A hybrid approach for enhancement of abdominal CT images. (September 2022)

• Record Type: Journal Article
• Authors: Kaur, Ravinder; Juneja, Mamta
• Is Part Of: Computers & electrical engineering. Volume 102(2022)
• Publication Date: 2022-09
• Subjects: DICOM -- CT -- Window level/center -- Medical imaging -- Visualization; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.108291
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Image quality plays an essential role for performing precise clinical diagnosis. CT images corrupted with image artifacts hampers the diagnosis accuracy. A hybrid method has been suggested for performing CT image enhancement. The results suggested that the proposed method outperforms existing techniques. Abstract: In this technological era, digital imaging is integrated into every aspect of our life, and it has a predominant and essential role in medical investigations. Different types of diseases are being diagnosed using Computed Tomography (CT) images, acquired with varying bit depths, and kept in conventional Digital Imaging and Communications in Medicine (DICOM) format. However, one of the key challenges faced by the practitioners experimenting in the field of biomedical image analysis is that the images acquired are of low contrast. This research work aims to propose a method centered on the windowing technique and anisotropic diffusion filter to enhance the contrast of medical images for further analysis. The suggested approach does adjustments of contrast and brightness using the information extracted from the DICOM header of considered abdominal CT images to efficiently visualize the details for further processing. Furthermore, we have analyzed and compared its performance with distinct notable techniques which have been typically used for the enhancement of the medical image. The recorded outcomes revealed that the suggested algorithm gives better results with average values of Peak Signal to Noise Ratio (PSNR), Absolute Mean Brightness Error (AMBE), and Feature Similarity (FSIM) index equivalent to 27.71 dB, 8.38, and 0.96, respectively. Further, the average value of the Universal Image Quality (UIQ) index and Edge Content (EC) is equivalent to 0.83 and 9.37, respectively. Based on the experimental results performed on 15, 000 images of a real human CT dataset, we have observed that the suggested algorithm gives much better results from subjective and objective perspectives. Graphical abstract: Figure: (a) Original CT image viewed in radiant DICOM viewer (b) The Same Image when displayed in Matlab. Image, graphical abstract