CoChat

SHAP-Based Identification of Potential Acoustic Biomarkers in Patients with Post-Thyroidectomy Voice Disorder

Open access
Celepli, Salih, Bigat, Irem, Karakas, Bilgi, Tezcan, Huseyin Mert, Yar, Mehmet Dincay, Celepli, Pinar, Aksahin, Mehmet Feyzi, Hancerliogullari, Oguz, Yilmaz, Yavuz Fuat, Erogul, Osman · 2025 · Diagnostics

The DOI matches its Crossref record; 93 of the 93 references it cites were found in a registry or an index.

Abstract

Objective: The objective of this study was to identify potential robust acoustic biomarkers for functional post-thyroidectomy voice disorder (PTVD) that may support early diagnosis and personalized treatment strategies, using acoustic analysis and explainable machine learning methods. Methods: Spectral and cepstral features were extracted from /a/ and /i/ voice recordings collected preoperatively and 4–6 weeks postoperatively from a total of 126 patients. Various Support Vector Machine (SVM) and Boosting models were trained. SHapley Additive exPlanations (SHAP) analysis was applied to enhance interpretability. SHAP values from training and test sets were compared via scatter plots to identify stable candidate biomarkers with high consistency. Results: GentleBoost (AUC = 0.85) and LogitBoost (AUC = 0.81) demonstrated the highest classification performance. Performance metrics across all models were evaluated for statistical significance. DeLong’s test was conducted to assess differences between ROC curves. The features iCPP, aCPP, and aHNR were identified as stable candidate biomarkers, exhibiting consistent SHAP distributions in both training and test sets in terms of direction and magnitude. These features showed statistically significant correlations with PTVD (p < 0.05) and demonstrated strong effect sizes (Cohen’s d = −2.95, −1.13, −0.60). Their diagnostic relevance was further supported by post hoc power analyses (iCPP: 1.00; aCPP: 0.998). Conclusions: SHAP-supported machine learning models offer an objective and clinically meaningful approach for evaluating PTVD. The identified features may serve as potential biomarkers to guide individualized voice therapy decisions during the early postoperative period.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/diagnostics15162065

Cites

93 of 93 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Machine learning workflow diagram (LASSO: Least Absolute Shrinkage and Selection Operator; SVM: Support Vector Machine; RBF: Radial Basis Function; XAI: explainable artificial intelligence; SHAP: SHapley Additive exPlanations; ROC: receiver operating characteristic).
Figure 1 Machine learning workflow diagram (LASSO: Least Absolute Shrinkage and Selection Operator; SVM: Support Vector Machine; RBF: Radial Basis Function; XAI: explainable artificial intelligence; SHAP: SHapley Additive exPlanations; ROC: receiver operating characteristic).CC BY · PubMed Central
Figure 2 Heatmap of most frequently contributing features across all configurations.
Figure 2 Heatmap of most frequently contributing features across all configurations.CC BY · PubMed Central
Figure 3 Comparison of SHAP contributions between training and test sets for five prominent acoustic features. (A) SHAP value distribution scatter plot for aHNR showing close alignment of points to the diagonal, indicating similar contribution levels in both training and test sets; (B) SHAP value distribution scatter plot for aCPP showing close alignment of points to the diagonal, indicating similar contribution levels in both training and test sets; (C) SHAP value distribution scatter plot for iCPP showing close alignment of points to the diagonal, indicating similar contribution levels in both training and test sets; (D) SHAP value distribution scatter plot for aSpectralCentroid showing noticeable variance in the direction and magnitude of SHAP values, suggesting limited generalizability; (E) SHAP value distribution scatter plot for iSpectralCentroid showing noticeable variance in the direction and magnitude of SHAP values, suggesting limited generalizability.
Figure 3 Comparison of SHAP contributions between training and test sets for five prominent acoustic features. (A) SHAP value distribution scatter plot for aHNR showing close alignment of points to the diagonal, indicating similar contribution levels in both training and test sets; (B) SHAP value distribution scatter plot for aCPP showing close alignment of points to the diagonal, indicating similar contribution levels in both training and test sets; (C) SHAP value distribution scatter plot for iCPP showing close alignment of points to the diagonal, indicating similar contribution levels in both training and test sets; (D) SHAP value distribution scatter plot for aSpectralCentroid showing noticeable variance in the direction and magnitude of SHAP values, suggesting limited generalizability; (E) SHAP value distribution scatter plot for iSpectralCentroid showing noticeable variance in the direction and magnitude of SHAP values, suggesting limited generalizability.CC BY · PubMed Central
Figure 4 Violin plots of key SHAP-based acoustic features (iCPP, aCPP, aHNR) by PTVD group. PTVD participants show notably reduced delta values in iCPP and aCPP compared to non-PTVD participants. Each plot includes the p-value and Cohen’s d effect size annotation above the violins.
Figure 4 Violin plots of key SHAP-based acoustic features (iCPP, aCPP, aHNR) by PTVD group. PTVD participants show notably reduced delta values in iCPP and aCPP compared to non-PTVD participants. Each plot includes the p-value and Cohen’s d effect size annotation above the violins.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 3

  1. Putri, 2024, Analysis of Noise Removal Performance in Speech Signals through Comparison of Median Filter, Low FIR Filter, and Butterworth Filter: Simulation and Evaluation, Ultim. Comput. J. Sist. Komput., 16, 56
  2. Peeters, 2004, A large set of audio features for sound description (similarity and classification) in the CUIDADO project, Cuid. Ist Proj. Rep., 54, 1
  3. Vapnik, V. (2013). The Nature of Statistical Learning Theory, Springer Science & Business Media.

Cited by

See all 7

Metadata

Authors
Celepli, S., Bigat, I., Karakas, B., Tezcan, H. M., Yar, M. D., Celepli, P., Aksahin, M. F., Hancerliogullari, O., Yilmaz, Y. F., & Erogul, O.
Affiliations
  • TOBB University of Economics and Technology (TR)
  • Department of Biomedical Engineering, Graduate School of Science and Technology, TOBB University of Economics and Technology, Ankara 06510, Turkey
  • Department of General Surgery, Gulhane Training and Research Hospital, Ankara 06010, Turkey
  • Gülhane Eğitim ve Araştırma Hastanesi (TR)
  • Department of Pathology, Gulhane Training and Research Hospital, Ankara 06010, Turkey
  • Department of Electrical and Electronic Engineering, Gazi University, Ankara 06560, Turkey
  • Department of Otorhinolaryngology, Gulhane Training and Research Hospital, Ankara 06010, Turkey
Corresponding author
Celepli, Salih
Venue
Diagnostics
Publisher
MDPI AG
Published
2025-08-18
Volume
15
Issue
16
Pages
2065
ISSN
2075-4418 (ISSN-L)
OpenAlex
W4413283240
Cited by
6
References
92
Citations by year
2026: 5 · 2025: 1
FWCI
3.86
Type
journal article
Language
en
Open access
Locations
  • Diagnostics (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Health Sciences › Health Professions › Speech and Hearing › Voice and Speech Disorders
  • Health Sciences › Medicine › Surgery › Thyroid and Parathyroid Surgery
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
Keywords
acoustic biomarkers, SHAP, explainable machine learning, cepstral features, spectral properties, ACPP, voice therapy, thyroidectomy, early diagnosis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.