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

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Май 30, 2025

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 Gated Recurrent Unit (GRU) networks, optimized using the Hippopotamus Optimization Algorithm (HOA). Initially, DenseNet-41 extracts intricate spatial features from images. These are then processed by hybrid AlexNet-GRU model, leveraging AlexNet's robust feature extraction GRU's sequential capabilities. HOA is employed to fine-tune hyperparameters, ensuring optimal model performance. The proposed approach evaluated on benchmark datasets (BreakHis BACH), achieving classification accuracy of 99.60%, surpassing existing state-of-the-art models. results demonstrate efficacy with bio-inspired optimization techniques breast detection. research offers computationally efficient improving early clinical decision-making, potentially enhancing patient outcomes.

Язык: Английский

---

Hybrid Bayesian Deep Learning for Explainable Breast Cancer Diagnosis in Telemedicine: Integrating Multi-Modal Data

Опубликована: Янв. 1, 2025

Язык: Английский

---

Enhancing Breast Cancer Classification using a Modified GoogLeNet Architecture with Attention Mechanism

Al-Iraqia Journal of Scientific Engineering Research, Год журнала: 2024, Номер 3(1)

Опубликована: Март 1, 2024

Breast cancer incidence has been soaring sharply, and this is causing grave concern worldwide due to its high mortality rates. It should be correctly diagnosed in the early stages order achieve better patient outcomes. Over last decade, there a great demand for diagnosis systems based on AI that could used breast detection classification. These computerized devices utilize deep learning algorithms analyze medical scans thereby allowing subtle abnormality recognition distinguishing malignant from benign tumors. Computer-aided named CAD can assist radiologists pathologists more precise with their diagnoses while at same time increasing productivity. Furthermore, recent advances CNN architectures coupled attention mechanisms have further improved diagnosis. Attention-based models focus crucial regions hence enhancing classification accuracy reliability. In study, we introduce new approach improves of using GoogLeNet architecture modified by an mechanism image regions. The spatial transformer network (STN), which allows it significant areas histopathology images selective attention. Through mechanism, model becomes discriminatory features indicate different subtypes cancer. evaluate effectiveness method, implemented experiments BreaKHis dataset classifying carcinomas. This intentionally collected under various magnifications so as facilitate binary multiple tasks. outcomes clearly show outperforms original terms accuracy. For classification, proposed demonstrated rate 98.08%, whereas GoogLeNet's was 94.99%. multi-class 100x magnification, achieved 94.63% 85.06%. evident these findings efficiency significantly approach. study incorporating framework improve performance. Combining together lead accurate treatment decisions results. More efforts are needed develop area ongoing endeavors towards upgrading them ultimately contribute saving many lives fight against

Язык: Английский

---

Network traffic grant classification based on 1DCNN-TCN-GRU hybrid model

Applied Intelligence, Год журнала: 2024, Номер 54(6), С. 4834 - 4847

Опубликована: Март 1, 2024

Язык: Английский

---

Leveraging Deep Learning for Breast Cancer Detection : An Comprehensive study

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Май 2, 2024

Breast cancer is a significant health concern globally, emphasizing the need for accurate diagnostic tools early detection and treatment. In this research, we propose novel methodology leveraging deep learning techniques classifying breast histopathology images as benign or malignant. We employ state of art convolutional neural networks (CNNs), including VGG16, VGG19, ResNet50, ResNet101, custom CNN architecture, to extract discriminative features from images. One key contributions work incorporation class weights into training process, aiming address imbalance in da-taset enhance model performance. evaluate efficacy our approach using various performance metrics, accuracy, precision, recall, F1-score, on publicly available BreaKHis dataset.

Язык: Английский

---

Advancements in Histopathologic Cancer Detection: A Deep Learning Odyssey

Опубликована: Июнь 11, 2024

Язык: Английский

---

IoT based healthcare system using Fractional dung beetle optimization enabled deep learning for breast cancer classification

Computational Biology and Chemistry, Год журнала: 2024, Номер 114, С. 108277 - 108277

Опубликована: Ноя. 10, 2024

Язык: Английский

---

Accurate breast cancer diagnosis strategy (BCDS) based on deep learning techniques

Neural Computing and Applications, Год журнала: 2024, Номер unknown

Опубликована: Дек. 24, 2024

Язык: Английский

---

Survey on Distributed AI-Enhanced Deep Learning for Predicting Chemo Response in Non-Hormone Receptor Breast Cancer

Asian Journal of Applied Science and Technology, Год журнала: 2023, Номер 07(04), С. 27 - 34

Опубликована: Янв. 1, 2023

This study offers a novel method for forecasting the response to chemotherapy in non-hormone receptor breast cancer, difficult and complicated condition. TensorFlow-powered Spatial Temporal Integration (CNN-RNN) Architecture is used methods integrate clinical data histological images. Heuristic-driven deep learning techniques use domain-specific knowledge build models choose features. Using knowledge, Hybrid Differential Evolution Particle Swarm Optimization (DE-PSO) optimizes model's parameters. Because Lime comprehensible justifications predictions, its adoption guarantees transparency interpretability. Furthermore, federated distributed training approach preserve scalability safeguard patient privacy. precision empathy better treatment decisions cancer by fusing AI with expertise.

Язык: Английский

---

Effective degradation of bentazone by two-dimensional and three-phase, three-dimensional electro-oxidation system: kinetic studies and optimization using ANN

Environmental Science and Pollution Research, Год журнала: 2024, Номер 31(39), С. 51267 - 51299

Опубликована: Авг. 6, 2024

Язык: Английский

---