From Support Vector Machines to Neural Networks
Although these results are promising and support the ability of ML models to detect VPD, the results likely indicate that the ML models are also picking up confounding data. Efforts are underway to address this problem while simultaneously employing disentanglement tactics to allow for multilingual speech analysis. The ability to clinically operationalize such a model could instantaneously enhance VPD care in LMICs for patients with cleft palate with little changes to existing healthcare infrastructure.
Authors
- Maria Powell (ORCID: https://orcid.org/0000-0002-6643-8991)
- Matthew E. Pontell (ORCID: https://orcid.org/0000-0002-2329-9899)
- Michael S. Golinko (ORCID: https://orcid.org/0000-0002-0715-1915)
- Andrew F. James (ORCID: https://orcid.org/0000-0002-2035-9489)
- Ricardo A. Torres‐Guzman (ORCID: https://orcid.org/0000-0002-4546-551X)
- Scott Corlew (ORCID: https://orcid.org/0000-0002-6101-0443)
- Noah Alter (ORCID: https://orcid.org/0009-0004-7233-3450)
- Andrea R. Hiller (ORCID: https://orcid.org/0000-0001-6965-7818)
- Claiborne Lucas
- Zhijun Yin (ORCID: https://orcid.org/0000-0002-3075-1337)
- Amy L. Stone (ORCID: https://orcid.org/0000-0003-3153-151X)
- Weixin Liu (ORCID: https://orcid.org/0000-0002-1650-0371)
- Bowen Qu
Institutions
- Prisma Health (US)
- Monroe Carell Jr. Children's Hospital (US)
- Harvard Global Health Institute (US)
- Vanderbilt University Medical Center (US)
Publication Details
- Journal
- Annals of Plastic Surgery
- Published
- 2025-09-01
- DOI
- https://doi.org/10.1097/sap.0000000000004460
- Citations
- 2
- Primary Topic
- Cleft Lip and Palate Research
- Type
- article
- Field-Weighted Citation Impact
- 3.63