Abstract
Objective: 1) Evaluate the feasibility of an automated zonal facial function grading system through comparison with the House-Brackmann II (HBII) scale. 2) Determine whether such a computerized tool for facial analysis patients can provide objective, accurate, and reproducible assessment in both global and zonal regions of the face.
Method: A prospective clinical trial was performed with 77 consecutive new facial paralysis patients undergoing standardized facial videography. Videos were independently scored by 3 facial nerve specialists using the HBII scale. Subsequent digital analysis was performed with Facogram software, and computer-generated scores were recorded and compared.
Results: Interobserver correlation was high using the HBII scale, with correlation coefficients of 0.81, 0.86, and 0.84. Automated Facogram scores correlated well with manual HBII (scores were within 1.0 ± 0.59, 1.4 ± 1.1, 1.1 ± 1.0, 1.1 ± 0.8, and 1.4 ± 0.8 for overall grade, brow, eye closure, nasolabial fold, and oral commissure, respectively). The HBII required more clinician time to calculate, with an average of 72 ± 21 seconds per case, compared with the Facogram which did not require any clinician time. Specific patient characteristics and video techniques that improved accuracy of the software were identified.
Conclusion: An automated, zone-specific facial analysis tool would eliminate inter-rater subjectivity and allow assessment of facial paralysis patients in a standardized manner across geographic regions and medical specialties. Facogram correlates well with the HBII system. Clinicians should understand the ideal patient population and videotaping technique to optimize accuracy of the program.
Get full access to this article
View all access options for this article.