Histopathological image based breast cancer diagnosis using deep learning and bio inspired optimization

Sep 24, 2026

Uploaded by Đào

Breast cancer diagnosis remains a major challenge requiring accurate, automated detection from histopathological images. This study presents a deep learning framework that combines DenseNet-41 feature extraction with a hybrid AlexNet-GRU classifier, where AlexNet supports strong feature extraction and GRU models sequential dependencies. Hyperparameters are fine-tuned using the Hippopotamus Optimization Algorithm (HOA) to maximize performance. Evaluation on BreakHis and BACH benchmarks achieves 99.60% classification accuracy, improving early diagnosis and clinical decision support.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages24

Reading Time1 h 0 min

AccessFree to download

Introduction

Breast cancer burden and early diagnosis needs

Histopathology workflow and limitations of manual analysis

Related Methods

Computer-aided diagnosis and machine learning approaches

Role of deep learning (CNNs, RNNs) in medical imaging

Proposed Framework

DenseNet-41 feature extraction

Hybrid AlexNet-GRU classification

HOA-based hyperparameter optimization

Experimental Evaluation

Benchmark datasets and performance results

[www. nature.com/scientificreports](www. nature.com/scientificreports) OPEN Histopathological image based breast cancer diagnosis using deep learning and bio inspired optimization Venkata Nagaraju Thatha1, M. Ganesh Karthik2, Venu Gopal Gaddam3, D. Pramodh Krishna4, S. Venkataramana5, Kranthi Kumar Lella6􀀍 & Udayaraju Pamula7 Breast cancer diagnosis remains a crucial challenge in medical research, necessitating accurate and automated detection methods. This study introduces an advanced deep learning framework for histopathological image classification, integrating AlexNet and Gated Recurrent Unit (GRU) networks, optimized using the Hippopotamus Optimization Algorithm (HOA) . Initially, DenseNet-41 extracts intricate spatial features from histopathological images. These features are then processed by the hybrid AlexNet-GRU model, leveraging AlexNet’s robust feature extraction and GRU’s sequential learning capabilities. HOA is employed to fine-tune hyperparameters, ensuring optimal model performance. The proposed approach is evaluated on benchmark datasets (BreakHis and BACH), achieving a classification accuracy of 99.60%, surpassing existing state-of-the-art models. The results demonstrate the efficacy of integrating deep learning with bio-inspired optimization techniques in breast cancer detection. This research offers a robust and computationally efficient framework for improving early diagnosis and clinical decision-making, potentially enhancing patient outcomes. Keywords Breast Cancer, Histopathological image classification, AlexNet, Gated recurrent unit (GRU), DenseNet-41, Hippopotamus optimization algorithm (HOA), Deep learning Almost any region of the body in which cells begin to multiply excessively can become cancerous1. Cancer is believed to be the primary cause of death globally, with yearly death rates continuing to rise2–4. Particularly in developing nations like Melanesia, Micronesia/Polynesia, Australia, Western Africa, and the Caribbean, breast cancer is one of the main reasons for female mortality5. Nonetheless, the highest proportions of breast cancer cases are found in Northern Europe, Western Europe, Australia, and North America6. While men are also at risk, breast cancer predominantly affects women7. Adipose tissue, lobules, and milk ducts make up the majority of a woman’s breast structure8. Breast cancer can originate in the milk-producing lobules or the ducts that carry milk to the nipple8,9. Ductal and lobular subtypes account for 40–75% of all reported cases globally10. Early diagnosis and treatment are crucial to halting breast cancer progression. Various imaging techniques like magnetic resonance imaging (MRI), ultrasound, mammograms, and biopsies aid in detection11–14. However, definitive confirmation requires tissue biopsy15. Histopathology images derived from these biopsies are crucial for improving breast cancer classification accuracy16. The process involves embedding tissue samples in paraffin wax, slicing them into slides, and examining them under a microscope17–19. Manual analysis of these biopsies is time-consuming, labor-intensive, and heavily reliant on histopathologist expertise and image quality20,21. To address these challenges, Computer-Aided Diagnosis (CAD) systems analyze histopathology images using machine learning methods, improving classification accuracy22–24. Deep learning, particularly Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), offers advanced capabilities for handling complex, high-dimensional data. These techniques aid in early 1Department of Information Technology, MLR Institute of Technology, Hyderabad, India. 2Department of Computer Science and Engineering, GITAM School of Technology, GITAM University-Bengaluru Campus, Bengaluru, India. 3Department of CSE (AIML), BV Raju Institute of Technology, Narsapur, India. 4Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vijayawada, India. 5Dep

FAQs

What problem does this research address in breast cancer detection?

It targets the need for accurate, automated diagnosis from histopathological images, which is difficult with time-consuming manual analysis and dependence on image quality and expert expertise.

How does the proposed model classify histopathological images?

DenseNet-41 extracts detailed spatial features, then a hybrid AlexNet-GRU model processes them, combining AlexNet’s feature extraction with GRU’s ability to learn sequential dependencies.

What is the role of the Hippopotamus Optimization Algorithm (HOA)?

HOA is used to fine-tune the model’s hyperparameters to ensure optimal classification performance.

Sep 24, 2026

Uploaded by Đào