Classifying voice disorders for machine learning: a pilot study using the USVAC-C2025 diagnostic framework - Scorecard - MDSpire

Classifying voice disorders for machine learning: a pilot study using the USVAC-C2025 diagnostic framework

  • By

  • Catherine Madill

  • Zhou Hao Leong

  • Dharshini Manoharan

  • Dhanshree Gunjawate

  • Charu Grover

  • Katrina Sandham

  • Rijul Gupta

  • Craig Jin

  • Duy Duong Nguyen

  • James Jordan Johnson

  • Daniel Novakovic

  • June 23, 2026

  • 0 min

Share

Clinical Scorecard: Developing a Multilayer Classification System for Voice Disorders in Machine Learning: A Preliminary Study Utilizing the USVAC-C2025 Diagnostic Framework

At a Glance

CategoryDetail
ConditionVoice Disorders
Key MechanismsMultilayer classification framework for diagnostic consistency
Target PopulationAdults with voice disorders
Care SettingTertiary voice clinic

Key Highlights

  • High intra-rater reliability (ICC range, 0.768–0.865)
  • Strong inter-rater reliability for disordered vs. non-disordered voices (Îș = 0.812)
  • Multilayer framework developed through multidisciplinary consensus
  • Agreement declines with increasing diagnostic specificity
  • Framework supports future machine learning applications

Guideline-Based Recommendations

Diagnosis

  • Utilize a structured multilayer classification framework for voice disorders.

Management

  • Apply the framework for consistent diagnostic labeling in clinical settings.

Monitoring & Follow-up

  • Assess inter- and intra-rater reliability to ensure diagnostic consistency.

Risks

  • Inconsistencies in diagnosis may hinder machine learning model training.

Patient & Prescribing Data

Adults with diagnosed voice disorders

Framework aids in systematic annotation for clinical decision support.

Clinical Best Practices

  • Implement structured classification to enhance diagnostic transparency.
  • Encourage multidisciplinary collaboration in voice disorder assessment.
  • Standardize diagnostic frameworks to improve machine learning integration.

Related Resources & Content

    Original Source(s)

    Classifying voice disorders for machine learning: a pilot study using the USVAC-C2025 diagnostic framework

    • by Catherine Madill, Zhou Hao Leong, Dharshini Manoharan, Dhanshree Gunjawate, Charu Grover, Katrina Sandham, Rijul Gupta, Craig Jin, Duy Duong Nguyen, James Jordan Johnson, Daniel Novakovic

    • June 23, 2026

    Related Content

    Data-driven refinements for voice disorder classification: improving accuracy and generalisability

    Tissue-Resident NK Cells Slow Solid Tumor Growth in Mice

    The study points to a possible off-the-shelf cell therapy for solid tumors

    • June 16, 2026

    • 3 min
    Enhanced parathyroid identification and preservation in total thyroidectomy using combined near-infrared autofluorescence and carbon nanoparticles