• Peter Savadjiev, Caroline Reinhold, Diego Martin, and Reza Forghani.
• Department of Diagnostic Radiology, McGill University, Room B02 9389, 1001 Decarie Boulevard, Montreal, Quebec H4A 3J1, Canada; School of Computer Science, McGill University, Montreal, Quebec, Canada; Medical Physics Unit, Department of Oncology, McGill University, Montreal, Quebec, Canada; Augmented Intelligence & Precision Health Laboratory (AIPHL), Department of Diagnostic Radiology, Research Institute of the McGill University Health Centre, Montreal, Quebec, Canada. Electronic address: peter.savadjiev@mcgill.ca.
• Neuroimaging Clin. N. Am. 2020 Nov 1; 30 (4): 401-415.

AbstractThe advent of big data and deep learning algorithms has promoted a major shift toward data-driven methods in medical image analysis recently. However, the medical image analysis field has a long and rich history inclusive of both knowledge-driven and data-driven methodologies. In the present article, we provide a historical review of an illustrative sample of medical image analysis methods and locate them along a knowledge-driven versus data-driven continuum. In doing so, we highlight the historical importance as well as current-day relevance of more traditional, knowledge-based artificial intelligence approaches and their complementarity with fully data-driven techniques such as deep learning.Copyright © 2020 Elsevier Inc. All rights reserved.

12 keywords...

hide…