Toward automated severe pharyngitis detection with smartphone camera using deep learning networks

Deep Learning Model and its Application for the Diagnosis of Exudative Pharyngitis

OBJECTIVES

Telemedicine is firmly established in the healthcare landscape of many countries. Acute respiratory infections are the 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 images of their throat to a web application. This study aimed to develop a deep learning model for the automated diagnosis of exudative pharyngitis. Thereafter, the model will be deployed online.

METHODS

We used 343 throat images (139 with exudative pharyngitis and 204 without pharyngitis) in the study. ImageDataGenerator was used to augment the training data. The convolutional 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 training loss 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 can diagnose exudative pharyngitis. Our model was able to achieve the highest accuracy, at 95.5%, out of all previous studies that used machine learning for the diagnosis of exudative pharyngitis. We have deployed the model on a web application that can be used to augment the doctor's diagnosis of exudative pharyngitis.

Treatment of recurrent acute tonsillitis-a systematic review and clinical practice recommendations

Background

There is an ongoing debate on the indications for tonsil surgery in both children and adults with recurrent acute tonsillitis. The aim is to provide practical recommendations for diagnostics and treatment for recurrent acute tonsillitis including evidence-based decision making for tonsillectomy.

Methods

A systematic literature search in PubMed, Embase, Web of Science, and ScienceDirect from 2014 until April 2023 resulted in 68 articles. These were the basis for the review and a comprehensive series of consensus statements on the most important diagnostics and indications for both non-surgical and surgical therapy. A consensus paper was circulated among the authors and members of the International Head and Neck Scientific Group until a final agreement was reached for all recommendations.

Results

The differentiation between sore throat and tonsillitis patient episodes is mostly not feasible and hence is not relevant for diagnostic decision making. Diagnostics of a tonsillitis/sore throat episode should always include a classification with a scoring system (Centor, McIssac, FeverPAIN score) to estimate the probability of a bacterial tonsillitis, mainly due to group A streptococcus (GAS). In ambiguous cases, a point-of-care test GAS swab test is helpful. Consecutive counting of the tonsillitis/sore throat episodes is important. In addition, a specific quality of life score (Tonsillectomy Outcome Inventory 14 or Tonsil and Adenoid Health Status Instrument) should be used for each episode. Conservative treatment includes a combination of paracetamol and/or non-steroidal anti-inflammatory drugs. In case of high probability of bacterial tonsillitis, and only in such cases, especially in patients at risk, standard antibiotic treatment is initiated directly or by delayed prescription. Tonsillectomy is indicated and is highly effective if the patient has had ≥7 adequately treated episodes in the preceding year, ≥5 such episodes in each of the preceding 2 years, or ≥3 such episodes in each of the preceding 3 years. An essential part of surgery is standardized pain management because severe postoperative pain can be expected in most patients.

Conclusion

It is necessary to follow a stringent treatment algorithm for an optimal and evidence-based treatment for patients with recurrent acute tonsillitis. This will help decrease worldwide treatment variability, antibiotic overuse, and avoid ineffective tonsillectomy.

Telemedicine in Primary Practice in the Age of the COVID-19 Pandemic-Review

Background and Objectives: In the era of the COVID-19 pandemic, telemedicine, so far underestimated, has gained in value. Currently, telemedicine is not only a telephone or chat consultation, but also the possibility of the remote recording of signals (such as ECG, saturation, and heart rate) or even remote auscultation of the lungs. The objective of this review article is to present a potential role for, and disseminate knowledge of, telemedicine during the COVID-19 pandemic. Material and Methods: In order to analyze the research material in accordance with PRISMA guidelines, a systematic search of the ScienceDirect, Web of Science, and PubMed databases was conducted. Out of the total number of 363 papers identified, 22 original articles were subjected to analysis. Results: This article presents the possibilities of remote patient registration, which contributes to an improvement in remote diagnostics and diagnoses. Conclusions: Telemedicine is, although not always and not by everyone, an accepted form of providing medical services. It cannot replace direct patient-doctor contact, but it can undoubtedly contribute to accelerating diagnoses and improving their quality at a distance.

