Automated detection of glottic laryngeal carcinoma in laryngoscopic images from a multicentre database using a convolutional neural network.
Clin Otolaryngol
; 48(3): 436-441, 2023 05.
Article
em En
| MEDLINE
| ID: mdl-36624555
ABSTRACT
OBJECTIVE:
Little is known about the efficacy of using artificial intelligence (AI) to identify laryngeal carcinoma from images of vocal lesions taken in different hospitals with multiple laryngoscope systems. This multicentre study aimed to establish an AI system and provide a reliable auxiliary tool to screen for laryngeal carcinoma. STUDYDESIGN:
Multicentre case-control study.SETTING:
Six tertiary care centres.PARTICIPANTS:
Laryngoscopy images were collected from 2179 patients with vocal fold lesions. OUTCOMEMEASURES:
An automatic detection system of laryngeal carcinoma was established and used to distinguish malignant and benign vocal lesions in 2179 laryngoscopy images acquired from 6 hospitals with 5 types of laryngoscopy systems. Pathological examination was the gold standard for identifying malignant and benign vocal lesions.RESULTS:
Out of 89 cases in the malignant group, the classifier was able to correctly identify laryngeal carcinoma in 66 patients (74.16%, sensitivity). Out of 640 cases in the benign group, the classifier was able to accurately assess the laryngeal lesion in 503 cases (78.59%, specificity). Furthermore, the region-based convolutional neural network (R-CNN) classifier achieved an overall accuracy of 78.05%, with a 95.63% negative predictive value and a 32.51% positive predictive value for the testing data set.CONCLUSION:
This automatic diagnostic system has the potential to assist clinical laryngeal carcinoma diagnosis which may improve and standardise the diagnostic capacity of laryngologists using different laryngoscopes.Palavras-chave
Texto completo:
1
Coleções:
01-internacional
Base de dados:
MEDLINE
Assunto principal:
Prega Vocal
/
Carcinoma
/
Neoplasias Laríngeas
/
Laringoscopia
Tipo de estudo:
Clinical_trials
/
Diagnostic_studies
/
Observational_studies
/
Prognostic_studies
/
Risk_factors_studies
Limite:
Humans
Idioma:
En
Ano de publicação:
2023
Tipo de documento:
Article