BBox-Guided Segmentor: Leveraging Expert Knowledge for Accurate Stroke Lesion Segmentation
Using Weakly Supervised Bounding Box Prior

Yanglan Ou (1), Sharon X. Huang (1), Kelvin K. Wong (2), Jonathon Cummock (2), John Volpi (3), James Z. Wang (1), Stephen T.C. Wong (2)

(1) The Pennsylvania State University, University Park, Pennsylvania, USA
(2) T.T. and W.F. Chao Center for BRAIN & Houston Methodist Cancer Center, Houston Methodist Hospital, Houston, Texas, USA
(3) Eddy Scurlock Comprehensive Stroke Center, Department of Neurology, Houston Methodist Hospital, Houston, Texas, USA

Stroke is one of the leading causes of death and disability in the world. Despite intensive research on automatic stroke lesion segmentation from non-invasive imaging modalities including diffusion-weighted imaging (DWI), challenges remain such as a lack of sufficient labeled data for training deep learning models and failure in detecting some small lesions. In this paper, we propose BB-Guided Segmentor, a method that significantly improves the accuracy of stroke lesion segmentation by leveraging expert knowledge. Specifically, our model uses a very coarse bounding box label provided by the expert and then performs accurate segmentation automatically. The small overhead of having the expert provide a rough bounding box leads to large performance improvement in segmentation, which is paramount to accurate stroke diagnosis. To train our model, we employ a weakly-supervised approach that uses a large number of weakly-labeled images with only bounding boxes and a small number of fully labeled images. The scarce fully labeled images are used to train a generator segmentation network, while adversarial training is used to leverage the large number of weakly-labeled images to provide additional learning signals. We evaluate our method extensively using a unique clinical dataset of 99 fully labeled cases (i.e., with full segmentation map labels) and 831 weakly labeled cases (i.e., with only bounding box labels), and the results demonstrate the superior performance of our approach over state-of-the-art stroke lesion segmentation models. We also achieve competitive performance as a SOTA fully supervised method using less than one-tenth of the complete labels. Our proposed approach has the potential to improve stroke diagnosis and treatment planning, which may lead to better patient outcomes.

Full Paper
(high-quality PDF, 2.0MB)

More information

Citation: Yanglan Ou, Sharon X. Huang, Kelvin K. Wong, Jonathon Cummock, John Volpi, James Z. Wang and Stephen T.C. Wong, ``BBox-Guided Segmentor: Leveraging Expert Knowledge for Accurate Stroke Lesion Segmentation Using Weakly Supervised Bounding Box Prior,'' Computerized Medical Imaging and Graphics, vol. , article , pp. -, 2023.

© 2023 Elsevier. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the Elsevier.

Last Modified: April 15, 2023
© 2023