement is necessary for clinical applications, our system has exhibited promising potential in differentiating between benign and malignant soft-tissue tumors based on MR images.
