Deep hybrid model for Mpox disease diagnosis from skin lesion images

International Journal of Imaging Systems and Technology, Journal Year: 2024, Volume and Issue: 34(2)

Published: Feb. 26, 2024

Abstract This research presents DNLR‐NET, a novel model designed for automated and accurate diagnosis of MPox disease. The model's performance is constructed validated using carefully collected dataset from online repositories. DNLR‐NET begins by extracting deep features the DenseNet201 pre‐trained model, which exhibited superior compared to other models during comparison. obtained each dense layer are then used train six classifiers, among logistic regression showcases best with extracted deep, feature. A comparative study earlier advanced CNN classifying same demonstrates that achieves an impressive accuracy 97.55%, outperforming base only attains 95.91% accuracy. emphasizes efficacy combining regression. Grid Search algorithm employed optimal hyperparameter extraction, creating multiple unified feature sets achieving highest classification fusion yields results ensemble techniques such as random forest support vector machines also reduces training time complexity. surpasses existing models, ML demonstrating its effectiveness potential clinical implementation in diagnosing MPox. promising outcomes advantage learning algorithms, particularly transfer learning, highlight significance adopting methodologies CNN‐based settings. Researchers clinicians strongly encouraged explore implement these improve efficiency diagnosis.

Language: Английский

---

Breast cancer diagnosis: A systematic review

Journal of Applied Biomedicine, Journal Year: 2024, Volume and Issue: 44(1), P. 119 - 148

Published: Jan. 1, 2024

The second-leading cause of death for women is breast cancer. Consequently, a precise early diagnosis essential. With the rapid development artificial intelligence, computer-aided can efficiently assist radiologists in diagnosing problems. Mammography images, thermal and ultrasound images are three ways to diagnose paper will discuss some recent developments machine learning deep different cancer methods. components conventional methods image preprocessing, segmentation, feature extraction, classification. Deep includes convolutional neural networks, transfer learning, other Additionally, benefits drawbacks thoroughly contrasted. Finally, we also provide summary challenges potential futures diagnosis.

Language: Английский

---

Big data analytics deep learning techniques and applications: A survey

Information Systems, Journal Year: 2023, Volume and Issue: 120, P. 102318 - 102318

Published: Nov. 21, 2023

Language: Английский

---

Intelligent Ultrasound Imaging for Enhanced Breast Cancer Diagnosis: Ensemble Transfer Learning Strategies

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 22243 - 22263

Published: Jan. 1, 2024

According to WHO statistics for 2018, there are 1.2 million cases and 700,000 deaths from breast cancer (BC) each year, making it the second-highest cause of mortality women globally. In recent years, advances in artificial (AI) intelligence machine (ML) learning have shown incredible potential increasing accuracy efficiency BC diagnosis. This research describes an intelligent image analysis system that leverages capabilities transfer (TLs) with ensemble stacking ML models. As part this research, we created a model analyzing ultrasound images using cutting-edge TL models such as Inception V3, VGG-19, VGG-16. We implemented models, including MLP (Multi-Layer Perceptron) different architectures (10 10, 20 20, 30 30) Support Vector Machines (SVM) RBF Polynomial kernels. analyzed effectiveness proposed performance parameters (accuracy (CA), sensitivity, specificity, AUC). Compared results existing diagnostic systems, method (Inception V3 + Staking) is superior, 0.947 AUC 0.858 CA values. The BCUI consists data collection, pre-processing, learning, evaluation, comparative demonstrating its superiority over methods.

Language: Английский

---

Attention-map augmentation for hypercomplex breast cancer classification

Pattern Recognition Letters, Journal Year: 2024, Volume and Issue: 182, P. 140 - 146

Published: April 18, 2024

Breast cancer is the most widespread neoplasm among women and early detection of this disease critical. Deep learning techniques have become great interest to improve diagnostic performance. However, distinguishing between malignant benign masses in whole mammograms poses a challenge, as they appear nearly identical an untrained eye, region (ROI) constitutes only small fraction entire image. In paper, we propose framework, parameterized hypercomplex attention maps (PHAM), overcome these problems. Specifically, deploy augmentation step based on computing maps. Then, are used condition classification by constructing multi-dimensional input comprised original breast image corresponding map. step, neural network (PHNN) employed perform classification. The framework offers two main advantages. First, provide critical information regarding ROI allow model concentrate it. Second, architecture has ability local relations dimensions thanks algebra rules, thus properly exploiting provided We demonstrate efficacy proposed both mammography images well histopathological ones. surpass attention-based state-of-the-art networks real-valued counterpart our approach. code work available at https://github.com/elelo22/AttentionBCS.

Language: Английский

---

Privacy-Preserving Breast Cancer Classification: A Federated Transfer Learning Approach

Deleted Journal, Journal Year: 2024, Volume and Issue: 37(4), P. 1488 - 1504

Published: Feb. 29, 2024

Breast cancer is deadly causing a considerable number of fatalities among women in worldwide. To enhance patient outcomes as well survival rates, early and accurate detection crucial. Machine learning techniques, particularly deep learning, have demonstrated impressive success various image recognition tasks, including breast classification. However, the reliance on large labeled datasets poses challenges medical domain due to privacy issues data silos. This study proposes novel transfer approach integrated into federated framework solve limitations limited collaborative healthcare settings. For classification, mammography MRO images were gathered from three different centers. Federated an emerging privacy-preserving paradigm, empowers multiple institutions jointly train global model while maintaining decentralization. Our proposed methodology capitalizes power pre-trained ResNet, neural network architecture, feature extractor. By fine-tuning higher layers ResNet using diverse centers, we enable learn specialized features relevant domains leveraging comprehensive representations acquired large-scale like ImageNet. overcome shift caused by variations distributions across introduce adversarial training. The learns minimize discrepancy maximizing classification accuracy, facilitating acquisition domain-invariant features. We conducted extensive experiments obtained Comparative analysis was performed evaluate against traditional standalone training without adaptation. When compared with models, our showed accuracy 98.8% computational time 12.22 s. results showcase promising enhancements generalization, underscoring potential method improving performance upholding environment.

Language: Английский

---

Integrated Model (IM- LTS) for Lung Tumor Segmentation using Neural Networks and IoMT].

MethodsX, Journal Year: 2025, Volume and Issue: 14, P. 103201 - 103201

Published: Feb. 7, 2025

Language: Английский

---

AI-Assisted Breast Cancer Prediction, Classification, and Future Directions: A Narrative Review Involving Histopathological Image Datasets

The Open Public Health Journal, Journal Year: 2025, Volume and Issue: 18(1)

Published: Feb. 7, 2025

Breast cancer-related deaths in women have increased significantly the past decade, emphasizing need for an accurate and early diagnosis. AI-assisted diagnosis using deep learning machine (DML) approaches has become a key method analysing breast tissue identifying tumour stages. DML algorithms are particularly effective classifying cancer images due to their ability handle large datasets, work with unstructured data, generate automated features, improve over time. However, performance of these models is heavily on datasets used training, performing inconsistently between different datasets. Given prediction that by 2050, there will be more than 30 million new cases 10 worldwide, it crucial focus recent advancements histopathological image systems. Histopathological provide critical information identify abnormalities, which directly impact model performance. This review discusses analyses various DML-based implementation, highlighting research gaps offering suggestions future improvements. The goal develop efficient early-stage cancer. In addition, this detection assists healthcare professional guiding prevention methods smart

Language: Английский

---

An interpretable framework for gastric cancer classification using multi-channel attention mechanisms and transfer learning approach on histopathology images

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: April 16, 2025

Abstract The importance of gastric cancer (GC) and the role deep learning techniques in categorizing GC histopathology images have recently increased. Identifying drawbacks traditional models, including lack interpretability, inability to capture complex patterns, adaptability, sensitivity noise. A multi-channel attention mechanism-based framework is proposed that can overcome limitations conventional models by dynamically focusing on relevant features, enhancing extraction, capturing relationships medical data. uses three different mechanism channels convolutional neural networks extract multichannel features during classification process. framework’s strong performance confirmed competitive experiments conducted a publicly available Gastric Histopathology Sub-size Image Database, which yielded remarkable accuracies 99.07% 98.48% validation testing sets, respectively. Additionally, HCRF dataset, achieved high accuracy 99.84% 99.65% effectiveness interchangeability are further ablation experiments, highlighting histopathological image tasks. This offers an advanced pragmatic artificial intelligence solution addresses challenges posed unique characteristics for intricate analysis. approach engineering demonstrates significant potential diagnostic precision achieving treatment outcomes.

Language: Английский

---

The Mountain Gazelle Optimizer for truss structures optimization

Applied Computing and Intelligence, Journal Year: 2023, Volume and Issue: 3(2), P. 116 - 144

Published: Jan. 1, 2023

<abstract> <p>Computational tools have been used in structural engineering design for numerous objectives, typically focusing on optimizing a process. We first provide detailed literature review truss structures with metaheuristic algorithms. Then, we evaluate an effective solution designing through method called the mountain gazelle optimizer, which is nature-inspired meta-heuristic algorithm derived from social behavior of wild gazelles. use benchmark problems optimization and penalty handling constraints. The performance proposed will be evaluated by solving complex challenging problems, are common design. include high number locally optimal solutions non-convex search space function, as these considered suitable to capabilities This work its kind, it examines optimizer applied field while assessing ability handle such effectively. results compared other algorithms, showing that can efficient lowest possible weight.</p> </abstract>

Language: Английский

---