An automatic classification of breast cancer using fuzzy scoring based ResNet CNN model

• Published in: Scientific reports Vol. 15; no. 1; pp. 20739 - 22
• Main Authors: Y., Nisha, Gopal, Jagadeesh
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.07.2025; Nature Portfolio
• Subjects: 631/67; 692/699; 692/700; Adaptive grey wolf optimization algorithm; Algorithms; Breast cancer; Breast Neoplasms - classification; Breast Neoplasms - diagnosis; Breast Neoplasms - diagnostic imaging; Deep Learning; Fast discrete wavelet transform; Feature extraction; Female; Fuzzy Logic; Humanities and Social Sciences; Humans; Hybrid deep learning; Image Processing, Computer-Assisted - methods; Machine Learning; multidisciplinary; Neural Networks, Computer; Science; Science (multidisciplinary); Statistical correlation analysis; Wavelet Analysis; Hybrid deep learning; Fast discrete wavelet transform; Statistical correlation analysis; Feature extraction; Adaptive grey wolf optimization algorithm; Fuzzy scoring ResNet; Breast cancer
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-025-07013-6

The expansion rate of medical data during the past ten years has rapidly expanded due to the vast fields. The automated disease diagnosis system is proposed using a deep learning (DL) algorithm, which automates and helps speed up the process efficiently. Further, this research concentrates on improving computation time based on the detection process. So, this research introduces a hybrid DL model for improving prediction performance andreducing time consumption compared to the machine learning (ML)model.Describing a pre-processing method utilizing statistical co-relational evaluation to improve the classifier’s accuracy.The features are then extracted from the Region of Interest (ROI) images using the wrapping technique and a fast discrete wavelet transform (FDWT). The extracted curvelet coefficients and the turn-time difficulty are too excessive to be categorized. Utilizing swarm intelligence, the Adaptive Grey Wolf Optimization Algorithm (AGWOA) was presented to reduce the time difficulty and choose the key characteristics. Here, it introduces a new building block identified as the Fuzzy Scoring Resnet-Convolutional Neural Network(FS-Resnet CNN) framework to optimize the network. The performance of the proposed model was assessedutilizing metrics such as recall, precision, F-measure, and accuracy.Furthermore, the suggested framework is computationally effective, less noise-sensitive, and efficiently saves memory. The simulation findings indicate that the suggested framework has a higher detection rate than the existing prediction model.