Machine learning for functional outcome prediction after vestibular schwannoma surgery: a systematic review and diagnostic test accuracy meta-analysis

Machine learning (ML) models have been increasingly applied to predict postoperative facial nerve dysfunction and hearing preservation after vestibular schwannoma (VS) surgery. However, reported performance varies substantially, and the overall diagnostic accuracy and clinical reliability of these models remain uncertain....

Continue reading at PubMed →

Machine learning prediction models for vestibular schwannoma surgical outcomes are still in the research phase; no specific model is validated sufficiently for routine clinical use, so no change in pre-surgical counseling or decision-making protocols is warranted yet.

If machine learning models can reliably forecast facial nerve and hearing outcomes before vestibular schwannoma surgery, they could reshape patient counseling, surgical planning, and shared decision-making across neurotology and audiology.

1. 01Systematic review + meta-analysis evaluating ML models for predicting outcomes after vestibular schwannoma (acoustic neuroma) surgery.
2. 02Two key outcomes assessed: postoperative facial nerve dysfunction and hearing preservation.
3. 03Diagnostic test accuracy meta-analysis framework used to pool sensitivity/specificity across included studies.
4. 04Findings reflect the current state of ML in neurotology — promising but not yet clinically validated at scale.
5. 05Highlights need for standardized outcome reporting and external validation before clinical adoption.

PMID

42637963

DOI

10.1007/s11060-026-05747-5.

Journal

Journal of Neuro-Oncology

Publication type

meta_analysis

Evidence level

1a

Population

Patients undergoing vestibular schwannoma (acoustic neuroma) surgery

Intervention

Machine learning models for predicting surgical outcomes

Comparator

Standard clinical predictors / reference standard outcome measures