ABSTRACT
Iridology is a fairly new scientific discipline interested in the study of the beauty-contact areas of the body that are termed the iris. It can even assess problems with a variety of biological activities. Since the nervous system and brain connect the body’s tissues and organs, the condition of each organ may be immediately revealed through the iris, and in this way, the cardia characteristics were retrieved through eye and feature expression and image pre-processing methods. In the system proposed, Explainable AI is now combined with machine learning techniques like K nearest neighbor and the random forest algorithm. If they unite, they will feel even stronger, more efficient, and self-reliant. With this component, estimating the probability of coronary artery disease becomes possible.
CONCLUSION AND FUTURE WORK
To sum up, the use of random forest and K-nearest neighbors (KNN) alongside the iris dataset offers valuable insights into predicting cardiac disease. Both models demonstrate strengths in heart disease prediction, with random forest providing robustness through ensembles and KNN relying on proximity for classification. The selection between the two models depends on specific needs, and further adjustments could improve predictive accuracy. Furthermore, the use of Explainable AI methods enhances model interpretability, highlighting the crucial features influencing predictions. This study sets a strong foundation for informed decision-making when applying machine learning methods to forecast heart disease. Future research exploring various organs, such as brain tumors and kidney deficiencies, holds significant potential for advancements.
