Support vector machine and deep-learning object detection for localisation of hard exudates

Support Vector Machine Outperforms Other Machine Learning Models in Early Diagnosis of Dengue Using Routine Clinical Data

Background: There is a dire need for the establishment of active dengue surveillance to continuously detect cases, circulating serotypes, and determine the disease burden of dengue fever (DF) in the country and region. Predicting dengue PCR results using machine learning (ML) models represents a significant advancement in pre-emptive healthcare measures. This study outlines the comprehensive process of data preprocessing, model selection, and the underlying mechanisms of each algorithm employed to accurately predict dengue PCR outcomes. Methods: We analyzed data from 300 suspected dengue patients in Islamabad and Rawalpindi, Pakistan, from August to October 2023. NS1 antigen ELISA, IgM and IgG antibody tests, and serotype-specific real-time polymerase chain reaction (RT-PCR) were used to detect the dengue virus (DENV). Representative PCR-positive samples were sequenced by Sanger sequencing to confirm the circulation of various dengue serotypes. Demographic information, serological test results, and hematological parameters were used as inputs to the ML models, with the dengue PCR result serving as the output to be predicted. The models used were logistic regression, XGBoost, LightGBM, random forest, support vector machine (SVM), and CatBoost. Results: Of the 300 patients, 184 (61.33%) were PCR positive. Among the total positive cases detected by PCR, 9 (4.89%), 171 (92.93%), and 4 (2.17%) were infected with serotypes 1, 2, and 3, respectively. A total of 147 (79.89%) males and 37 (20.11%) females were infected, with a mean age of 33 ± 16 years. In addition, the mean platelet and leukocyte counts and the hematocrit percentages were 75,447%, 4189.02%, and 46.05%, respectively. The SVM was the best-performing ML model for predicting RT-PCR results, with 71.4% accuracy, 97.4% recall, and 71.6% precision. Hyperparameter tuning improved the recall to 100%. Conclusion: Our study documents three circulating serotypes in the capital territory of Pakistan and highlights that the SVM outperformed other models, potentially serving as a valuable tool in clinical settings to aid in the rapid diagnosis of DF.

Nonproliferative diabetic retinopathy dataset(NDRD): A database for diabetic retinopathy screening research and deep learning evaluation

OBJECTIVES

In this article, we provide a database of nonproliferative diabetes retinopathy, which focuses on early diabetes retinopathy with hard exudation, and further explore its clinical application in disease recognition.

METHODS

We collect the photos of nonproliferative diabetes retinopathy taken by Optos Panoramic 200 laser scanning ophthalmoscope, filter out the pictures with poor quality, and label the hard exudative lesions in the images under the guidance of professional medical personnel. To validate the effectiveness of the datasets, five deep learning models are used to perform learning predictions on the datasets. Furthermore, we evaluate the performance of the model using evaluation metrics.

RESULTS

Nonproliferative diabetes retinopathy is smaller than proliferative retinopathy and more difficult to identify. The existing segmentation models have poor lesion segmentation performance, while the intersection over union (IOU) value for deep lesion segmentation of models targeting small lesions can reach 66.12%, which is higher than ordinary lesion segmentation models, but there is still a lot of room for improvement.

CONCLUSION

The segmentation of small hard exudative lesions is more challenging than that of large hard exudative lesions. More targeted datasets are needed for model training. Compared with the previous diabetes retina datasets, the NDRD dataset pays more attention to micro lesions.

Prediction of Five-Year Survival Rate for Rectal Cancer Using Markov Models of Convolutional Features of RhoB Expression on Tissue Microarray

The ability to predict survival in cancer is clinically important because the finding can help patients and physicians make optimal treatment decisions. Artificial intelligence in the context of deep learning has been increasingly realized by the informatics-oriented medical community as a powerful machine-learning technology for cancer research, diagnosis, prediction, and treatment. This paper presents the combination of deep learning, data coding, and probabilistic modeling for predicting five-year survival in a cohort of patients with rectal cancer using images of RhoB expression on biopsies. Using about one-third of the patients' data for testing, the proposed approach achieved 90% prediction accuracy, which is much higher than the direct use of the best pretrained convolutional neural network (70%) and the best coupling of a pretrained model and support vector machines (70%).

