• Ramesh
• Sathiamoorthy

Cardiovascular diseases (CVDs) are a main world-wide health problem, requiring earlier diagnosis for efficient intervention. Recently, deep learning (DL) methods have been developed as robust tools for automatically diagnosing disease. This study considers leveraging DL approach for the detection of CVDs employing retinal fundus images, a non-invasive and simply available imaging modality. In this study, we present an innovative Grasshopper Optimization Algorithm with Deep Learning based Blood Vessel Segmentation and Classification (GOADL-BVSC) model for grading the CVD. This analysis suggests a robust method for CVD detection by connecting DL methodologies, particularly utilizing DenseNet for feature extraction, Grasshopper Optimization Algorithm (GOA) for parameter tuning, and Deep Belief Network (DBN) for classification. Retinal fundus images provide a useful resources for evaluating cardiovascular conditions, giving a non-invasive and quickly accessible process of detecting CVDs. DenseNet, a deep neural network (DNN) framework known for its extensive feature- representations, has been employed for extracting useful features from these images. GOA, stimulated by the foraged behavior of grasshoppers, is exploited to fine-tune the hyperparameters of the method, improving its performance. The DBN classification model is trained for differentiating among normal and CVD-affected retinal images depends on the extracted features. GOA iteratively enhances the hyperparameters of the DBN, confirming that the system attains its maximum accuracy capability. The simulation outcomes represented the excellent outcomes of the GOADL-BVSC method over other existing models with regard to different levels.

Cardiovascular diseases; Retinal fundus images; Blood vessel segmentation; Grasshopper Optimization Algorithm; Deep learning

"A Deep Learning Model for Blood Vessel Segmentation and Classification for Cardiovascular Diseases Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 10, page no.c115-c128, October-2023, Available :http://www.jetir.org/papers/JETIR2310214.pdf

"A Deep Learning Model for Blood Vessel Segmentation and Classification for Cardiovascular Diseases Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 10, page no. ppc115-c128, October-2023, Available at : http://www.jetir.org/papers/JETIR2310214.pdf