EUVIMEDEUVIMED entdecken
Crossref

Blink Reflex and Facial Nerve Outcomes: Using a Machine Learning Tool

Jihad Abdelgadir, Harrison Hockenberry, Tanner Zachem, Syed Adil, Daniel Sexton, Holly Johnson, Patrick Codd, Rory Goodwin, Ali Zomorodi

Journal of Neurological Surgery Part B: Skull Base · 01.02.2024 · 85 (S 01 S 01) · S1-S398

DOI: 10.1055/s-0044-1780090 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Neurological Surgery Part B: Skull Base, erschienen am 01.02.2024. Verfasst von Jihad Abdelgadir und 8 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Despite robust EMG monitoring efforts during surgery, complications are somewhat common with around a 40% chance of facial nerve sequelae following surgery. This research will explore the feasibility of utilizing large-scale EMG data to determine the prognostic value of blink reflex data in the preservation of facial nerve function.

Methods

This study involves assessing EMG data collected prospectively from patients undergoing translabyrinthine approach for suspected vestibular schwannoma. To assess the blink reflex, a stimulus train of four to seven individual pulses was used to elicit a response. Intensity was measured in mA and increased until a response was elicited. To test feasibility, we collected EMG data from six patients and analyzed tumor volume, House-Brackmann score, and changes to blink reflex amplitude over time. This data was quantified using the MNE EMG data analysis package within a Jupyter Python 3 environment. Kruskal–Wallis tests were performed to test the association between blink reflex EMG changes and House-Brackmann scores. Mann–Whitney tests were used to identify significant differences between tumor-side blink reflex, contralateral amplitude and House-Brackmann scores. Statistical significance was determined to be p ≤ 0.05.

Results

We collected demographic data from a total of 6 patients, 3 were female and 3 were male, the average age was 53 and the median age at diagnosis was 52.5. We observed an average tumor volume of 5.05 cm 3 (CI: 5.05–2.15, SD: 3.47) and an average House-Brackmann score of 1.44 (CI: 1.44–0.66, SD: 1.01), range 1–4. The average blink reflex amplitude for tumor side was -1.60 μV (CI: −9.65 to 6.45, SD: 5.8) while the baseline average contralateral amplitude was −1.67 (CI: −13.01 to 9.67, SD: 8.18). We did not observe a significant difference between tumor-side blink reflex and baseline contralateral amplitude ( p = 0.67). Our Mann–Whitney test revealed a nonsignificant difference between House-Brackmann score and right-sided blink reflex amplitude ( p = 0.60) or left-sided blink reflex amplitude ( p = 0.17). Only one patient in our sample had facial nerve deficit with a House-Brackmann score of 4. Interestingly, our algorithm was able to detect a difference in the tumor side blink reflex over time as well as compared to the contralateral side blink reflex. The difference in amplitude was 1.82 μV (−9.27 μV on the Tumor side and −7.45 μV baseline), and the change in amplitude over time was measured to be −5.71 μV ([ Fig. 2 ]).

Conclusion

In this study, we demonstrate the feasibility of using machine learning to interpret facial nerve monitoring data as Blink reflex EMG and correlate it with Facial nerve outcomes. Our small sample size and the heterogenous nature of neuromonitoring significantly limit our analysis of the association between blink reflex and tumor side. Future work will delve into larger-scale databases to determine the prognostic value of perioperative neuromonitoring and the preservation of facial nerve function following surgical intervention for vestibular schwannoma and other pathologies where the facial nerve is deemed at risk. Fig. 1 Change in amplitude on the tumor side overtime. Fig. 2 House-Brackmann score and blink reflex response. Publication History Article published online: 05 February 2024 © 2024. Thieme. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany

Zitieren

APA

Abdelgadir, J., Hockenberry, H., Zachem, T., Adil, S., Sexton, D., Johnson, H., Codd, P., Goodwin, R., & Zomorodi, A. (2024). Blink Reflex and Facial Nerve Outcomes: Using a Machine Learning Tool. Journal of Neurological Surgery Part B: Skull Base, 85(S 01 S 01), S1-S398. https://doi.org/10.1055/s-0044-1780090

Vancouver

Abdelgadir J, Hockenberry H, Zachem T, Adil S, Sexton D, Johnson H, et al. Blink Reflex and Facial Nerve Outcomes: Using a Machine Learning Tool. Journal of Neurological Surgery Part B: Skull Base. 2024;85(S 01 S 01):S1-S398. doi:10.1055/s-0044-1780090

BibTeX
@article{Abdelgadir2024,
  author = {Abdelgadir, Jihad and Hockenberry, Harrison and Zachem, Tanner and Adil, Syed and Sexton, Daniel and Johnson, Holly and Codd, Patrick and Goodwin, Rory and Zomorodi, Ali},
  title = {Blink Reflex and Facial Nerve Outcomes: Using a Machine Learning Tool},
  journal = {Journal of Neurological Surgery Part B: Skull Base},
  year = {2024},
  volume = {85},
  number = {S 01 S 01},
  pages = {S1--S398},
  doi = {10.1055/s-0044-1780090}
}
Kostenlos starten →