Diabetic Macular Edema Detection Using End-to-End Deep Fusion Model and Anatomical Landmark Visualization on an Edge Computing Device

Purpose

Diabetic macular edema (DME) is a common cause of vision impairment and blindness in patients with diabetes. However, vision loss can be prevented by regular eye examinations during primary care. This study aimed to design an artificial intelligence (AI) system to facilitate ophthalmology referrals by physicians.

Methods

We developed an end-to-end deep fusion model for DME classification and hard exudate (HE) detection. Based on the architecture of fusion model, we also applied a dual model which included an independent classifier and object detector to perform these two tasks separately. We used 35,001 annotated fundus images from three hospitals between 2007 and 2018 in Taiwan to create a private dataset. The Private dataset, Messidor-1 and Messidor-2 were used to assess the performance of the fusion model for DME classification and HE detection. A second object detector was trained to identify anatomical landmarks (optic disc and macula). We integrated the fusion model and the anatomical landmark detector, and evaluated their performance on an edge device, a device with limited compute resources.

Results

For DME classification of our private testing dataset, Messidor-1 and Messidor-2, the area under the receiver operating characteristic curve (AUC) for the fusion model had values of 98.1, 95.2, and 95.8%, the sensitivities were 96.4, 88.7, and 87.4%, the specificities were 90.1, 90.2, and 90.2%, and the accuracies were 90.8, 90.0, and 89.9%, respectively. In addition, the AUC was not significantly different for the fusion and dual models for the three datasets (p = 0.743, 0.942, and 0.114, respectively). For HE detection, the fusion model achieved a sensitivity of 79.5%, a specificity of 87.7%, and an accuracy of 86.3% using our private testing dataset. The sensitivity of the fusion model was higher than that of the dual model (p = 0.048). For optic disc and macula detection, the second object detector achieved accuracies of 98.4% (optic disc) and 99.3% (macula). The fusion model and the anatomical landmark detector can be deployed on a portable edge device.

Conclusion

This portable AI system exhibited excellent performance for the classification of DME, and the visualization of HE and anatomical locations. It facilitates interpretability and can serve as a clinical reference for physicians. Clinically, this system could be applied to diabetic eye screening to improve the interpretation of fundus imaging in patients with DME.

Artificial Intelligence in Orthodontic Smart Application for Treatment Coaching and Its Impact on Clinical Performance of Patients Monitored with AI-TeleHealth System

BACKGROUND

Treatment of malocclusion with clear removable appliances like Invisalign^® or Spark™, require considerable higher level of patient compliance when compared to conventional fixed braces. The clinical outcomes and treatment efficiency strongly depend on the patient's discipline. Smart treatment coaching applications, like strojCHECK^® are efficient for improving patient compliance.

PURPOSE

To evaluate the impact of computerized personalized decision algorithms responding to observed and anticipated patient behavior implemented as an update of an existing clinical orthodontic application (app).

MATERIALS AND METHODS

Variables such as (1) patient app interaction, (2) patient app discipline and (3) clinical aligner tracking evaluated by artificial intelligence system (AI) system-Dental monitoring^® were observed on the set of 86 patients. Two 60-day periods were evaluated; before and after the app was updated with decision tree processes.

RESULTS

All variables showed significant improvement after the update except for the manifestation of clinical non-tracking in men, evaluated by artificial intelligence from video scans.

CONCLUSIONS

Implementation of application update including computerized decision processes can significantly enhance clinical performance of existing health care applications and improve patients' compliance. Using the algorithm with decision tree architecture could create a baseline for further machine learning optimization.