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
. 2023 Nov 2:14:1291692.
doi: 10.3389/fmicb.2023.1291692. eCollection 2023.

A novel deep learning model for glioma epilepsy associated with the identification of human cytomegalovirus infection injuries based on head MR

Wei Wang  1 , Xuanyi Li  2 , Lou Ye  3 , Jian Yin  4
Affiliations

A novel deep learning model for glioma epilepsy associated with the identification of human cytomegalovirus infection injuries based on head MR

Wei Wang et al. Front Microbiol. .

Abstract

Purpose: In this study, a deep learning model was established based on head MRI to predict a crucial evaluation parameter in the assessment of injuries resulting from human cytomegalovirus infection: the occurrence of glioma-related epilepsy. The relationship between glioma and epilepsy was investigated, which serves as a significant indicator of labor force impairment.

Methods: This study enrolled 142 glioma patients, including 127 from Shengjing Hospital of China Medical University, and 15 from the Second Affiliated Hospital of Dalian Medical University. T1 and T2 sequence images of patients' head MRIs were utilized to predict the occurrence of glioma-associated epilepsy. To validate the model's performance, the results of machine learning and deep learning models were compared. The machine learning model employed manually annotated texture features from tumor regions for modeling. On the other hand, the deep learning model utilized fused data consisting of tumor-containing T1 and T2 sequence images for modeling.

Results: The neural network based on MobileNet_v3 performed the best, achieving an accuracy of 86.96% on the validation set and 75.89% on the test set. The performance of this neural network model significantly surpassed all the machine learning models, both on the validation and test sets.

Conclusion: In this study, we have developed a neural network utilizing head MRI, which can predict the likelihood of glioma-associated epilepsy in untreated glioma patients based on T1 and T2 sequence images. This advancement provides forensic support for the assessment of injuries related to human cytomegalovirus infection.

Keywords: deep learning; epilepsy; glioma; human cytomegalovirus; injury assessment.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Human cytomegalovirus reinfection is associated with a potential glioma risk.
Figure 2
Figure 2
Flowchart of the sample collection process in this study, which involved data from two centers. We primarily excluded patients with poor image quality, those with brain diseases, and patients with low-grade gliomas that were difficult to annotate on MRI. To improve data quality, patients with a history of epilepsy or those whose epilepsy diagnosis was challenging based on clinical symptoms were also excluded.
Figure 3
Figure 3
Schematic diagram of labeled images, (A) T1 sequence image, (B)T2 sequence image, (C) manually labeled image.
Figure 4
Figure 4
(A) The structure diagram of convolutional block attention module (CBAM). (B) We use early-fusion data to greatly improve the speed of neural network training.
Figure 5
Figure 5
Age distribution of patients in training sets, test sets and validation sets.
Figure 6
Figure 6
Age distribution in patients with and without epilepsy.
Figure 7
Figure 7
Results of Four Machine Learning Methods on the Validation Set, with the four parameters from left to right being accuracy, precision, recall, and F1-score.
Figure 8
Figure 8
Results of the four machine learning methods on the test set, with the four parameters shown from left to right: Accuracy, Precision, Recall, and F1-score.
Figure 9
Figure 9
Results of deep learning on the training set, with the four performance metrics displayed from left to right: accuracy, precision, recall, and F1-score.
Figure 10
Figure 10
Results of deep learning on the validation set, with the four performance metrics displayed from left to right: accuracy, precision, recall, and F1-score.
Figure 11
Figure 11
Deep learning results on the test set, with the four metrics displayed from left to right as accuracy, precision, recall, and F1-score.
Figure 12
Figure 12
(A) Structure diagram, (B) Loss curve and (C) accuracy curve of GE-Net.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Benbadis S. R., Beniczky S., Bertram E., MacIver S., Moshé S. L. (2020). The role of EEG in patients with suspected epilepsy. Epileptic Disord. 22, 143–155. doi: 10.1684/epd.2020.1151, PMID: - DOI - PubMed
    1. Bernasconi A., Cendes F., Theodore W. H., Gill R. S., Koepp M. J., Hogan R. E., et al. . (2019). Recommendations for the use of structural magnetic resonance imaging in the care of patients with epilepsy: a consensus report from the international league against epilepsy neuroimaging task force. Epilepsia 60, 1054–1068. doi: 10.1111/epi.15612, PMID: - DOI - PubMed
    1. Boison D. (2010). Adenosine dysfunction and adenosine kinase in epileptogenesis. Open Neurosci J 4:93. doi: 10.2174/1874082001004010093, PMID: - DOI - PMC - PubMed
    1. Chen D. Y., Chen C. C., Crawford J. R., Wang S. G. (2018). Tumor-related epilepsy: epidemiology, pathogenesis and management. J. Neuro-Oncol. 139, 13–21. doi: 10.1007/s11060-018-2862-0, PMID: - DOI - PubMed
    1. Cobbs C. S. (2011). Evolving evidence implicates cytomegalovirus as a promoter of malignant glioma pathogenesis. Herpesviridae 2, 1–7. doi: 10.1186/2042-4280-2-10 - DOI - PMC - PubMed

Grants and funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work was supported by grants from the National Natural Science Foundation of China (81671298).

LinkOut - more resources