Analysis of tuberculosis severity levels from CT pulmonary images based on enhanced residual deep learning architecture

Gao, Xiaohong W. ORCID logoORCID: https://orcid.org/0000-0002-8103-6624, James-Reynolds, Carl ORCID logoORCID: https://orcid.org/0000-0002-5892-5415 and Currie, Edward ORCID logoORCID: https://orcid.org/0000-0003-1186-5547 (2020) Analysis of tuberculosis severity levels from CT pulmonary images based on enhanced residual deep learning architecture. Neurocomputing, 392 . pp. 233-244. ISSN 0925-2312 [Article] (doi:10.1016/j.neucom.2018.12.086)

[img]
Preview
PDF - Final accepted version (with author's formatting)
Available under License Creative Commons Attribution-NonCommercial-NoDerivatives 4.0.

Download (1MB) | Preview

Abstract

This research investigates the application of CT pulmonary images to the detection and characterisation of TB at five levels of severity, in order to monitor the efficacy of treatment. To contend with smaller datasets (i.e. in hundreds) and the characteristics of CT TB images in which abnormalities occupy only limited regions, a 3D block-based residual deep learning network (ResNet) coupled with injection of depth information (depth-Resnet) at each layer was implemented. Progress in evaluation has been accomplished in two ways. One is to assess the proposed depth-Resnet in prediction of severity scores and another is to analyse the probability of high severity of TB. For the former, delivered results are of 92.70 ± 5.97% and 67.15 ± 1.69% for proposed depth-Resnet and ResNet-50 respectively. For the latter, two additional measures are put forward, which are calculated using (1) the overall severity (1 to 5) probability, and (2) separate probabilities of both high severity (scores of 1 to 3) and low severity (scores of 4 and 5) respectively, when scores of 1 to 5 are mapped into initial probabilities of (0.9, 0.7, 0.5, 0.3, 0.2) respectively. As a result, these measures achieve the averaged accuracies of 75.88% and 85.29% for both methods respectively.

Item Type: Article
Research Areas: A. > School of Science and Technology > Computer Science
Item ID: 25919
Notes on copyright: © 2019 Elsevier B.V. This author's accepted manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/
Useful Links:
Depositing User: Xiaohong Gao
Date Deposited: 07 Jan 2019 12:54
Last Modified: 29 Nov 2022 18:24
URI: https://eprints.mdx.ac.uk/id/eprint/25919

Actions (login required)

View Item View Item

Statistics

Activity Overview
6 month trend
378Downloads
6 month trend
358Hits

Additional statistics are available via IRStats2.