Grants Funded
ASPS/PSF leadership is committed to continuing to provide high levels of investigator-initiated research support to ensure that plastic surgeons have the needed research resources to be pioneers and innovators in advancing the practice of medicine.
Research Abstracts
Search The PSF database to have easy access to full-text grant abstracts from past PSF-funded research projects 2003 to present. All abstracts are the work of the Principal Investigators and were retrieved from their PSF grant applications. Several different filters may be applied to locate abstracts specific to a particular focus area or PSF funding mechanism.
Using AI to Predict Palatal Dynamics in Patients with Velopharyngeal Insufficency
Principal Investigator
Miles Pfaff MD, MHS
Miles Pfaff MD, MHS
Year
2025
2025
Institution
The Regents of the University of California, Irvine
The Regents of the University of California, Irvine
Funding Mechanism
Focus Area
Cranio/Maxillofacial/Head and Neck, Technology Based
Cranio/Maxillofacial/Head and Neck, Technology Based
Abstract
Project Summary
Velopharyngeal Insufficiency (VPI) is a significant condition associated with impaired speech characterized by hypernasality, nasal air emission, and articulation errors, which can severely impact patients' quality of life. Nasopharyngoscopy (NP) is the gold standard for evaluating velopharyngeal function, but its effectiveness is limited by the subjectivity of clinician interpretation, resulting in
inconsistent diagnoses and treatment planning. The proposed research aims to develop an artificial intelligence (AI)-based system, VPI-Net, to provide objective, consistent, and repeatable assessments of velopharyngeal function through speech samples. The goal of VPI-Net will be to automatically link speech with velar assessment to assist with VPI diagnosis. The AI system will utilize deep
learning models to analyze NP video frames. The model will also be trained to automate NP studies using a standardized NP assessment tool for VPI. This approach will enable anatomical and functional assessment, which are crucial for understanding the
manifestations of VPI during various speech tasks. Additionally, representative control data will be incorporated into the model to improve specificity and minimize false positives. The end goal is to create a tool that will assist clinicians with the diagnosis and assessment of VPI.
Impact Statement
Our proposal aims to build a clinically impactful AI-based model to assess patients with VPI and to provide more consistent and accessible evaluations for providers of patients with speech differences. By linking and automating nasopharyngoscopy assessment with speech, this proposal will offer objective metrics that enhance clinical decision-making. Such developments could minimize the
need for invasive diagnostics while expanding the availability of reliable assessments to more healthcare facilities, especially those in resource-limited settings.
Project Summary
Velopharyngeal Insufficiency (VPI) is a significant condition associated with impaired speech characterized by hypernasality, nasal air emission, and articulation errors, which can severely impact patients' quality of life. Nasopharyngoscopy (NP) is the gold standard for evaluating velopharyngeal function, but its effectiveness is limited by the subjectivity of clinician interpretation, resulting in
inconsistent diagnoses and treatment planning. The proposed research aims to develop an artificial intelligence (AI)-based system, VPI-Net, to provide objective, consistent, and repeatable assessments of velopharyngeal function through speech samples. The goal of VPI-Net will be to automatically link speech with velar assessment to assist with VPI diagnosis. The AI system will utilize deep
learning models to analyze NP video frames. The model will also be trained to automate NP studies using a standardized NP assessment tool for VPI. This approach will enable anatomical and functional assessment, which are crucial for understanding the
manifestations of VPI during various speech tasks. Additionally, representative control data will be incorporated into the model to improve specificity and minimize false positives. The end goal is to create a tool that will assist clinicians with the diagnosis and assessment of VPI.
Impact Statement
Our proposal aims to build a clinically impactful AI-based model to assess patients with VPI and to provide more consistent and accessible evaluations for providers of patients with speech differences. By linking and automating nasopharyngoscopy assessment with speech, this proposal will offer objective metrics that enhance clinical decision-making. Such developments could minimize the
need for invasive diagnostics while expanding the availability of reliable assessments to more healthcare facilities, especially those in resource-limited settings.
Biography
Miles J. Pfaff, MD, MHS, FACS, FAAP is a craniomaxillofacial surgeon at the University of California, Irvine, and at Children's Hospital
of Orange County. He currently holds the title of Assistant Professor within the Department of Plastic Surgery. Dr. Pfaff completed his undergraduate studies in cellular and molecular biology at the University of California, Berkeley. Dr. Pfaff then went on to complete his medical training at Yale School of Medicine. While at Yale, Dr. Pfaff completed a National Institutes of Health research fellowship and earned a Master’s degree focused on investigating the complex biological mechanisms that result in premature suture fusion, also known as craniosynostosis. He completed his Plastic Surgery training and Craniomaxillofacial Surgery Fellowship at
UCLA and the University of Pittsburgh, respectively.
Dr. Pfaff directs the Children’s Hospital of Orange County’s Ear Anomalies Program and co-leads the CHOC Speech Surgery Clinic.
Dr. Pfaff’s research interests include craniofacial clinical outcomes, translational research, where he is a member of the UC Irvine Center for Tissue Engineering, and clinical innovation, which includes the use of artificial intelligence to improve clinical outcomes for his patients.
Miles J. Pfaff, MD, MHS, FACS, FAAP is a craniomaxillofacial surgeon at the University of California, Irvine, and at Children's Hospital
of Orange County. He currently holds the title of Assistant Professor within the Department of Plastic Surgery. Dr. Pfaff completed his undergraduate studies in cellular and molecular biology at the University of California, Berkeley. Dr. Pfaff then went on to complete his medical training at Yale School of Medicine. While at Yale, Dr. Pfaff completed a National Institutes of Health research fellowship and earned a Master’s degree focused on investigating the complex biological mechanisms that result in premature suture fusion, also known as craniosynostosis. He completed his Plastic Surgery training and Craniomaxillofacial Surgery Fellowship at
UCLA and the University of Pittsburgh, respectively.
Dr. Pfaff directs the Children’s Hospital of Orange County’s Ear Anomalies Program and co-leads the CHOC Speech Surgery Clinic.
Dr. Pfaff’s research interests include craniofacial clinical outcomes, translational research, where he is a member of the UC Irvine Center for Tissue Engineering, and clinical innovation, which includes the use of artificial intelligence to improve clinical outcomes for his patients.