Showing 2 results for Medical Image Processing
Sahand Shahalinejad, Atefeh Bahador, Ali Niapour,
Volume 12, Issue 1 (11-2022)
Abstract
Background and Aim: Noise reduction in medical images is important. Excessive distortion in medical images reduces the accuracy of diagnosis of various diseases or structures. Violet conversion and filtering are among the most widely used methods for reducing noise in medical images. The aim of this study was to compare noise reduction using filtration (low-pass, mid-pass, and high-pass filters) as well as violet conversion from MRI images.
Methods: In this study, using MATLAB software, noisy MRI data were entered into the program environment and each of the proposed algorithms including filtering (low-pass, mid-pass, and high-pass filters) as well as violet conversion were implemented separately on images. And the ideal output was obtained due to the nature of the noise.
Results: The results obtained from the proposed violet and filtering methods were compared and analyzed. The signal-to-noise ratio (SNR) of all filters used and the violet conversion displayed a value above 30 dB. Violet conversion for selected images has a higher SNR value, and in some images, this difference is more than 40 dB. According to the images and relative PSNR values, among all the studied methods, the best dehumidification is when the CWT method was used. In this case, the PSNR is high and there is the most similarity between the degenerate image and the original image.
Sahand Shahalinejad, Atefeh Bahadori, Ali Niapour,
Volume 12, Issue 1 (11-2022)
Abstract
Background and Aim: Identifying and demarcating the masses and diagnosing the disease in breast tissue is a serious challenge in diagnosing this cancer. Mammography is currently the most common method to diagnose breast cancer, in which incorrectly identifying the masses can lead to misdiagnosis or sampling of breast tissue. In this study, using feature extraction in medical image processing, we tried to make a diagnosis with better accuracy than in the past.
Materials and Methods: Mammographic image features were extracted using the Harris feature extraction algorithm. Mammographic images were analyzed using Matlab2019a software and ideal outputs were obtained.
Findings: Mammographic images were pre-processed and Harris algorithm was applied. In the output of the proposed algorithm, the accuracy and speed of detection of the algorithm were higher in comparison to other routine methods.
Conclusion: The purpose of extracting the Harris property is to make the raw data more usable for future statistical processing. it is expected that in the future, feature extraction will be more accurate, and more details will be provided to the machine vision systems to identify objects in the image.
