KMID : 1100520240300010042
|
|
Healthcare Informatics Research 2024 Volume.30 No. 1 p.42 ~ p.48
|
|
Deep Learning Model and its Application for the Diagnosis of Exudative Pharyngitis
|
|
Seo Yi Chng
Paul Jie Wen Tern Matthew Rui Xian Kan Lionel Tim-Ee Cheng
|
|
Abstract
|
|
|
Objectives: Telemedicine is firmly established in the healthcare landscape of many countries. Acute respiratory infections arethe most common reason for telemedicine consultations. A throat examination is important for diagnosing bacterial pharyngitis,but this is challenging for doctors during a telemedicine consultation. A solution could be for patients to upload imagesof their throat to a web application. This study aimed to develop a deep learning model for the automated diagnosis ofexudative pharyngitis. Thereafter, the model will be deployed online.
Methods: We used 343 throat images (139 with exudativepharyngitis and 204 without pharyngitis) in the study. ImageDataGenerator was used to augment the training data. Theconvolutional neural network models of MobileNetV3, ResNet50, and EfficientNetB0 were implemented to train the dataset,with hyperparameter tuning.
Results: All three models were trained successfully; with successive epochs, the loss and trainingloss decreased, and accuracy and training accuracy increased. The EfficientNetB0 model achieved the highest accuracy(95.5%), compared to MobileNetV3 (82.1%) and ResNet50 (88.1%). The EfficientNetB0 model also achieved high precision(1.00), recall (0.89) and F1-score (0.94).
Conclusions: We trained a deep learning model based on EfficientNetB0 that candiagnose exudative pharyngitis. Our model was able to achieve the highest accuracy, at 95.5%, out of all previous studies thatused machine learning for the diagnosis of exudative pharyngitis. We have deployed the model on a web application that canbe used to augment the doctor¡¯s diagnosis of exudative pharyngitis.
|
|
KEYWORD
|
|
Artificial Intelligence, Deep Learning, Diagnosis, Pharyngitis, Telemedicine
|
|
FullTexts / Linksout information
|
|
|
|
Listed journal information
|
|
|
|