Our team adapted machine learning and computer vision methods to develop a technique that rapidly and automatically detects CT-identifiable lesions. Specifically, we use scale invariant feature transform (SIFT)1 and deep convolutional neural networks (CNN)2 to identify important image features that can distinguish TBI lesions from background data. Our learning algorithm is a linear support vector machine (SVM)3. Further, we also employ tools from topological data analysis (TDA) for gleaning insights into the correlation patterns between healthy and pathological data. The technique was validated using 409 CT scans of the brain, acquired via the Progesterone for the Treatment of Traumatic Brain Injury phase III clinical trial (ProTECT_III) which studied patients with moderate to severe TBI4. CT data were annotated by a central radiologist and included patients with positive and negative scans. Additionally, the largest lesion on each positive scan was manually segmented. We reserved 80% of the data for training the SVM and used the remaining 20% for testing. Preliminary results are promising with 92.55% prediction accuracy (sensitivity = 91.15%, specificity = 93.45%), indicating the potential usefulness of this technique in clinical scenarios.
ACCESS THE FULL ARTICLE
Krishna N. Keshavamurthy, Owen P. Leary, Lisa H. Merck, Benjamin Kimia, Scott Collins, David W. Wright, Jason W. Allen, Jeffrey F. Brock, Derek Merck, "Machine learning algorithm for automatic detection of CT-identifiable hyperdense lesions associated with traumatic brain injury," Proc. SPIE 10134, Medical Imaging 2017: Computer-Aided Diagnosis, 101342G (23 March 2017);