Speech-based AI-driven Diagnostics and Rehabilitation for Oral and Oropharyngeal Cancer. (speech omics)

A Preliminary Study on the Phonetics of Resonance and Articulation Disorders Caused by Defects of the Oral and Maxillofacial Speech Organs and Speech Therapy

Participants diagnosed with oral cancer, oropharyngeal cancer, or oral potentially malignant lesions, as well as healthy controls, had their speech audio recordings collected for the development and validation of AI-driven models for diagnosis and prognosis prediction of oral cancer and oropharyngeal cancer.

Study Overview

• Status: Completed
• Conditions: Oral Cancer; Oropharyngeal Cancer

Detailed Description

Participants were instructed to articulate three sustained vowels (/a/, /i/, /u/) repeatedly at a moderate volume and pace, with three repetitions per vowel and each utterance lasting at least one second. We developed a neuromorphic computing framework that orthogonally decomposes acoustic features into ultra-dimensional omics representations, enabling the characterization of both localized lesions and systemic physiological conditions. The study collected a comprehensive spectrum of biological profiles, including sociodemographic characteristics, tumor metrics, oral function-related factors, patient-reported outcome measures (PROMs), immunoinflammatory indices, and general health status indicators, to thoroughly investigate the paralinguistic representations of transformed speech omics features. These features were then rigorously evaluated for their clinical efficacy across multiple diagnostic tasks, including screening, early detection, pathological diagnosis, disease staging, and risk factor identification.

• Study Type: Observational
• Enrollment (Actual): 501

Contacts and Locations

Study Locations

• China
  • Guangdong
    • Guangzhou, Guangdong, China, 510055
      • Department of oral and maxillofacial surgery, Hospital of Stomatology, Sun Yat-sen University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

In-patient participants diagnosed with oral cancer, oropharyngeal cancer or oral potentially malignant lesion and healthy volunteers.

Description

Inclusion Criteria:

• diagnosed as having OC or OPC regardless of concrete pathological subtype (e.g., epithelial, adenoid, odontogenic)
• without a history of other tumor-related treatment before the initial evaluation, such as radiotherapy and chemotherapy
• native Chinese speakers

Exclusion Criteria:

• hearing impairment
• history of stuttering, cerebrovascular accident, brain trauma, neurodegenerative diseases
• severe dental or maxillofacial deformity
• cleft lip, cleft palate, and related post-treatment status
• severe cardiology conditions, breathing/pulmonology disorders, psychiatric disorders or other illnesses preventing patients from receiving standard surgery

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort
healthy control and oral potentially malignant lesion; Speakers with oral potentially malignant lesions or healthy status
oral cancer and oropharyngeal cancer; Patients diagnosed with oral cancer or oropharyngeal cancer

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
AUC value	Area under the receiver operating characteristic curve (AUC) for discriminating OC/OPC from healthy controls	From enrollment to the report of surgical pathology，up to two weeks.

Collaborators and Investigators

• Sponsor: Yudong Xiao

Study record dates

Study Major Dates

• Study Start (Actual): July 17, 2017
• Primary Completion (Actual): April 30, 2022
• Study Completion (Actual): December 31, 2022

Study Registration Dates

• First Submitted: December 4, 2025
• First Submitted That Met QC Criteria: January 11, 2026
• First Posted (Estimated): January 16, 2026

Study Record Updates

• Last Update Posted (Estimated): January 16, 2026
• Last Update Submitted That Met QC Criteria: January 11, 2026
• Last Verified: January 1, 2026

More Information

Terms related to this study

• Keywords: speech signal analysis; diagnostic models; vowel acoustics; bio-inspired computing
• Additional Relevant MeSH Terms: Mouth Diseases; Stomatognathic Diseases; Neoplasms by Site; Neoplasms; Head and Neck Neoplasms; Otorhinolaryngologic Diseases; Pharyngeal Neoplasms; Otorhinolaryngologic Neoplasms; Pharyngeal Diseases; Mouth Neoplasms; Oropharyngeal Neoplasms
• Other Study ID Numbers: ERC-[2014]-35

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: Raw audio files and corresponding metadata files for this study are uploaded and shared. The underlying code for this study, including PRAAT scripts for speech preprocessing and automatic feature extraction, and python codes for speech omics representation, can be accessed.
• IPD Sharing Time Frame: permanent valid
• IPD Sharing Supporting Information Type: ANALYTIC_CODE

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No