Image Processing and CNN-Based Anxiety, Depression, and Hypertension Identification through Iridology: A Systematic Review

The extensiveness of stress-related physiological and other psychological conditions such as anxiety and depression requires new diagnostic approaches. This research will aim at finding out the efficiency of using iris analysis in diagnosing such conditions as a complementary to the current methods that are either invasive or rely on the patient’s perception. Based on the interviews with ophthalmologists and psychologists, it was discovered that the movements of the iris and pupil controlled by the autonomic nervous system may contain information about stress and mental health disorders. Ophthalmologists noted that iris could be used as a bio-marker while clinical psychologists noted that speech based assessment was still the common practice in the assessment of mental health. Although iris analysis cannot be used as the sole diagnostic tool, this research shows that it could be a valuable adjunct when combined with standard approaches. The proposed system will employ high quality of iris image capture and the state of the art machine learning especially Convolutional Neural Networks (CNNs) to detect abnormalities concerning psychological and physiological disorders. The integration of image processing with real time data analysis is a less invasive and may be earlier diagnostic tool than the conventional methods in non-ocular diseases and enhances the accuracy and speed of diagnosis. This approach is meant to encourage a holistic practice of medicine.