1150 The Use of Artificial Intelligence in the Detection of Chondrosarcomas: A Scoping Review

Abstract

Abstract Aim Chondrosarcoma is a challenging primary bone tumour to diagnose radiographically. This review investigates the role that artificial intelligence (AI) may have in improving the diagnostic accuracy of chondrosarcoma. Method A review of public databases, including PubMed, Scopus, and IEEE Xplore was performed. Predefined search terms, including but not limited to "chondrosarcoma," “bone tumour” "artificial intelligence," and "diagnosis." were used. The inclusion criteria were research articles and reviews focusing on the application of AI in the detection of chondrosarcoma as applied to various imaging modalities including X-ray, CT scans, and Magnetic Resonance Imaging. Details on the types of algorithms used, dataset characteristics, performance metrics, and any comparisons to traditional diagnostic methods were captured. Results The review highlighted that there is a growing body of literature and increasing potential for AI in the identification of chondrosarcomas across different imaging modalities. AI algorithms, including machine learning and deep learning models, showed promising results with improved sensitivity and specificity compared to conventional diagnostic methods. Conclusions The Integration of AI in chondrosarcoma detection holds promise for enhancing diagnostic accuracy. Standardised methodologies, larger datasets, and prospective validation studies are needed for robust application. Collaborative efforts between clinicians, radiologists, and AI developers are crucial for translating these advancements into clinically meaningful tools, ultimately improving early detection and management of chondrosarcoma.

Ähnliche Arbeiten