A Hybrid ArtInt Model to Predict Diabetic Retinopathy

1. Introduction

Diabetes risk and loss of sight are both directly linked to Diabetic Retinopathy (DR). The retina is the back of the eye’s protective tissue. The sensitivity to light here is extraordinary. It converts all incoming light into signals that the brain can infer. These sensory impressions allow the eye to function. DR causes fluid leakage and permanent vision loss by damaging blood vessels in the retina. Non-proliferative DR (nPDR) and proliferative DR (pDR) are the two main categories of DR. Nasopharyngeal carcinoma represents an asymptomatic or mildly symptomatic form of the disease. The retinal blood vessels get weakened and leads to the formation of nPDR. Macula Edoema can be caused by fluid leakage into the retinal blood vessels from micro aneurysms. pDR can occur when blood sugar levels fluctuate frequently over time. This condition is called proliferative DR because new blood vessels sprout from the retina's surface. When these blood vessels harden, fluid or blood can leak out of them. The end result is permanent vision loss due to retinal damage. The development of new blood vessels causes scar tissue that can eventually cause the retina to isolate from the back of the eye. The pressure inside the eye can rise due to the leakage of the liquid from an affected eye. Glaucoma can develop if the processes described above create injury to the nerves that transmit visual information. Symptoms of DR are uncommon in the earliest stages. When symptoms manifest, the disease is usually well along in its progression. Most of the time, sudden and complete blindness is the only symptom present. The major indications of DR comprise of blurred vision, appearance of floaters like colourless spots, dark threads, the blocking of the patient's vision by blotches or smudges, meagre night vision and quick loss of vision. Figure 1 indicates the difference in blood vessels of normal eye and DR affected eye.

Figure 1.

Difference Between a Normal Eye and an Eye with Retinopathy

Source: Understanding the Stages of Diabetic Retinopathy | Elman Retina Group

In recent years, the integration of artificial intelligence (AI) into healthcare has emerged as a transformative force, revolutionizing traditional medical practices. From streamlining administrative tasks to enhancing diagnostic accuracy and personalized treatment plans, AI holds immense potential to improve patient outcomes while optimizing healthcare delivery. This article inculcates the machine learning and deep leaning methodology to predict diabetic retinopathy at an earlier stage. The results obtained by the proposed method is compared with the existing traditional methods to prove its efficacy.