Disease Prediction Research Articles

Intelligent Diagnosis and Prediction of Plant Diseases Based on Efficientnet_B3a

Intelligent Diagnosis and Prediction of Plant Diseases Based on Efficientnet_B3a

Challenges in application of artificial intelligence in healthcare field and response strategies

The rapid development of artificial intelligence in the field of healthcare has greatly improved diagnosis accuracy, disease prediction, personalized treatment and healthcare resource management. However, with the widespread application of medical artificial intelligence, challenges has emerged in the aspects of medical data, model development and evaluation, and societal considerations. Therefore, this study aims to explore challenges in the application of artificial intelligence in healthcare and suggest a series of feasible solutions to improve medical professional and researchers' understanding of medical artificial intelligence and enhance the quality of healthcare in clinical practice.

Predicting Chronic Diseases Using Nonlinear Systems

Healthcare heavily relies on advanced analytics to predict diseases and risks, with an abundance of health data being gathered through IoT and smart healthcare. Nonlinear systems and synchronization techniques play a crucial role in analyzing this data and predicting chronic diseases, such as cancer, cardiometabolic disease, and Parkinson’s disease. Using machine learning and computational intelligence, nonlinear analysis offers valuable insights into the enormous amounts of data collected in smart healthcare settings, enabling more accurate and efficient disease prediction. This chapter explores the various aspects of nonlinear systems and synchronization techniques in predictive analytics, providing a holistic view of their applications in chronic disease prediction

Multiplex cerebrospinal fluid proteomics identifies biomarkers for diagnosis and prediction of Alzheimer's disease.

Recent expansion of proteomic coverage opens unparalleled avenues to unveil new biomarkers of Alzheimer's disease (AD). Among 6,361 cerebrospinal fluid (CSF) proteins analysed from the ADNI database, YWHAG performed best in diagnosing both biologically (AUC = 0.969) and clinically (AUC = 0.857) defined AD. Four- (YWHAG, SMOC1, PIGR and TMOD2) and five- (ACHE, YWHAG, PCSK1, MMP10 and IRF1) protein panels greatly improved the accuracy to 0.987 and 0.975, respectively. Their superior performance was validated in an independent external cohort and in discriminating autopsy-confirmed AD versus non-AD, rivalling even canonical CSF ATN biomarkers. Moreover, they effectively predicted the clinical progression to AD dementia and were strongly associated with AD core biomarkers and cognitive decline. Synaptic, neurogenic and infectious pathways were enriched in distinct AD stages. Mendelian randomization did not support the significant genetic link between CSF proteins and AD. Our findings revealed promising high-performance biomarkers for AD diagnosis and prediction, with implications for clinical trials targeting different pathomechanisms.

Development of a machine learning-based model for the prediction and progression of diabetic kidney disease: A single centred retrospective study

BackgroundDiabetic kidney disease (DKD) is a diabetic microvascular complication often characterized by an unpredictable progression. Hence, early detection and recognition of patients vulnerable to progression is crucial. ObjectiveTo develop a prediction model to identify the stages of DKD and the factors contributing to progression to each stage using machine learning. MethodologyA retrospective study was conducted in a South Indian tertiary care hospital and collected the details of patients diagnosed with DKD from January 2017 to January 2022. Bayesian optimization-based machine learning techniques such as classification and regression were employed. The model was developed with the help of an optimization framework that effectively balances classification, prediction accuracy, and explainability. ResultsOf the 311 patients diagnosed with DKD, 227 were selected for the study. A system for predicting DKD has been created for a patient dataset utilizing a variety of machine-learning approaches. The eXtreme gradient (XG) Boost method excelled, achieving 88.75% accuracy, 88.57% precision, 91.4% sensitivity,100% specificity, and 89.49% F1-score. An interpretable data-driven method highlights significant features for early DKD diagnosis. The best explainable prediction model uses the XG Boost classifier, revealing serum uric acid, urea, phosphorous, red blood cells, calcium, and absolute eosinophil count as the major predictors influencing the progression of DKD. In the case of regression models, the gradient boost regressor performed the best, with an R2 score of 0.97. ConclusionMachine learning algorithms can effectively predict the stages of DKD and thus help physicians in providing patients with personalized care at the right time.

Hybrid Deep Learning and Squirrel Search Algorithm for Early Prediction of Heart Disease with IoT-Integrated Health Monitoring Systems

This research proposes a hybrid approach combining deep learning models with the Squirrel Search Algorithm (SSA) for early heart disease prediction. Integrating Internet of Things (IoT) devices for real-time health monitoring enhances predictive capability and provides continuous patient data. The proposed model optimizes feature selection using SSA, ensuring that the most relevant and impactful features are used for prediction, which improves the accuracy and efficiency of the deep learning models. This study aims to provide a robust, real-time heart disease prediction system that can be seamlessly integrated into modern healthcare infrastructures.

Health Analysis and Recommendation Based on Food Using Data Mining

Suitable nutritional diets have been widely recognized as important measures to prevent and control non-communicable diseases (NCDs). However, there is little research on nutritional ingredients in food now, which are beneficial to the rehabilitation of NCDs. In this project, we profoundly analyzed the relationship between nutritional ingredients and diseases by using data mining methods. First, more than n number of diseases was obtained, and we collected the recommended food and taboo food for each disease. The experiments on reallife data show that our method based on data mining improves the performance compared with the traditional statistical approach. We can assist doctors and disease researchers to find out positive nutritional ingredients that are conducive to the rehabilitation of the diseases as accurately as possible. At present, some data is not available, because they are still in the medical verification. The dataset uploaded will be pre-processed, Feature Extraction, noisy data will be removed, and classification of dataset will take places using random forest algorithm based on this analysis the diseases prediction takes places for the food intake by the individual

Liver Diseases Prediction Using Machine Learning with Comparison Graph

Chronic Liver Disease is the leading cause of global death that impacts the massive quantity of humans around the world. This disease is caused by an assortment of elements that harm the liver. For example, obesity, an undiagnosed hepatitis infection, alcohol misuse. Which is responsible for abnormal nerve function, coughing up or vomiting blood, kidney failure, liver failure, jaundice, liver encephalopathy and there are many more. This disease diagnosis is very costly and complicated. Therefore, the goal of this work is to evaluate the performance of different Machine Learning algorithms to reduce the high cost of chronic liver disease diagnosis by prediction. In this work, we used five algorithms Logistic Regression, K Nearest Neighbors, Decision Tree, Support Vector Machine, and Random Forest. The performance of different classification techniques was evaluated on different measurement techniques such as accuracy, precision, recall, f-1 score, and specificity. The analysis result shown the LR achieved the highest accuracy. Moreover, our present study mainly focused on the use of clinical data for liver disease prediction and explore different ways of representing such data through our analysis.

Advancements in Ophthalmic Healthcare with Deep Learning-Driven Segmentation for Multi-Stage Eye Fundus Disease Diagnosis

The global rise in eye diseases highlights the need for advanced diagnostic tools in ophthalmic care. This project introduces a deep learning model for classifying eye diseases, streamlining diagnosis, and improving accuracy. Using real-time images from reputable healthcare facilities like Bajwa Hospital in Punjab and Shang gong Medical Tech in China, the model is fine-tuned to clinical nuances. Segmentation of the optic disc and blood vessels is key for precise retinal structure delineation, enhancing disease identification. Various CNN models, including Mobile Net, Dense Net, Reset, and a custom CNN, were utilized for retinal image analysis. Additionally, the Vision Transformer (ViT) model was integrated to capture intricate patterns. The model is deployed as a web application using Django, HTML, SQLite, and Bootstrap, featuring a secure, user-friendly interface. Users can input images to receive prompt disease predictions, along with verified information on prevention, treatment options, and medications. This system not only automates and improves diagnostic processes but also provides reliable medical guidance.

Exploring COVID-19 Pandemic Disparities with Transcriptomic Meta-analysis from the Perspective of Personalized Medicine

Infection with SARS-CoV2, which is responsible for COVID-19, can lead to differences in disease development, severity and mortality rates depending on gender, age or the presence of certain diseases. Considering that existing studies ignore these differences, this study aims to uncover potential differences attributable to gender, age and source of sampling as well as viral load using bioinformatics and multi-omics approaches. Differential gene expression analyses were used to analyse the phenotypic differences between SARS-CoV-2 patients and controls at the mRNA level. Pathway enrichment analyses were performed at the gene set level to identify the activated pathways corresponding to the differences in the samples. Drug repurposing analysis was performed at the protein level, focusing on host-mediated drug candidates to uncover potential therapeutic differences. Significant differences (i.e. the number of differentially expressed genes and their characteristics) were observed for COVID-19 at the mRNA level depending on the sample source, gender and age of the samples. The results of the pathway enrichment show that SARS-CoV-2 can be combated more effectively in the respiratory tract than in the blood samples. Taking into account the different sample sources and their characteristics, different drug candidates were identified. Evaluating disease prediction, prevention and/or treatment strategies from a personalised perspective is crucial. In this study, we not only evaluated the differences in COVID-19 from a personalised perspective, but also provided valuable data for further experimental and clinical efforts. Our findings could shed light on potential pandemics.

Unsupervised representation learning on high-dimensional clinical data improves genomic discovery and prediction

Although high-dimensional clinical data (HDCD) are increasingly available in biobank-scale datasets, their use for genetic discovery remains challenging. Here we introduce an unsupervised deep learning model, Representation Learning for Genetic Discovery on Low-Dimensional Embeddings (REGLE), for discovering associations between genetic variants and HDCD. REGLE leverages variational autoencoders to compute nonlinear disentangled embeddings of HDCD, which become the inputs to genome-wide association studies (GWAS). REGLE can uncover features not captured by existing expert-defined features and enables the creation of accurate disease-specific polygenic risk scores (PRSs) in datasets with very few labeled data. We apply REGLE to perform GWAS on respiratory and circulatory HDCD—spirograms measuring lung function and photoplethysmograms measuring blood volume changes. REGLE replicates known loci while identifying others not previously detected. REGLE are predictive of overall survival, and PRSs constructed from REGLE loci improve disease prediction across multiple biobanks. Overall, REGLE contain clinically relevant information beyond that captured by existing expert-defined features, leading to improved genetic discovery and disease prediction.

Revolutionizing Gastrointestinal Disorder Management: Cutting-Edge Advances and Future Prospects.

In recent years, remarkable strides have been made in the management of gastrointestinal disorders, transforming the landscape of patient care and outcomes. This article explores the latest breakthroughs in the field, encompassing innovative diagnostic techniques, personalized treatment approaches, and novel therapeutic interventions. Additionally, this article emphasizes the use of precision medicine tailored to individual genetic and microbiome profiles, and the application of artificial intelligence in disease prediction and monitoring. This review highlights the dynamic progress in managing conditions such as inflammatory bowel disease, gastroesophageal reflux disease, irritable bowel syndrome, and gastrointestinal cancers. By delving into these advancements, we offer a glimpse into the promising future of gastroenterology, where multidisciplinary collaborations and cutting-edge technologies converge to provide more effective, patient-centric solutions for individuals grappling with gastrointestinal disorders.

Feature Selection AI Technique for Predicting Chronic Kidney Disease

The kidney is a vital organ that plays a crucial role in eliminating waste and excess water from the bloodstream. When renal function is impaired, the filtration process also ceases. This leads to an elevation of harmful molecules in the body, a condition referred to as chronic kidney disease (CKD). Early-stage chronic kidney disease often lacks noticeable symptoms, making it challenging to detect in its early stages. Diagnosing chronic kidney disease (CKD) typically involves advanced blood and urine tests, but unfortunately, by the time these tests are conducted, the disease may already be life-threatening. Our research focuses on the early prediction of chronic kidney disease (CKD) using machine learning (ML) and deep learning (DL) techniques. Utilized a dataset from the machine learning repository at the University of California, Irvine (UCI) to train various machine learning algorithms in conjunction with a Convolutional Neural Network (CNN) model. The algorithms encompassed in this set are Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and Gradient Boosting (GB). Based on the experimental results, the CNN model achieves a prediction accuracy of precisely 97% after feature selection, the highest among all models tested. Hence, the objective of this project is to develop a deep learning-based prediction model to aid healthcare professionals in the timely identification of chronic kidney disease (CKD), potentially leading to life-saving interventions for patients.

Serum and Urine Metabolic Fingerprints Characterize Renal Cell Carcinoma for Classification, Early Diagnosis, and Prognosis.

Renal cell carcinoma (RCC) is a substantial pathology of the urinary system with a growing prevalence rate. However, current clinical methods have limitations for managing RCC due to the heterogeneity manifestations of the disease. Metabolic analyses are regarded as a preferred noninvasive approach in clinics, which can substantially benefit the characterization of RCC. This study constructs a nanoparticle-enhanced laser desorption ionization mass spectrometry (NELDI MS) to analyze metabolic fingerprints of renal tumors (n=456) and healthy controls (n=200). The classification models yielded the areas under curves (AUC) of 0.938 (95% confidence interval (CI), 0.884-0.967) for distinguishing renal tumors from healthy controls, 0.850 for differentiating malignant from benign tumors (95% CI, 0.821-0.915), and 0.925-0.932 for classifying subtypes of RCC (95% CI, 0.821-0.915). For the early stage of RCC subtypes, the averaged diagnostic sensitivity of 90.5% and specificity of 91.3% in the test set is achieved. Metabolic biomarkers are identified as the potential indicator for subtype diagnosis (p<0.05). To validate the prognostic performance, a predictive model for RCC participants and achieve the prediction of disease (p=0.003) is constructed. The study provides a promising prospect for applying metabolic analytical tools for RCC characterization.

