Validation of Artificial Intelligence-Based Facial Paralysis Assessment in Patients With Bell's Palsy

ConditionsBell's Palsy Facial Nerve Paralysis

Trial statusNot yet recruiting

Trial phaseNot listed

Trial typeObservational

Biological sexAll

Age25-40

SponsorCairo University

About this trial

This observational study aims to assess the concurrent validity of an artificial intelligence (AI)-based facial paralysis assessment system in patients with unilateral Bell's palsy. Currently, clinical assessment relies on subjective scales like the Sunnybrook Facial Grading System, which can vary between different observers. This study will compare AI-generated composite asymmetry scores-derived from real-time computer vision analysis of facial landmarks-with scores from the Sunnybrook system. The goal is to determine if AI can provide a valid, objective method for monitoring facial nerve recovery.

Eligibility criteria

This trial does not accept healthy volunteers

Qualifiers

Patients with unilateral Bell's palsy.

Patients must be within one month of onset of Bell's palsy symptoms at the time of enrollment.

Body mass index (BMI) less than 30 $kg/m²$.

Patients must be cooperative and able to follow simple verbal instructions during facial movement tasks.

Disqualifiers

Bilateral facial paralysis or recurrent Bell's palsy.

Facial nerve palsy due to known secondary causes (e.g., trauma, neoplasm, infection, stroke, Ramsay Hunt syndrome, or otitis media).

Facial deformities, scars, or burns that interfere with facial motion detection.

Uncooperative or cognitively impaired individuals unable to follow instructions or maintain required facial postures.

Trial population

Sixty-three patients of both sexes with unilateral Bell's palsy will be recruited from private clinics. Patients will be diagnosed by a neurologist based on clinical examination.

Trial design

Design model

Cohort

Time perspective

Cross-sectional

Treatments tested in this trial

Sunnybrook Facial Grading System (FGS)
Other intervention
Clinical grading of facial muscle paralysis based on resting symmetry, symmetry of voluntary movements, and synkinesis detection.
AI-Based Facial Assessment
Other intervention
Real-time computer vision analysis using deep-learning-based landmark detection to track 468 facial points during standardized facial expressions.

Treatment groups

63 Participants

are divided into 1 treatment group

Group A: Patients with Unilateral Bell's Palsy2 interventions

Trial outcomes

Primary outcomes

Spearman's Rank Correlation Coefficient between AI-derived scores and Sunnybrook Facial Grading System scores.

This measure evaluates the concurrent validity of the AI-based assessment system. The AI system uses 468 facial landmarks to calculate a composite asymmetry score (0-100%). These results will be correlated with the clinical scores from the Sunnybrook Facial Grading System (0-100), where higher scores indicate better facial function.

Time frame

Baseline (single assessment at the time of enrollment).

Secondary outcomes

AI-Based Composite Asymmetry Score.

The specific numerical output generated by the computer-vision system. It quantifies facial symmetry by measuring the amplitude of movement (in pixels/displacement) during five standardized facial expressions: brow lift, eye closure, broad smile, snarl, and lip pucker.

Time frame

Baseline.

Sponsors and contacts

Click on the lead sponsor to view all of their trials.

Cairo University

Lead sponsor

Source ClinicalTrials.gov

This trial is not recruiting at the moment. You can still explore other options: