Quantum Fruit Fly algorithm and ResNet50-VGG16 for medical diagnosis

Article Chemistry, Multidisciplinary

Early Risk Prediction of Diabetes Based on GA-Stacking

Yaqi Tan et al.

Summary: This paper proposes a GA-stacking ensemble learning model for early risk prediction of diabetes, and verifies its accuracy and efficiency improvement through experiments.

APPLIED SCIENCES-BASEL (2022)

Add to Collection

Article Computer Science, Information Systems

A Breast Cancer Diagnosis Method Based on VIM Feature Selection and Hierarchical Clustering Random Forest Algorithm

Zexian Huang et al.

Summary: Breast cancer is a neoplastic disease that seriously threatens women's health. Accurate detection and effective treatment are crucial for reducing the death rate. In this study, a Hierarchical Clustering Random Forest (HCRF) model is proposed to improve the classification performance of decision trees in breast cancer diagnosis. By measuring the similarity among decision trees and using hierarchical clustering, a random forest with low similarity and high accuracy is constructed. The proposed method achieved better accuracy compared to other algorithms on the tested datasets.

IEEE ACCESS (2022)

Add to Collection

Article Computer Science, Information Systems

Prediction model using SMOTE, genetic algorithm and decision tree (PMSGD) for classification of diabetes mellitus

Chandrashekhar Azad et al.

Summary: The study introduced a novel prediction model, PMSGD, and through different layers of processing, it improved the accuracy and effectiveness of diabetes classification.

MULTIMEDIA SYSTEMS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Deep convolutional neural network for diabetes mellitus prediction

Suja A. Alex et al.

Summary: Diabetes mellitus, a widely known disease, leads to abnormal metabolism in various organs due to decreased hormone production and increased glucose. However, the classification performance using machine learning techniques is often hindered by missing values and class imbalance. This study proposes an efficient prediction algorithm using Deep 1D-Convolutional Neural Network values, which has shown successful results in balancing and mitigating these challenges for diabetes mellitus classification.

NEURAL COMPUTING & APPLICATIONS (2022)

Add to Collection

Article Computer Science, Information Systems

Speaker Gender Recognition Based on Deep Neural Networks and ResNet50

Abeer Ali Alnuaim et al.

Summary: This study constructs a gender identification model using voice data and deep neural networks, achieving high accuracy and demonstrating good generalization capacity.

WIRELESS COMMUNICATIONS & MOBILE COMPUTING (2022)

Add to Collection

Article Biotechnology & Applied Microbiology

Enhanced Watershed Segmentation Algorithm-Based Modified ResNet50 Model for Brain Tumor Detection

Arpit Kumar Sharma et al.

Summary: This paper presents a novel technique that combines enhanced watershed modeling with modified ResNet50 architecture for brain tumor detection and classification. The technique achieves effective diagnosis by extracting deep features and demonstrates good performance.

BIOMED RESEARCH INTERNATIONAL (2022)

Add to Collection

Article Medical Informatics

Expert system based on fuzzy rules for diagnosing breast cancer

Isha Thani et al.

Summary: Breast cancer is a major threat globally, and early diagnosis and treatment are crucial; existing expert systems are not fully utilized, especially in terms of reaching rural areas.

HEALTH AND TECHNOLOGY (2022)

Add to Collection

Article Health Care Sciences & Services

A comparative study on deep learning models for text classification of unstructured medical notes with various levels of class imbalance

Hongxia Lu et al.

Summary: This study compares the performance of various deep learning algorithms in text classification tasks on medical notes. The results show that the Transformer encoder model performs the best in most scenarios, while the CNN model achieves comparable performance when the disease prevalence is close to or greater than 50% and has a shorter training time. Specific word embedding methods can also improve the model performance.

BMC MEDICAL RESEARCH METHODOLOGY (2022)

Add to Collection

Article Mathematics, Interdisciplinary Applications

A Novel Feature Selection Method for Classification of Medical Data Using Filters, Wrappers, and Embedded Approaches

Saba Bashir et al.

Summary: The study aims to enhance classification accuracy using effective feature selection algorithms, utilizing different feature selection methods and SVM classifier to classify data, achieving high accuracy on microarray and heart disease datasets.

COMPLEXITY (2022)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging

Classification Algorithms for Brain Magnetic Resonance Imaging Images of Patients with End-Stage Renal Disease and Depression

Yan Cheng et al.

Summary: This study used a brain MRI image classification algorithm to explore the relationship between depression and brain function. The results showed that the capsule network can improve the heterogeneity of depression, and the combination of fALFF, DC, and ALFF can achieve higher accuracy.

CONTRAST MEDIA & MOLECULAR IMAGING (2022)

Add to Collection

Article Chemistry, Analytical

An Ensemble Approach to Predict Early-Stage Diabetes Risk Using Machine Learning: An Empirical Study

Umm e Laila et al.

Summary: Diabetes is a long-lasting disease that can cause various health issues. Timely disease prediction can save lives and support healthcare decision making. In this study, ensemble learning technique was used to predict diabetes, with Random Forest showing the best performance.

SENSORS (2022)

Add to Collection

Article Multidisciplinary Sciences

Hybrid Model Based on ReliefF Algorithm and K-Nearest Neighbor for Erythemato-Squamous Diseases Forecasting

Abdullah S. Alotaibi

Summary: Machine learning techniques have been used in medical sciences for decades, but conventional methods have limitations in dealing with large datasets. Researchers have suggested hybrid solutions to integrate the merits of different methods for better results.

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING (2022)

Add to Collection

Article Engineering, Electrical & Electronic

Residual spatiotemporal autoencoder for unsupervised video anomaly detection

K. Deepak et al.

Summary: Modeling abnormal spatiotemporal events using a normality modeling approach and a residual autoencoder is effective in detecting anomalies in surveillance videos and shows better results compared to deeper layer approaches. The reconstruction loss method is utilized for identifying irregularities, and good practices for training autoencoders for video anomaly detection are presented using benchmark datasets.

SIGNAL IMAGE AND VIDEO PROCESSING (2021)

Add to Collection

Article Computer Science, Hardware & Architecture

A multiple multilayer perceptron neural network with an adaptive learning algorithm for thyroid disease diagnosis in the internet of medical things

Mehdi Hosseinzadeh et al.

Summary: This study focuses on improving the diagnosis accuracy of thyroid diseases in IoMT systems using artificial neural networks. The proposed Multiple Multilayer Perceptron (MMLP) neural network, combined with an adaptive learning rate algorithm, significantly increases the overall accuracy of disease classification. By comparing with standard back-propagation, the MMLP with adaptive learning rate algorithm achieves a 4.6% improvement in accuracy.

JOURNAL OF SUPERCOMPUTING (2021)

Add to Collection

Article Chemistry, Analytical

Flame Image Processing and Classification Using a Pre-Trained VGG16 Model in Combustion Diagnosis

Zbigniew Omiotek et al.

Summary: The method combining flame image processing with a deep convolutional neural network (DCNN) ensures high accuracy in identifying undesired combustion states, with a detection accuracy ranging from 82% to 98%. Its high accuracy and fast processing time make it suitable for application in real systems.

SENSORS (2021)

Add to Collection

Article Computer Science, Information Systems

BARF: A new direct and cross-based binary residual feature fusion with uncertainty-aware module for medical image classification

Moloud Abdar et al.

Summary: Automatic medical image analysis, particularly using deep learning-based computer-aided diagnosis systems, plays a crucial role in early disease diagnosis and treatment. Uncertainty quantification has not been well addressed in existing deep learning methods for medical research. A novel fusion model called Binary Residual Feature fusion (BARF) with uncertainty-aware module has been proposed, showing efficient performance in medical image classification with Monte Carlo dropout for uncertainty handling in real clinical settings.

INFORMATION SCIENCES (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Multi class SVM algorithm with active learning for network traffic classification

Shi Dong

Summary: This paper explores the use of network flow-level characteristics for identifying application types of traffic, introducing an improved support vector machine algorithm called CMSVM to address the data imbalance issue. Experimental results show that the CMSVM algorithm can reduce computation cost, improve classification accuracy, and solve the imbalance problem compared to other machine learning techniques.

EXPERT SYSTEMS WITH APPLICATIONS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

MetaMed: Few-shot medical image classification using gradient-based meta-learning

Rishav Singh et al.

Summary: The study addresses the challenges of long-tailed distributions and lack of high-quality annotated images in medical datasets. By formulating a few-shot learning problem and proposing a meta-learning-based MetaMed approach, the model achieved promising results with an accuracy of over 70% on three medical datasets, showcasing improved generalization capability.

PATTERN RECOGNITION (2021)

Add to Collection

Article Biochemical Research Methods

A hybrid model based on general regression neural network and fruit fly optimization algorithm for forecasting and optimizing paclitaxel biosynthesis in Corylus avellana cell culture

Mina Salehi et al.

Summary: The study utilized GRNN-FOA models for the first time to predict and optimize growth and paclitaxel biosynthesis in Corylus avellana cell culture. The results showed superior prediction performances and great accordance between the predicted and observed values, indicating that GRNN-FOA may serve as a new mathematical tool for forecasting and optimizing secondary metabolite production in plant in vitro culture.

PLANT METHODS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Deep virtual adversarial self-training with consistency regularization for semi-supervised medical image classification

Xi Wang et al.

Summary: A novel semi-supervised deep learning method incorporating self-training and consistency regularization is proposed in this study to improve the discrimination capability of training models by effectively utilizing useful information from unlabeled data.

MEDICAL IMAGE ANALYSIS (2021)

Add to Collection

Article Biology

BreastNet18: A High Accuracy Fine-Tuned VGG16 Model Evaluated Using Ablation Study for Diagnosing Breast Cancer from Enhanced Mammography Images

Sidratul Montaha et al.

Summary: Early stage breast cancer diagnosis through mammography is crucial for treatment planning and survival rates. This study focuses on developing an effective method to classify breast images accurately, utilizing image processing, data augmentation, and fine-tuned CNN models. The proposed BreastNet18 model, based on VGG16, achieved high accuracy of 98.02%, outperforming other CNN architectures.

BIOLOGY-BASEL (2021)

Add to Collection

Article Computer Science, Information Systems

Automatic prediction of COVID-19 from chest images using modified ResNet50

Marwa Elpeltagy et al.

Summary: The recent pandemic of COVID-19 has highlighted the urgent need for accurate diagnostic tools. In addition to laboratory methods, medical imaging techniques and artificial intelligence are being utilized for detection and diagnosis. Transfer learning from pre-trained CNNs shows promise for accurate diagnosis of coronavirus.

MULTIMEDIA TOOLS AND APPLICATIONS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

OAA-SVM-MS: A fast and efficient multi-class classification algorithm*

Yuze Duan et al.

Summary: This paper introduces the idea of learning uniformly ergodic Markov chain for OAA-SVM algorithm, and proposes a new OAA-SVM-MS method. Experimental results confirm that the OAA-SVM-MS algorithm significantly outperforms the traditional OAA-SVM algorithm and other multi-class SVM algorithms in performance.

NEUROCOMPUTING (2021)

Add to Collection

Article Chemistry, Analytical

A Tri-Stage Wrapper-Filter Feature Selection Framework for Disease Classification

Moumita Mandal et al.

Summary: In this study, a feature selection framework for disease detection based on medical reports was proposed, which utilized a tri-stage process involving ensemble of filter methods, classification algorithms, Pearson correlation, and a meta-heuristic algorithm. The results showed that the proposed method outperformed state-of-the-art methods in detecting important features, leading to less medical tests for accurate diagnosis and saving time and cost.

SENSORS (2021)

Add to Collection

Article Mathematics

Hybrid Fruit-Fly Optimization Algorithm with K-Means for Text Document Clustering

Timea Bezdan et al.

Summary: This study introduces a hybrid swarm intelligence algorithm with K-means algorithm for text clustering. The experimental evaluations on both unconstrained functions and text-based documents demonstrate the robustness and superiority of the proposed approach over other state-of-the-art methods.

MATHEMATICS (2021)

Add to Collection

Article Mathematics

A ResNet50-Based Method for Classifying Surface Defects in Hot-Rolled Strip Steel

Xinglong Feng et al.

Summary: This paper proposes a deep learning model for strip steel defect classification, achieving a high classification accuracy by introducing FcaNet and CBAM technologies. Additionally, through ensemble learning optimization, the recognition rate of oxide scale defects and overall defect classification accuracy were significantly improved.

MATHEMATICS (2021)

Add to Collection

Article Computer Science, Information Systems

An Efficient Slime Mould Algorithm Combined With K-Nearest Neighbor for Medical Classification Tasks

Yaser M. Wazery et al.

Summary: The study introduces an improved version of the slime mould algorithm (ISMA) hybridized with opposition-based learning strategy based on the k-nearest neighbor (kNN) classifier for disease classification. The experiment results show the superiority of ISMA-kNN in various evaluation metrics compared to other algorithms.

IEEE ACCESS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence