An Ensemble Hierarchical Deep Learning Framework for Breast Cancer Prediction
Research Journal for Social Affairs · 29.03.2026 · 4 (3) · 367-373
DOI: 10.71317/rjsa.004.03.0965 ↗Kurz gefasst
Fachartikel in der Zeitschrift Research Journal for Social Affairs, online erschienen am 29.03.2026. Verfasst von Shahan Yamin Siddiqui und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Breast cancer is one of the major cancers in the world, and also the number one killer among all other causes of cancer in women. An early, quick, and precise diagnosis could contribute to better diagnosis and survival rate, but conventional techniques require an expert’s analysis, which varies from case to case. With the progress in AI and deep learning, possibilities exist that could help improve the computer-assisted diagnostic technique. In this study, an ensemble of hierarchical deep learning algorithm is proposed for breast cancer detection and histopathological classification through mammograms and histopathology images. Using the BreakHis dataset, the authors tested the accuracy of their model against 9,109 breast cancer images from biopsy specimens from 82 patients. There were two phases to the experimental evaluation of the proposed model: training and validating. The results showed BCP-HDL produced an accuracy of 99.63% and an average error rate of only 0.37%. Additionally, individual accuracy was determined for ductal (99.98%), lobular (98.72%), mucinous (99.49%), and papillary (99.28%) carcinomas. When the data was compared to other existing studies, it was demonstrated that this research represents the most accurate system. The results reported indicate that the combination of Convolutional Neural Networks (CNN), Long Short-Term Memory Networks (LSTM), and Support Vector Machines (SVM) as part of a Multi-Level System produces significant improvements in the performance of breast cancer diagnosis and classification, thereby providing an effective early detection tool for breast cancer and a clinical decision support system.
Zitieren
APA
Siddiqui, S. Y., Khan, M. F., Rehman, U., Tauheed, N., & Naveed, S. A. (2026). An Ensemble Hierarchical Deep Learning Framework for Breast Cancer Prediction. Research Journal for Social Affairs, 4(3), 367-373. https://doi.org/10.71317/rjsa.004.03.0965
Vancouver
Siddiqui SY, Khan MF, Rehman U, Tauheed N, Naveed SA. An Ensemble Hierarchical Deep Learning Framework for Breast Cancer Prediction. Research Journal for Social Affairs. 2026;4(3):367-373. doi:10.71317/rjsa.004.03.0965
BibTeX
@article{Siddiqui2026,
author = {Siddiqui, Shahan Yamin and Khan, Muhammad Farrukh and Rehman, Unaiza and Tauheed, Nusratullah and Naveed, Shahzada Atif},
title = {An Ensemble Hierarchical Deep Learning Framework for Breast Cancer Prediction},
journal = {Research Journal for Social Affairs},
year = {2026},
volume = {4},
number = {3},
pages = {367--373},
doi = {10.71317/rjsa.004.03.0965}
}
EUVIMED