Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Sep 18;10(18):e38088.
doi: 10.1016/j.heliyon.2024.e38088. eCollection 2024 Sep 30.

Research on improved YOLOv8s model for detecting mycobacterium tuberculosis

Hao Chen  1 , Wenye Gu  2 , Haifei Zhang  1 , Yuwei Yang  1 , Lanmei Qian  1
Affiliations

Research on improved YOLOv8s model for detecting mycobacterium tuberculosis

Hao Chen et al. Heliyon. .

Abstract

Accurate identification of Mycobacterium tuberculosis (M. tuberculosis) is a critical step in the diagnosis of tuberculosis. Existing object detection methods struggle with the challenges posed by the varied morphology and size of M. tuberculosis in sputum smear images, which makes precise targeting difficult. To solve these problems, an improved YOLOv8s model is proposed. Specifically, an additional detection head is added to focus on small target information. Second, a multi-scale feature fusion module is introduced to adapt the model to different sizes of M. tuberculosis. In addition, a convolutional layer is added to the Coordinate Attention (CA) module to extract more advanced semantic features. Finally, a self-attention mechanism is added after the CA module to enhance the model's ability to accurately understand and localize the varied morphology of M. tuberculosis. Our model performed well with an average precision of 85.7 % when tested on a publicly available dataset. This clearly demonstrates the effectiveness of our proposed model in M. tuberculosis detection.

Keywords: CA; M. tuberculosis; Multi-scale feature fusion; YOLOv8s.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Figures

Fig. 1
Fig. 1
Model network structure.
Fig. 2
Fig. 2
Multi-scale feature fusion structure.
Fig. 3
Fig. 3
Eccsa structure.
Fig. 4
Fig. 4
Examples of sputum sample image.
Fig. 5
Fig. 5
Total training loss curve of the model.
Fig. 6
Fig. 6
Detection results of the proposed model.
Fig. 7
Fig. 7
Trend plot of AP for each model in the comparison experiments.
Fig. 8
Fig. 8
Detection results in the comparison experiments. (a) SSD. (b) YOLOv3. (c) YOLOv5s. (d) Improved YOLOv4. (e) Improved YOLOv5s. (f) Ours.
Fig. 9
Fig. 9
Trend plot of AP in the ablation experiments.
Fig. 10
Fig. 10
Detection results in the ablation experiments. (a) Test Sample. (b) YOLOv8s. (c) Ours.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Natarajan A., Beena P.M., Devnikar A.V., Mali S. A systemic review on tuberculosis. Indian J. Tubercul. 2020;67(3):295–311. - PubMed
    1. An L., Peng K., Yang X., Huang P., Luo Y., Feng P., Wei B. E-TBNet: light deep neural network for automatic detection of tuberculosis with X-ray DR imaging. Sensors. 2022;22(3):821. - PMC - PubMed
    1. Kotei E., Thirunavukarasu R. A comprehensive review on advancement in deep learning techniques for automatic detection of tuberculosis from chest X-ray images. Arch. Comput. Methods Eng. 2024;31(1):455–474.
    1. Dinkele R., Gessner S., McKerry A., Leonard B., Seldon R., Koch A.S., Warner D.F. Capture and visualization of live Mycobacterium tuberculosis bacilli from tuberculosis patient bioaerosols. PLoS Pathog. 2021;17(2) - PMC - PubMed
    1. Zachariou M., Arandjelović O., Sloan D.J. Automated methods for tuberculosis detection/diagnosis: a literature review. BioMedInformatics. 2023;3(3):724–751.

LinkOut - more resources