Parita Oza

Work place: Institute of Technology, Nirma University, Ahmedabad, Gujarat, 382481, Ahmedabad, Gujarat, India



Research Interests: Graph and Image Processing, Computer Vision, Image Processing, Medical Image Computing


Prof. Parita Oza is working as an Assistant Professor in the Computer Science and Engineering Department, Institute of Technology, Nirma University, Ahmedabad. She has pursued her MTech in Information and Communication Technology from Nirma University. She is involved in teaching courses at both undergraduate and postgraduate levels. She has published several research papers in national and international conferences and journals. She is pursuing her PhD. from Pandit Deendayal Energy University (Gandhinagar, India). Her research area includes Image Processing, Computer Vision and Medical Imaging. She has mentored many BTech and MTech projects. She also serves as a member of the programme committees and session chair for international conferences and as a reviewer for indexed international journals

Author Articles
Mammogram Pre-processing Using filtering methods for Breast Cancer Diagnosis

By Shah Hemali Agrawal Smita Parita Oza Sudeep Tanwar Ahmed Alkhayyat

DOI:, Pub. Date: 8 Aug. 2023

Cancer is the second most found disease, and Breast cancer is the most common in women. Breast cancer is curable and can reduce mortality, but it needs to be identified early and treated accordingly. Radiologists use different modalities for the identification of Breast cancer. The superiority of Mammograms over other modalities is like minor radiation exposure and can identify different types of cancers. Therefore, mammograms are the most frequently used imaging modality for Breast Cancer Diagnosis. However, noise can be added while capturing the image, affecting the accuracy and analysis of the result. Therefore, using different filtering techniques to pre-process mammograms can enhance images and improve outcomes. For the study, the MIAS dataset has been used. This paper gives a comparative study on filters for Denoising and enhancement of mammograms. The study focuses on filters like Box Filter, Averaging filter, Gaussian Filter, Identical Filter, Convolutional 2D Filter, Median Filter, and Bilateral Filter. Performance measures used to compare these filters are Mean Squared Error (MSE), Structural Similarity Index Measure (SSIM), and Peak Signal-to-noise Ratio (PSNR). All Performance measures are evaluated for all images of MIAS dataset and compared accordingly. Results show that Gaussian Filter, Median Filter, and Bilateral Filter give better results than other filters.

[...] Read more.
Other Articles