Scholarly open access journals, Peer-reviewed, and Refereed Journals, Impact factor 8.14 (Calculate by google scholar and Semantic Scholar | AI-Powered Research Tool) , Multidisciplinary, Monthly, Indexing in all major database & Metadata, Citation Generator, Digital Object Identifier(DOI)
Among all cancers detected breast cancer rates are alarming. Ministry of Health and Family Welfare under Government of India has ranked breast cancer as commonly occurring cancer among women, both in terms of incidence as well as mortality. Mortality rate can be reduced if it is diagnosed early, using techniques like Mammography, X-Ray, MRI (Magnetic Resonance Image), Computed Tomography (CT) etc. In this paper, we propose a method for detection of breast cancer and segmentation of abnormalities for MRI scans. We first classify the MRI scans as malignant or benign using deep learning algorithm and then pass the malignant images for segmentation and sizing. Segmentation of abnormalities (e.g. tumor) is done using image processing and the size (length and breadth) of tumor is calculated. For classification we used Resnet whose performance is measured on both CPU and GPU systems. Resnet has advantage of tolerance to vanishing gradient compared to other deep neural networks. Instead of widening the network, Resnet increases the depth of network which results in less trainable parameters. In only about 5 epochs Resnet gets trained, with good accuracy. During training phase, on CPU system, Resnet demonstrates 98.07% accuracy in 48 min whereas on GPU system it shows 96.96% accuracy in 24 min. During inference phase, Resnet exhibits 99.56% and 99.07% accuracy on CPU and GPU systems respectively.
Keywords:
Breast Cancer, Deep Learning, Magnetic Resonance Imaging (MRI), Segmentation, Image processing
Cite Article:
"Detection Of Breast Cancer And Segmentation Of Abnormalities Using Deep Learning And Image Processing Techniques", International Journal for Research Trends and Innovation (www.ijrti.org), ISSN:2455-2631, Vol.7, Issue 8, page no.1599 - 1607, August-2022, Available :http://www.ijrti.org/papers/IJRTI2208256.pdf
Downloads:
000205292
ISSN:
2456-3315 | IMPACT FACTOR: 8.14 Calculated By Google Scholar| ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 8.14 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator