Title: A comparative study and implementation of neuro-fuzzy and decision tree for malignant tumour detection system

Authors: Sanjeev Kumar; Rajesh Kumar Maurya; Sanjay Kumar Yadav; Baij Nath Kaushik

Addresses: CSE Department, ABESIT, Ghaziabad, India ' Department of Computer Science and Information Technology, Sam Higginbottom University of Agriculture, Technology and Sciences, Allahabad UP, India ' Department of Computer Science and Information Technology, Sam Higginbottom University of Agriculture, Technology and Sciences, Allahabad UP, India ' School of Computer Science and Engineering, Shri Mata Vaishno Devi University, Katra, J&K, India

Abstract: Breast cancer is one of the most chronic diseases found in women. There are two types of tumours found in the breasts: malignant and benign. A patient who has more percentage of malignant tumours is suffering from breast cancer. A model based on neuro-fuzzy is proposed to classify the tumour as malignant or benign. The designed system works on various attributes of tumour like tumour thickness, shape, size, etc. The classification process completes in three phases; phase 1 classifies the attributes as cat1 or cat2 on the basis of information gain. Then in phase 2, cat1 attributes are used to select the class of tumour by using the radial bias function neural network while the cat2 attributes use the fuzzy to select the class of tumour. The results of both techniques are collaborated by using the fuzzy inference system in the phase 3. The effectiveness of the technique is easily identified by the results. The results are compared for the accuracy of cancer detection of cat1 and cat2 with neuro-fuzzy system and decision tree.

Keywords: breast cancer; malignant; benign; tumour; radial basis function; RBF; fuzzy; decision tree.

DOI: 10.1504/IJAIP.2022.126699

International Journal of Advanced Intelligence Paradigms, 2022 Vol.23 No.3/4, pp.410 - 422

Received: 13 Mar 2018
Accepted: 13 Nov 2018
Published online: 03 Nov 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article