Objective: This study investigated whether artificial intelligence (AI) models combining voice signals, demographics, and structured medical records can detect glottic neoplasm from benign voice disorders.
Methods: We used a primary dataset containing 2-3 s of vowel "ah", demographics, and 26 items of structured medical records (e.g., symptoms, comorbidity, smoking and alcohol consumption, vocal demand) from 60 patients with pathology-proved glottic neoplasm (i.e., squamous cell carcinoma, carcinoma in situ, and dysplasia) and 1940 patients with benign voice disorders. The validation dataset comprised data from 23 patients with glottic neoplasm and 1331 patients with benign disorders. The AI model combined convolutional neural networks, gated recurrent units, and attention layers. We used 10-fold cross-validation (training-validation-testing: 8-1-1) and preserved the percentage between neoplasm and benign disorders in each fold.
Results: Results from the AI model using voice signals reached an area under the ROC curve (AUC) value of 0.631, and additional demographics increased this to 0.807. The highest AUC of 0.878 was achieved when combining voice, demographics, and medical records (sensitivity: 0.783, specificity: 0.816, accuracy: 0.815). External validation yielded an AUC value of 0.785 (voice plus demographics; sensitivity: 0.739, specificity: 0.745, accuracy: 0.745). Subanalysis showed that AI had higher sensitivity but lower specificity than human assessment (p < 0.01). The accuracy of AI detection with additional medical records was comparable with human assessment (82% vs. 83%, p = 0.78).
Conclusions: Voice signal alone was insufficient for AI differentiation between glottic neoplasm and benign voice disorders, but additional demographics and medical records notably improved AI performance and approximated the prediction accuracy of humans.
Level Of Evidence: NA Laryngoscope, 134:4585-4592, 2024.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1002/lary.31563 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
[Clinical treatment strategy for pT3N0 laryngeal squamous cell carcinoma].
Lin Chuang Er Bi Yan Hou Tou Jing Wai Ke Za Zhi
January 2025
To investigate optimal treatment strategy for pT3N0 laryngeal squamous cell carcinoma（SCC）. A retrospective study of 150 patients with pT3N0 laryngeal SCC treated in the First Affiliated Hospital of Chongqing Medical University was performed. The efficacies of partial laryngectomy and total laryngectomy, as well as surgery alone and postoperative radiotherapy were evaluated.
View Article and Find Full Text PDFMachine learning in personalized laryngeal cancer management: insights into clinical characteristics, therapeutic options, and survival predictions.
Eur Arch Otorhinolaryngol
December 2024
King Hussein Cancer Center, Medical Oncology Department, Amman, Jordan.
Purpose: Over the last 40 years, there has been an unusual trend where, even though there are more varied treatments, survival rates have not improved much. Our study used survival analysis and machine learning (ML) to investigate this odd situation and to improve prediction methods for treating non-metastatic LSCC.
Methods: The surveillance, epidemiology and end results (SEER) database provided the data used for this study's analysis.
[Study on the prevention of vocal cord adhesion after operation of T1b glottic carcinoma by repairing the wound with transposition of ventricular mucosal flap].
Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi
December 2024
Department of Otorhinolaryngology, the Second Affiliated Hospital Zhejiang University School of Medicine, Hangzhou310009, China.
To explore the feasibility of one-stage repair and reconstruction of glottic area wounds with the ventricular mucosal flap to prevent postoperative vocal cord adhesion in patients with T1b glottic laryngeal cancer. This case series study involved the research and analysis of clinical data of 12 patients with T1b glottic laryngeal cancer treated in the Department of Otorhinolaryngology, the Second Affiliated Hospital of Zhejiang University School of Medicine from January 2021 to June 2023. All patients were male, aged 50-85 years (median age 64.
View Article and Find Full Text PDFAnaesthetic management of a breast cancer patient with cardiac tamponade and bilateral vocal cord paralysis.
BMJ Case Rep
November 2024
Department of Anesthesiology, Philippine General Hospital, University of the Philippines Manila, Manila, Metro Manila, Philippines.
Article Synopsis
- Metastatic breast cancer can lead to unusual complications, such as cardiac tamponade and bilateral vocal cord paralysis, as shown in a case of an elderly patient who had prior surgery for breast cancer.
- The patient underwent multiple surgical procedures, including thoracotomy and tracheostomy, due to these complications, which posed significant challenges for anaesthesia management.
- This case underscores the seriousness of advanced breast cancer and the need for careful consideration of anaesthetic strategies in similar situations.
[Analysis of vocal characteristics in hypopharyngeal cancer patients with vocal cord dysfunction].
Lin Chuang Er Bi Yan Hou Tou Jing Wai Ke Za Zhi
November 2024
Article Synopsis
- The study analyzes the acoustic features of patients with hypopharyngeal cancer who also have vocal cord dysfunction, comparing two groups based on vocal cord activity: one with dysfunction and the other with normal function.
- Significant findings show that lesions in the experimental group are more common on the inner wall of the piriform fossa, with a statistically meaningful difference, while vocal parameters like sound intensity showed no significant difference between both groups.
- Stroboscopic laryngoscopy results indicated higher rates of glottic insufficiency and asymmetric arytenoid cartilage in the experimental group, although both groups maintained normal mucosal wave patterns on the vocal cords.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!