Optimized Ensemble Learning Approach with Explainable AI for Improved Heart Disease Prediction

Recent advances in machine learning (ML) have shown great promise in detecting heart disease. However, to ensure the clinical adoption of ML models, they must not only be generalizable and robust but also transparent and explainable. Therefore, this research introduces an approach that integrates the robustness of ensemble learning algorithms with the precision of Bayesian optimization for hyperparameter tuning and the interpretability offered by Shapley additive explanations (SHAP). The ensemble classifiers considered include adaptive boosting (AdaBoost), random forest, and extreme gradient boosting (XGBoost). The experimental results on the Cleveland and Framingham datasets demonstrate that the optimized XGBoost model achieved the highest performance, with specificity and sensitivity values of 0.971 and 0.989 on the Cleveland dataset and 0.921 and 0.975 on the Framingham dataset, respectively.

E-DigitTool: A New-Fangled Framework for Disease Prediction and Diagnosis in Remote Healthcare Applications

E-DigitTool: A New-Fangled Framework for Disease Prediction and Diagnosis in Remote Healthcare Applications

An Efficient Machine Learning based Framework for Alcoholic Liver Disease Prediction

An Efficient Machine Learning based Framework for Alcoholic Liver Disease Prediction

Optimization of vision transformer-based detection of lung diseases from chest X-ray images

BackgroundRecent advances in Vision Transformer (ViT)-based deep learning have significantly improved the accuracy of lung disease prediction from chest X-ray images. However, limited research exists on comparing the effectiveness of different optimizers for lung disease prediction within ViT models. This study aims to systematically evaluate and compare the performance of various optimization methods for ViT-based models in predicting lung diseases from chest X-ray images.MethodsThis study utilized a chest X-ray image dataset comprising 19,003 images containing both normal cases and six lung diseases: COVID-19, Viral Pneumonia, Bacterial Pneumonia, Middle East Respiratory Syndrome (MERS), Severe Acute Respiratory Syndrome (SARS), and Tuberculosis. Each ViT model (ViT, FastViT, and CrossViT) was individually trained with each optimization method (Adam, AdamW, NAdam, RAdam, SGDW, and Momentum) to assess their performance in lung disease prediction.ResultsWhen tested with ViT on the dataset with balanced-sample sized classes, RAdam demonstrated superior accuracy compared to other optimizers, achieving 95.87%. In the dataset with imbalanced sample size, FastViT with NAdam achieved the best performance with an accuracy of 97.63%.ConclusionsWe provide comprehensive optimization strategies for developing ViT-based model architectures, which can enhance the performance of these models for lung disease prediction from chest X-ray images.

Enhanced heart disease prediction in remote healthcare monitoring using IoT-enabled cloud-based XGBoost and Bi-LSTM

The advancement of medical technology has brought about a significant transformation in remote healthcare monitoring, which is crucial for providing customized care and ongoing observation. This is especially important when it comes to controlling long-term illnesses like high blood pressure, which raises the risk of heart disease considerably, especially in older people. This methodology achieves greater accuracy by combining regular medical monitoring and Electronic Clinical Data (ECD) from complete medical records with physical data from patients' routine medical monitoring. This innovative technique enhances the area of cardiac disease prediction. A technique that uses cutting-edge machine learning models and IoT technology to meet this demand. In particular, we use the powerful Extreme Gradient Boosting (XGBoost) algorithm to effectively examine big datasets and extract important characteristics to improve prediction accuracy. The deep learning model Bidirectional Long Short-Term Memory (Bi-LSTM) is used to further enhance prediction skills to extract complex temporal patterns from patient data. It outperformed naive Bayes, decision trees, and random forests with our approach, achieving a greater prediction accuracy of 99.4 %. With the combination of Internet of Things technologies and sophisticated machine learning models, this paper offers a novel approach to remote healthcare monitoring.

Advancements and Challenges in the Application of Machine Learning for Biomedical Diagnostics and Disease Prediction

This review paper delves into the transformative potential and challenges of Machine Learning (ML) in the field of biomedical diagnostics and disease prediction. With the advent of advanced computational models and an ever-increasing availability of biomedical data, ML has the potential to revolutionize diagnostic methodologies, enhance predictive accuracy, and streamline therapeutic interventions. However, the application of these technologies is not without its challenges, including issues of data quality, algorithmic bias, and ethical concerns, which must be addressed to leverage ML’s full potential effectively.