Neuroanesthesia & neurocritical care
Neural networks for estimation of facial palsy after vestibular schwannoma surgery
An erratum to this article was published (J Clin Monit Comput, Jun 2023 ↗)
Authors' conclusion
NN based on preoperative HB, traintime and tumor size provide good estimations of postoperative HB.
Study context
- Population
- Patients undergoing vestibular schwannoma surgery with free-running EMG
- Participants
- 200 consecutive patients
- Comparison
- NN input sets: traintime, preoperative House and Brackmann (HB) grade and tumor size vs tumor size or mean traintime alone
- Outcome
- Agreement of estimated vs clinical postoperative House and Brackmann (HB) grades
Methodological detailsconsecutive patients, neural feed-forward networks
- Consecutive patients
- Neural feed-forward networks
Main finding
As reported in the abstract
Estimation of postoperative HB grades (chi2)
Traintime + preoperative HB + tumor size
54.8
chi2
Tumor size
30.6
chi2
Mean traintime
31.9
chi2
Further findings
Added inputs: separate intermediate nerve or A-train cluster detection
- Did not improve performance
Removal of A-train cluster traintime (patients without separate intermediate nerve, subgroup)
- Improved results (chi2 = 54.8 vs. 51.3)
Original publication
Neural networks for estimation of facial palsy after vestibular schwannoma surgery. J Clin Monit Comput. Published 4 November 2022. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →