In Press                   Back to the articles list | Back to browse issues page

XML Print

A Practical Hybrid Framework for High-Quality Enhancement of Multi-Modality Medical Images
Nashaat Hassan , Tarik R. Al-Khateeb
Abstract:   (14 Views)
Medical imaging quality is crucial in accurate diagnosis and computer aided diagnosis (CAD). But medical images obtained through X-rays, CT scan, and MRI may contain poor contrast, noise, uneven lighting conditions, and poor anatomical visibility that may cause errors during diagnosis and further analysis of the medical image. This research work introduces a Hybrid Contrast and Detail Enhancement Framework (HCDEF) in order to enhance medical image quality. HCDEF incorporates both the traditional image processing techniques and the artificial intelligence technique in order to enhance medical image quality while preserving important anatomical structures. The enhancement process involves a number of stages such as Contrast Limited Adaptive Histogram Equalization (CLAHE), Gamma Correction, Haze Removal, Saturation Correction, Unsharp Masking, and Adaptive Luminance Control. A light-weight CNN is used for the estimation of spatially varying parameters which help adaptively enhance an image without adding any extra features to it. The proposed approach has been tested using x-ray, CT, and MRI images and has been compared to traditional enhancement techniques and other deep learning-based approaches like GANs and autoencoders. Experimental analysis showed good performance with a contrast of 64.09, entropy 6.81, standard deviation 64.09, and energy 0.029.
Keywords: Medical image enhancement, contrast improvement, hybrid preprocessing, classical image processing methods, AI-driven adaptive modules, diagnostic imaging quality.
Full-Text [PDF 1862 kb]   (2 Downloads)    
Type of Study: Research Paper | Subject: Biomedical Signal Processing
Received: 2025/11/22 | Revised: 2026/08/26 | Accepted: 2026/06/26
Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Creative Commons License
© 2022 by the authors. Licensee IUST, Tehran, Iran. This is an open access journal distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license.