An explainable deep-learning algorithm for the detection of acute intracranial haemorrhage from small datasets
Access StatusFull text of the requested work is not available in DASH at this time ("dark deposit"). For more information on dark deposits, see our FAQ.
Tajmir, Shahein H.
Guerrier, Claude E.
Ebert, Sarah A.
MetadataShow full item record
CitationLee, Hyunkwang, Sehyo Yune, Mohammad Mansouri, Myeongchan Kim, Shahein H. Tajmir, Claude E. Guerrier, Sarah A. Ebert et al. "An explainable deep-learning algorithm for the detection of acute intracranial haemorrhage from small datasets." Nature Biomedical Engineering 3, no. 3 (2018): 173-182. DOI: 10.1038/s41551-018-0324-9
AbstractOwing to improvements in image recognition via deep learning, machine-learning algorithms could eventually be applied to automated medical diagnoses that can guide clinical decision-making. However, these algorithms remain a 'black box' in terms of how they generate the predictions from the input data. Also, high-performance deep learning requires large, high-quality training datasets. Here, we report the development of an understandable deep-learning system that detects acute intracranial haemorrhage (ICH) and classifies five ICH subtypes from unenhanced head computed-tomography scans. By using a dataset of only 904 cases for algorithm training, the system achieved a performance similar to that of expert radiologists in two independent test datasets containing 200 cases (sensitivity of 98% and specificity of 95%) and 196 cases (sensitivity of 92% and specificity of 95%). The system includes an attention map and a prediction basis retrieved from training data to enhance explainability, and an iterative process that mimics the workflow of radiologists. Our approach to algorithm development can facilitate the development of deep-learning systems for a variety of clinical applications and accelerate their adoption into clinical practice.
Citable link to this pagehttps://nrs.harvard.edu/URN-3:HUL.INSTREPOS:37371238
- HMS Scholarly Articles