Magnetic Resonance Imaging and Ultrasound-based Breast Tumor Classification using Soft Voting Ensemble Deep Learning Method

Authors

DOI:

https://doi.org/10.31436/iiumej.v27i2.3387

Keywords:

Breast Tumor Classification, Dynamic Contrast-Enhanced Magnetic Resonance Imaging, Ensemble Deep Learning, Multi-Modality Adaptive Feature Fusion Framework, Ultrasound

Abstract

In recent years, breast cancer (BC) has been considered the most common cause of death in women worldwide. Commonly, BC occurs in breast cells or fatty tissues within the breast, and these tissues tend to grow faster and lead to death. The survival of patients with BC can be enhanced with early and accurate diagnosis. Therefore, this research proposes an ensemble learning approach that uses Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and a soft-voting mechanism to make early predictions and diagnoses of BC. The input sample images are initially passed through the convolutional layer to extract features from each modality separately. The Multi-Modality Adaptive Feature Fusion (MMAFF) technique combines generic features across modalities. Compared with existing techniques such as Deep Convolutional Neural Networks (DCNN) and Resolution Adaptive Network (RANet), the proposed ensemble learning technique achieves 89.6% accuracy for clinical indicator prediction and 99.25% accuracy for final tumor classification on the MRI-US dataset.

ABSTRAK: Dalam beberapa tahun kebelakangan ini, kanser payudara (Breast Cancer, BC) telah dikenal pasti sebagai antara punca utama kematian dalam kalangan wanita di seluruh dunia. Secara umum, BC berlaku pada sel-sel payudara atau tisu lemak dalam payudara, di mana sel-sel ini cenderung membiak dengan cepat dan berpotensi membawa kepada kematian. Kadar kelangsungan hidup pesakit BC dapat ditingkatkan melalui diagnosis awal yang tepat dan boleh dipercayai. Sehubungan itu, kajian ini mencadangkan satu pendekatan pembelajaran ansambel yang menggabungkan Rangkaian Neural Berulang (Recurrent Neural Networks, RNN), Rangkaian Neural Konvolusi (Convolutional Neural Networks, CNN), Memori Jangka Pendek Panjang (Long Short-Term Memory, LSTM), serta mekanisme pengundian lembut (soft voting) bagi tujuan ramalan awal dan diagnosis BC. Imej sampel input pada peringkat awal diproses melalui lapisan konvolusi untuk mengekstrak ciri daripada setiap modaliti secara berasingan. Seterusnya, teknik Multi-Modality Adaptive Feature Fusion (MMAFF) digunakan untuk menggabungkan ciri-ciri umum daripada setiap modaliti tersebut. Berbanding dengan teknik sedia ada seperti Rangkaian Neural Konvolusi Mendalam (Deep Convolutional Neural Networks, DCNN) dan Resolution Adaptive Network (RANet), pendekatan pembelajaran ansambel yang dicadangkan menunjukkan prestasi yang lebih baik dengan mencapai ketepatan sebanyak 89.6% bagi ramalan penunjuk klinikal dan 99.25% bagi pengelasan akhir tumor menggunakan set data MRI-US.

Downloads

Download data is not yet available.

Author Biographies

Vijaya Bathini, Sri Venkateswara College of Engineering Tirupati

Department of Computer Science and Engineering

Sankar Babu Jangili, Sri Venkateswara College of Engineering Tirupati

Computer Science and Engineering

Santosh Kumar Reddy Thikkam, Sri Venkateswara College of Engineering Tirupati

 Computer Science and Engineering

References

Zou Y, Zhang J, Huang S, Liu B. (2022) Breast cancer histopathological image classification using attention high?order deep network. International Journal of Imaging Systems and Technology, 32(1):266-279.

Liu M, He Y, Wu M, Zeng C. (2022) Breast histopathological image classification method based on autoencoder and siamese framework. Information, 13(3):107.

Ahmad N, Asghar S, Gillani SA. (2022). Transfer learning-assisted multi-resolution breast cancer histopathological images classification. The Visual Computer, 38(8):2751-2770.

Alqahtani Y, Mandawkar U, Sharma A, Hasan MNS, Kulkarni MH, Sugumar R. (2022) Breast cancer pathological image classification based on the multiscale CNN squeeze model. Computational Intelligence and Neuroscience, 2022(1):7075408.

Zhou Y, Zhang C, Gao S. (2022) Breast cancer classification from histopathological images using resolution adaptive network. IEEE Access, 10:35977-35991.

