IJRR

International Journal of Research and Review

| Home | Current Issue | Archive | Instructions to Authors | Journals |

Year: 2025 | Month: May | Volume: 12 | Issue: 5 | Pages: 617-624

DOI: https://doi.org/10.52403/ijrr.20250564

Leveraging Deep Learning with Morphological Operators, PCA, and SVM for Enhanced Detection and Classification of Kidney Abnormalities in Medical Imaging

Mahdi Koohi1, Hamid Reza Tavakoli2

1Electronic Engineering Department, STU, Tehran, Iran
2Environment and Energy, GIS Department, Science and Research Branch, Tehran, Iran

Corresponding Author: Mahdi Koohi

ABSTRACT

Magnetic Resonance Imaging (MRI) and CT (Computed Tomography) are vital tool in medical diagnostics, offering detailed visualization of internal body structures. However, challenges such as noise, intensity variations, and the complexity of tissue structures hinder accurate abnormality detection. This paper introduces a hybrid approach that combines deep learning with morphological operators, Principal Component Analysis (PCA), and Support Vector Machines (SVM) to improve feature extraction and classification of abnormalities in medical images. Morphological operations enhance image preprocessing, PCA reduces dimensionality while preserving key features, and SVM performs robust classification. A deep learning model is integrated to extract high-level spatial features, improving classification performance. Experimental results demonstrate that this combined approach enhances diagnostic accuracy and robustness This approach emphasizes the synergy between traditional image processing techniques and modern machine learning, aiming to achieve state-of-the-art performance in medical imaging diagnostics.

Keywords: Medical Image Processing, Classification, Deep Learning, Abnormalities, SVM

[PDF Full Text]