IIT Jodhpur researchers find an automated detection method for hypertensive retinopathy

The proposed system obtained average accuracy, sensitivity, specificity, and precision of 98.44%, 98.44%, 98.44%, 98.44%, and 98.48%, respectively

0
48
Image for the representational purpose only.
New Delhi: Indian Institute of Technology Jodhpur researchers have developed a novel Computer-aided diagnosis (deep learning) based approach for Hypertensive Retinopathy (HR) detection.
Computer-aided diagnosis (CAD) plays an important role in the early identification of HR with high diagnostic accuracy, which is cheap, portable, non-invasive, time-efficient and demands fewer resources.
The work has been published in Biomedical Signal Processing and Control Journal and a detailed review study for Hypertensive Retinopathy Diagnosis has also been published in Computer Methods and Programs in Biomedicine and has been co-authored by Ms. Supriya Suman, PhD research Scholar, Interdisciplinary Research Division: Smart Health Care, IIT Jodhpur, Dr. Anil Kumar Tiwari, Professor, Department of Electrical Engineering, IIT Jodhpur, Mr. Tejas Ingale, 4th Year – BTech Student, Department of Mechanical Engineering, IIT Jodhpur, and Dr. Kuldeep Singh, Professor, Department of Paediatrics, AIIMS Jodhpur.
Hypertension is a prevalent condition that affects both elderly and young people, resulting in increased morbidity and mortality. According to a recent World Health Organization (WHO) survey, around 1.13 billion people in developing nations suffer from hypertension. Hypertensive Retinopathy (HR) is a retinal disease caused by hypertension for a prolonged period of time. HR is a biomarker for a variety of diseases, including cardiovascular risks and target-organ damage. Early detection and treatment can prevent blindness as it may even lead to loss of vision. In the initial stage of this retinal disease, no visible signs are found. Through regular screening of patients, it can only be prevented.
The proposed method will play an important role in the early identification of Hypertensive Retinopathy with high diagnostic accuracy. Identifying early-stage HR patients will help in the timely and proper treatment before it starts damaging several other organs of the body, including the kidney, heart, brain, and eyes. The method obtained average accuracy, sensitivity, specificity, and precision of 98.44%, 98.44%, 98.44%, 98.44%, and 98.48%, respectively.
Talking about the significance of the research, Dr. Anil Kumar Tiwari, Professor, Department of Electrical Engineering, IIT Jodhpur, said, “The proposed system will not only provide a second opinion to the doctors but can provide better diagnosis accuracy. This system can be used specially in emergency treatment by the physicians, neurologists, and ophthalmologists.”
The future scope of this research includes the severity grading of HR and several other retinal disease diagnoses including papilledema, glaucoma, Age related Macular Degeneration (AMD), and other non-ocular diseases.
The research was done in collaboration with All India Institute of Medical Sciences (AIIMS), Jodhpur and was funded by the Ministry of Education, Government of India. The Interdisciplinary Program in Smart Health Care program by IIT Jodhpur-AIIMS Jodhpur facilitated collaborative research, enabling interdisciplinary teams of doctors and engineers to unite in the pursuit of technological advancements in healthcare. This synergy aims to bring about positive impacts on a broader scale, particularly benefiting patients in remote areas with limited access to healthcare facilities.