Weakly supervised semantic segmentation of tomographic images in the diagnosis of stroke Full article
| Conference |
International Conference «Marchuk Scientific Readings 04-08 Oct 2021 , Новосибирск |
||||||
|---|---|---|---|---|---|---|---|
| Journal |
Journal of Physics: Conference Series
ISSN: 1742-6588 , E-ISSN: 1742-6596 |
||||||
| Output data | Year: 2021, Volume: 2099, Number: 1, Article number : 012021, Pages count : DOI: 10.1088/1742-6596/2099/1/012021 | ||||||
| Authors |
|
||||||
| Affiliations |
|
Funding (1)
| 1 | Russian Foundation for Basic Research | 19-29-01175 |
Abstract:
This paper presents an automatic algorithm for the segmentation of areas affected by an acute stroke in the non-contrast computed tomography brain images. The proposed algorithm is designed for learning in a weakly supervised scenario when some images are labeled accurately, and some images are labeled inaccurately. Wrong labels appear as a result of inaccuracy made by a radiologist in the process of manual annotation of computed tomography images. We propose methods for solving the segmentation problem in the case of inaccurately labeled training data. We use the U-Net neural network architecture with several modifications. Experiments on real computed tomography scans show that the proposed methods increase the segmentation accuracy. © 2021 Institute of Physics Publishing. All rights reserved.
Cite:
Dobshik A.V.
, Tulupov A.A.
, Berikov V.B.
Weakly supervised semantic segmentation of tomographic images in the diagnosis of stroke
Journal of Physics: Conference Series. 2021. V.2099. N1. 012021 . DOI: 10.1088/1742-6596/2099/1/012021 Scopus OpenAlex
Weakly supervised semantic segmentation of tomographic images in the diagnosis of stroke
Journal of Physics: Conference Series. 2021. V.2099. N1. 012021 . DOI: 10.1088/1742-6596/2099/1/012021 Scopus OpenAlex
Identifiers:
| Scopus: | 2-s2.0-85123727143 |
| OpenAlex: | W3197468198 |