Convolutional neural network-based vocal cord tumor classification technique for home-based self-prescreening purpose

BACKGROUND

In this study, we proposed a deep learning technique that can simultaneously detect suspicious positions of benign vocal cord tumors in laparoscopic images and classify the types of tumors into cysts, granulomas, leukoplakia, nodules and polyps. This technique is useful for simplified home-based self-prescreening purposes to detect the generation of tumors around the vocal cord early in the benign stage.

RESULTS

We implemented four convolutional neural network (CNN) models (two Mask R-CNNs, Yolo V4, and a single-shot detector) that were trained, validated and tested using 2183 laryngoscopic images. The experimental results demonstrated that among the four applied models, Yolo V4 showed the highest F1-score for all tumor types (0.7664, cyst; 0.9875, granuloma; 0.8214, leukoplakia; 0.8119, nodule; and 0.8271, polyp). The model with the lowest false-negative rate was different for each tumor type (Yolo V4 for cysts/granulomas and Mask R-CNN for leukoplakia/nodules/polyps). In addition, the embedded-operated Yolo V4 model showed an approximately equivalent F1-score (0.8529) to that of the computer-operated Yolo-4 model (0.8683).

CONCLUSIONS

Based on these results, we conclude that the proposed deep-learning-based home screening techniques have the potential to aid in the early detection of tumors around the vocal cord and can improve the long-term survival of patients with vocal cord tumors.

SMD-YOLO: An efficient and lightweight detection method for mask wearing status during the COVID-19 pandemic

BACKGROUND AND OBJECTIVE

At present, the COVID-19 epidemic is still spreading worldwide and wearing a mask in public areas is an effective way to prevent the spread of the respiratory virus. Although there are many deep learning methods used for detecting the face masks, there are few lightweight detectors having a good effect on small or medium-size face masks detection in the complicated environments.

METHODS

In this work we propose an efficient and lightweight detection method based on YOLOv4-tiny, and a face mask detection and monitoring system for mask wearing status. Two feasible improvement strategies, network structure optimization and K-means++ clustering algorithm, are utilized for improving the detection accuracy on the premise of ensuring the real-time face masks recognition. Particularly, the improved residual module and cross fusion module are designed to aim at extracting the features of small or medium-size targets effectively. Moreover, the enhanced dual attention mechanism and the improved spatial pyramid pooling module are employed for merging sufficiently the deep and shallow semantic information and expanding the receptive field. Afterwards, the detection accuracy is compensated through the combination of activation functions. Finally, the depthwise separable convolution module is used to reduce the quantity of parameters and improve the detection efficiency. Our proposed detector is evaluated on a public face mask dataset, and an ablation experiment is also provided to verify the effectiveness of our proposed model, which is compared with the state-of-the-art (SOTA) models as well.

RESULTS

Our proposed detector increases the AP (average precision) values in each category of the public face mask dataset compared with the original YOLOv4-tiny. The mAP (mean average precision) is improved by 4.56% and the speed reaches 92.81 FPS. Meanwhile, the quantity of parameters and the FLOPs (floating-point operations) are reduced by 1/3, 16.48%, respectively.

CONCLUSIONS

The proposed detector achieves better overall detection performance compared with other SOTA detectors for real-time mask detection, demonstrated the superiority with both theoretical value and practical significance. The developed system also brings greater flexibility to the application of face mask detection in hospitals, campuses, communities, etc.

Application of generative adversarial networks (GAN) for ophthalmology image domains: a survey

BACKGROUND

Recent advances in deep learning techniques have led to improved diagnostic abilities in ophthalmology. A generative adversarial network (GAN), which consists of two competing types of deep neural networks, including a generator and a discriminator, has demonstrated remarkable performance in image synthesis and image-to-image translation. The adoption of GAN for medical imaging is increasing for image generation and translation, but it is not familiar to researchers in the field of ophthalmology. In this work, we present a literature review on the application of GAN in ophthalmology image domains to discuss important contributions and to identify potential future research directions.

