Towards subject-level cerebral infarction classification of CT scans using convolutional networks

• Published in: PloS one Vol. 15; no. 7; p. e0235765
• Main Authors: Schultheiss, Manuel, Noël, Peter B., Riederer, Isabelle, Thiele, Frank, Kopp, Felix K., Renger, Bernhard, Rummeny, Ernst J., Pfeiffer, Franz, Pfeiffer, Daniela
• Format: Journal Article
• Language: English
• Published: San Francisco Public Library of Science 15.07.2020; Public Library of Science (PLoS)
• Subjects: Algorithms; Analysis; Artificial neural networks; Bioengineering; Biology and Life Sciences; CAT scans; Cerebral infarction; Classification; Clinical decision making; Computed tomography; Datasets; Decision making; Diagnosis; Health aspects; Identification and classification; Indexing; Infarction; Machine learning; Magnetic resonance imaging; Medical imaging; Medicine; Medicine and Health Sciences; Methods; People and Places; Physics; Probability; Research and Analysis Methods; Segmentation; Stroke; Germany
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0235765

Automatic evaluation of 3D volumes is a topic of importance in order to speed up clinical decision making. We describe a method to classify computed tomography scans on volume level for the presence of non-acute cerebral infarction. This is not a trivial task, as the lesions are often similar to other areas in the brain regarding shape and intensity. A three stage architecture is used for classification: 1) A cranial cavity segmentation network is developed, trained and applied. 2) Region proposals are generated 3) Connected regions are classified using a multi-resolution, densely connected 3D convolutional network. Mean area under curve values for subject level classification are 0.95 for the unstratified test set, 0.88 for stratification by patient age and 0.93 for stratification by CT scanner model. We use a partly segmented dataset of 555 scans of which 186 scans are used in the unstratified test set. Furthermore we examine possible dataset bias for scanner model and patient age parameters. We show a successful application of the proposed three-stage model for full volume classification. In contrast to black-box approaches, the convolutional network's decision can be further assessed by examination of intermediate segmentation results.