This paper presents a novel head-mounted point-of-care diagnostic system for detection and continuous monitoring of Age-related Macular Degeneration (AMD). This wearable embedded open-source platform enables accurate monitoring of AMD by taking advantage of multiple standard graphical interface techniques such as Amsler Grid, Threshold Amsler Grid, Macular Computerized Psychophysical Test and Preferential Hyperacuity Perimeter (PHP). Here, we describe the proposed multi-Grid or so-called NGRID software and elaborate on the hardware prototype. This prototype includes a commercially available Oculus HMD incorporated with a single board computer. As the first step towards a fully integrated wearable system, this paper successfully proves the functionality of head-mounted graphical interface device ready for a live demonstration. Participants can experience this device and take a 10-minute AMD eye-exam. Furthermore, NGRID has been approved and permitted for an in-hospital clinical trial.

This paper presents a new mathematical model along with a measurement platform for accurate detection and monitoring of various visual distortions (VD) caused by macular disorders such as central serous chorioretinopathy (CSR) and age-related macular degeneration (AMD). This platform projects a series of graphical patterns on the patient’s retina and calculates the severity of VDs accordingly. The accuracy of this technique relies on the accurate detection of distorted lines by the patient. We also propose a simple mathematical model to evaluate the VD created by CSR. The model is used as a control for the test results achieved from the proposed platform. The proposed platform consists of the required hardware and software for the generation and projection of patterns along with the collection and processing of patients against their standard optical coherence tomography (OCT) images. Based on these results, the OCT images agree with the VD test results, and the proposed platform can be used as an alternative home monitoring method for various macular disorders.

Recent research has identified salivary oral polymorphonuclear neutrophils (oPMNs) as reliable cellular biomarkers for monitoring the progression of periodontal disease (PD). While conventional diagnostic tools such as periodontal probing and radiographic imaging are effective at detecting advanced disease stages, they lack the sensitivity required for early diagnosis. oPMNs, derived from circulating white blood cells (WBCs), transmigrate through the oral epithelium and appear in saliva with diverse morphological characteristics. Although deep learning-based WBC quantification has been extensively explored using peripheral blood datasets, the unique morphology of oPMNs necessitates the development of new image datasets and retraining of models tailored to their detection. To address these challenges, we introduce DePerio-an AI-powered diagnostic pipeline that integrates a novel oPMN isolation protocol with deep neural network (DNN) architectures. We evaluate the performance of multiple DNN models to achieve accurate detection and quantification of oPMNs. Validated against standard quantification methods for oPMNs and oral inflammatory load (OIL), DePerio achieved a detection error rate of less than 5%. In a clinical study involving 111 human saliva samples, spanning individuals from healthy to severe periodontitis cases, and successfully classified five distinct OIL levels. This robust and low-complexity platform provides a scalable and practical solution for early PD detection and longitudinal monitoring in clinical dental practice.