METHODS

We performed a survey on studies using GAN published before June 2021 only, and we introduced various applications of GAN in ophthalmology image domains. The search identified 48 peer-reviewed papers in the final review. The type of GAN used in the analysis, task, imaging domain, and the outcome were collected to verify the usefulness of the GAN.

RESULTS

In ophthalmology image domains, GAN can perform segmentation, data augmentation, denoising, domain transfer, super-resolution, post-intervention prediction, and feature extraction. GAN techniques have established an extension of datasets and modalities in ophthalmology. GAN has several limitations, such as mode collapse, spatial deformities, unintended changes, and the generation of high-frequency noises and artifacts of checkerboard patterns.

CONCLUSIONS

The use of GAN has benefited the various tasks in ophthalmology image domains. Based on our observations, the adoption of GAN in ophthalmology is still in a very early stage of clinical validation compared with deep learning classification techniques because several problems need to be overcome for practical use. However, the proper selection of the GAN technique and statistical modeling of ocular imaging will greatly improve the performance of each image analysis. Finally, this survey would enable researchers to access the appropriate GAN technique to maximize the potential of ophthalmology datasets for deep learning research.

Adopting low-shot deep learning for the detection of conjunctival melanoma using ocular surface images

BACKGROUND AND OBJECTIVE

The purpose of the present study was to investigate low-shot deep learning models applied to conjunctival melanoma detection using a small dataset with ocular surface images.

METHODS

A dataset was composed of anonymized images of four classes; conjunctival melanoma (136), nevus or melanosis (93), pterygium (75), and normal conjunctiva (94). Before training involving conventional deep learning models, two generative adversarial networks (GANs) were constructed to augment the training dataset for low-shot learning. The collected data were randomly divided into training (70%), validation (10%), and test (20%) datasets. Moreover, 3D melanoma phantoms were designed to build an external validation set using a smartphone. The GoogleNet, InceptionV3, NASNet, ResNet50, and MobileNetV2 architectures were trained through transfer learning and validated using the test and external validation datasets.

RESULTS

The deep learning model demonstrated a significant improvement in the classification accuracy of conjunctival lesions using synthetic images generated by the GAN models. MobileNetV2 with GAN-based augmentation displayed the highest accuracy of 87.5% in the four-class classification and 97.2% in the binary classification for the detection of conjunctival melanoma. It showed an accuracy of 94.0% using 3D melanoma phantom images captured using a smartphone camera.

CONCLUSIONS

The present study described a low-shot deep learning model that can detect conjunctival melanomas using ocular surface images. To the best of our knowledge, this study is the first to develop a deep learning model to detect conjunctival melanoma using a digital imaging device such as smartphone camera.

Feasibility study to improve deep learning in OCT diagnosis of rare retinal diseases with few-shot classification

Deep learning (DL) has been successfully applied to the diagnosis of ophthalmic diseases. However, rare diseases are commonly neglected due to insufficient data. Here, we demonstrate that few-shot learning (FSL) using a generative adversarial network (GAN) can improve the applicability of DL in the optical coherence tomography (OCT) diagnosis of rare diseases. Four major classes with a large number of datasets and five rare disease classes with a few-shot dataset are included in this study. Before training the classifier, we constructed GAN models to generate pathological OCT images of each rare disease from normal OCT images. The Inception-v3 architecture was trained using an augmented training dataset, and the final model was validated using an independent test dataset. The synthetic images helped in the extraction of the characteristic features of each rare disease. The proposed DL model demonstrated a significant improvement in the accuracy of the OCT diagnosis of rare retinal diseases and outperformed the traditional DL models, Siamese network, and prototypical network. By increasing the accuracy of diagnosing rare retinal diseases through FSL, clinicians can avoid neglecting rare diseases with DL assistance, thereby reducing diagnosis delay and patient burden.