A multi-stage transfer learning strategy for diagnosing a class of rare laryngeal movement disorders

Yu Yao; Maria Powell; Jules White; Jian Feng; Quchen Fu; Peng Zhang; Douglas C Schmidt

doi:10.1016/j.compbiomed.2023.107534

A multi-stage transfer learning strategy for diagnosing a class of rare laryngeal movement disorders

Comput Biol Med. 2023 Nov:166:107534. doi: 10.1016/j.compbiomed.2023.107534. Epub 2023 Sep 29.

Authors

Yu Yao¹, Maria Powell², Jules White³, Jian Feng⁴, Quchen Fu³, Peng Zhang³, Douglas C Schmidt³

Affiliations

¹ The College of Information Science and Engineering, Northeastern University, Wenhua Road 3-11, Shenyang, 110819, Liaoning, PR China; The Department of Electrical Engineering and Computer Science, Vanderbilt University, 2301 Vanderbilt Place, Nashville, 37235, TN, USA. Electronic address: yu.ethan.yao@gmail.com.
² Vanderbilt University Medical Center, 1211 Medical Center Drive, Nashville, 37232, TN, USA.
³ The Department of Electrical Engineering and Computer Science, Vanderbilt University, 2301 Vanderbilt Place, Nashville, 37235, TN, USA.
⁴ The College of Information Science and Engineering, Northeastern University, Wenhua Road 3-11, Shenyang, 110819, Liaoning, PR China.

PMID: 37801923
DOI: 10.1016/j.compbiomed.2023.107534

Abstract

Background: It remains hard to directly apply deep learning-based methods to assist diagnosing essential tremor of voice (ETV) and abductor and adductor spasmodic dysphonia (ABSD and ADSD). One of the main challenges is that, as a class of rare laryngeal movement disorders (LMDs), there are limited available databases to be investigated. Another worthy explored research question is which above sub-disorder benefits most from diagnosis based on sustained phonations. The question is from the fact that sustained phonations can help detect pathological voice from healthy voice.

Method: A transfer learning strategy is developed for LMD diagnosis with limited data, which consists of three fundamental parts. (1) An extra vocally healthy database from the International Dialects of English Archive (IDEA) is employed to pre-train a convolutional autoencoder. (2) The transferred proportion of the pre-trained encoder is explored. And its impact on LMD diagnosis is also evaluated, yielding a two-stage transfer model. (3) A third stage is designed following the initial two stages to embed information of pathological sustained phonation into the model. This stage verifies the different effects of applying sustained phonation on diagnosing the three sub-disorders, and helps boost the final diagnostic performance.

Results: The analysis in this study is based on clinician-labeled LMD data obtained from the Vanderbilt University Medical Center (VUMC). We find that diagnosing ETV shows sensitivity to sustained phonation within the current database. Meanwhile, the results show that the proposed multi-stage transfer learning strategy can produce (1) accuracy of 65.3% on classifying normal and other three sub-disorders all at once, (2) accuracy of 85.3% in differentiating normal, ABSD, and ETV, and (3) accuracy of 77.7% for normal, ADSD and ETV. These findings demonstrate the effectiveness of the proposed approach.

Keywords: Convolutional autoencoder; Laryngeal movement disorders; Limited data; Multi-stage transfer; Sustained phonation.

MeSH terms

Databases, Factual
Deep Learning*
Diagnosis, Computer-Assisted / methods
Female
Humans
Laryngeal Diseases / diagnosis
Larynx / physiopathology
Male
Movement Disorders / diagnosis
Movement Disorders / physiopathology