Michael E, Ma H, Li H, Qi S. (2022) An optimized framework for breast cancer classification using machine learning. BioMed Research International, 2022(1):8482022.

Wakili MA, Shehu HA, Sharif MH, Sharif MHU, Umar A, Kusetogullari H, Ince IF, Uyaver S. (2022) Classification of breast cancer histopathological images using DenseNet and transfer learning. Computational Intelligence and Neuroscience, 2022(1):8904768

Balkenende L, Teuwen J, Mann RM. (2022) Application of deep learning in breast cancer imaging. In Seminars in Nuclear Medicine. WB Saunders, 52(5):584-596

Liu M, Hu L, Tang Y, Wang C, He Y, Zeng C, Lin K, He Z, Huo W. (2022) A deep learning method for breast cancer classification in the pathology images. IEEE Journal of Biomedical and Health Informatics, 26(10):5025-5032.

Rakha EA, Tse GM, Quinn CM. (2023) An update on the pathological classification of breast cancer. Histopathology, 82(1):5-16.

Yu D, Lin J, Cao T, Chen Y, Li M, Zhang X. (2023) SECS: An effective CNN joint construction strategy for breast cancer histopathological image classification. Journal of King Saud University-Computer and Information Sciences, 35(2):810-820.

Jabeen K, Khan MA, Balili J, Alhaisoni M, Almujally NA, Alrashidi H, Tariq U, Cha, JH. (2023) BC2NetRF: breast cancer classification from mammogram images using enhanced deep learning features and equilibrium-jaya controlled regula falsi-based features selection. Diagnostics, 13(7):1238.

Shankar K, Dutta AK, Kumar S, Joshi GP, Doo IC. (2022) Chaotic sparrow search algorithm with deep transfer learning enabled breast cancer classification on histopathological images. Cancers, 14(11):2770.

Hamedani-KarAzmoudehFar F, Tavakkoli-Moghaddam R, Tajally AR, Aria SS. (2023) Breast cancer classification by a new approach to assessing deep neural network-based uncertainty quantification methods. Biomedical Signal Processing and Control, 79:104057.

Rashmi R, Prasad K, Udupa CBK. (2023) Region-based feature enhancement using channel-wise attention for classification of breast histopathological images. Neural Computing and Applications, 35(8):5839-5854.

Ahmad N, Asghar S, Gillani SA. (2022) Transfer learning-assisted multi-resolution breast cancer histopathological images classification. The Visual Computer, 38(8):2751-2770.

Ukwuoma CC, Hossain MA, Jackson JK, Nneji GU, Monday HN, Qin Z. (2022) Multi-classification of breast cancer lesions in histopathological images using DEEP_Pachi: Multiple self-attention head. Diagnostics, 12(5):1152.

Clement D, Agu E, Suleiman MA, Obayemi J, Adeshina S, Soboyejo W. (2022) Multi-class breast cancer histopathological image classification using multi-scale pooled image feature representation (MPIFR) and one-versus-one support vector machines. Applied Sciences, 13(1):156.

Umer MJ, Sharif M, Kadry S, Alharbi A. (2022) Multi-class classification of breast cancer using 6B-Net with deep feature fusion and selection method. Journal of Personalized Medicine, 12(5):683.

Zhou Y, Zhang C, Gao S. (2022) Breast cancer classification from histopathological images using resolution adaptive network. IEEE Access, 10:35977-35991.

Qiao M, Liu C, Li Z, Zhou J, Xiao Q, Zhou S, Chang C, Gu Y, Guo Y, Wang Y. (2022) Breast tumor classification based on MRI-US images by disentangling modality features. IEEE Journal of Biomedical and Health Informatics, 26(7):3059-3067.

Akyol K, Uçar E, Atila Ü, Uçar M. (2024) An ensemble approach for classification of tympanic membrane conditions using soft voting classifier. Multimedia Tools and Applications, 1-22.

Saleh H, Abd-El Ghany SF, Alyami H, Alosaimi W. (2022) Predicting breast cancer based on optimized deep learning approach. Computational Intelligence and Neuroscience, 2022(1):1820777.

Downloads

Published

2026-05-10

How to Cite

Ramayanam, S., Bathini, V., Jangili, S. B., & Thikkam, S. K. R. (2026). Magnetic Resonance Imaging and Ultrasound-based Breast Tumor Classification using Soft Voting Ensemble Deep Learning Method. IIUM Engineering Journal, 27(2), 74–86. https://doi.org/10.31436/iiumej.v27i2.3387

Issue

Section

Electrical, Computer and Communications Engineering