Artificial intelligence assisted food science and nutrition perspective for smart nutrition research and healthcare

Systems Microbiology and Biomanufacturing, Journal Year: 2023, Volume and Issue: 4(1), P. 86 - 101

Published: Aug. 9, 2023

Language: Английский

---

aiGeneR 1.0: An Artificial Intelligence Technique for the Revelation of Informative and Antibiotic Resistant Genes in Escherichia coli

Frontiers in Bioscience-Landmark, Journal Year: 2024, Volume and Issue: 29(2)

Published: Feb. 22, 2024

Background: There are several antibiotic resistance genes (ARG) for the Escherichia coli (E. coli) bacteria that cause urinary tract infections (UTI), and it is therefore important to identify these ARG. Artificial Intelligence (AI) has been used previously in field of gene expression data, but never adopted detection classification bacterial We hypothesize, if data correctly conferred, right features selected, Deep Learning (DL) models optimized, then (i) non-linear DL would perform better than Machine (ML) models, (ii) leads higher accuracy, (iii) can hub genes, and, (iv) pathways accurately. have designed aiGeneR, first its kind system uses DL-based ARG E. data. Methodology: The aiGeneR consists a tandem connection quality control embedded with feature extraction AI-based cross-validation approach evaluate performance using precision, recall, F1-score. Further, we analyzed effect sample size ensuring generalization compare against power analysis. was validated scientifically biologically pathways. benchmarked two linear other AI models. Results: identifies tetM (an ARG) showed an accuracy 93% area under curve (AUC) 0.99 (p < 0.05). mean 22% compared aiGeneR. Conclusions: successfully detected validating our four hypotheses.

Language: Английский

---

UltraAIGenomics: Artificial Intelligence-Based Cardiovascular Disease Risk Assessment by Fusion of Ultrasound-Based Radiomics and Genomics Features for Preventive, Personalized and Precision Medicine: A Narrative Review

Reviews in Cardiovascular Medicine, Journal Year: 2024, Volume and Issue: 25(5)

Published: May 22, 2024

Cardiovascular disease (CVD) diagnosis and treatment are challenging since symptoms appear late in the disease’s progression. Despite clinical risk scores, cardiac event prediction is inadequate, many at-risk patients not adequately categorised by conventional factors alone. Integrating genomic-based biomarkers (GBBM), specifically those found plasma and/or serum samples, along with novel non-invasive radiomic-based (RBBM) such as plaque area burden can improve overall specificity of CVD risk. This review proposes two hypotheses: (i) RBBM GBBM have a strong correlation be used to detect severity stroke precisely, (ii) introduces proposed artificial intelligence (AI)—based preventive, precision, personalized (aiP3) CVD/Stroke model. The PRISMA search selected 246 studies for It showed that using biomarkers, deep learning (DL) modelscould stratification aiP3 framework. Furthermore, we present concise overview platelet function, complete blood count (CBC), diagnostic methods. As part AI paradigm, discuss explainability, pruning, bias, benchmarking against previous their potential impacts. integration GBBM, an innovative solution streamlined DL paradigm predicting combination powerful assessment paradigm. model signifies promising advancement assessment.

Language: Английский

---

Four Transformer-Based Deep Learning Classifiers Embedded with an Attention U-Net-Based Lung Segmenter and Layer-Wise Relevance Propagation-Based Heatmaps for COVID-19 X-ray Scans

Diagnostics, Journal Year: 2024, Volume and Issue: 14(14), P. 1534 - 1534

Published: July 16, 2024

Background: Diagnosing lung diseases accurately is crucial for proper treatment. Convolutional neural networks (CNNs) have advanced medical image processing, but challenges remain in their accurate explainability and reliability. This study combines U-Net with attention Vision Transformers (ViTs) to enhance disease segmentation classification. We hypothesize that Attention will accuracy ViTs improve classification performance. The methodologies shed light on model decision-making processes, aiding clinical acceptance. Methodology: A comparative approach was used evaluate deep learning models segmenting classifying illnesses using chest X-rays. segmentation, architectures consisting of four CNNs were investigated Methods like Gradient-weighted Class Activation Mapping plus (Grad-CAM++) Layer-wise Relevance Propagation (LRP) provide by identifying areas influencing decisions. Results: results support the conclusion are outstanding disorders. obtained a Dice Coefficient 98.54% Jaccard Index 97.12%. outperformed tasks 9.26%, reaching an 98.52% MobileViT. An 8.3% increase seen while moving from raw data segmented Techniques Grad-CAM++ LRP provided insights into processes models. Conclusions: highlights benefits integrating analyzing diseases, demonstrating importance settings. Emphasizing clarifies enhancing confidence AI solutions perhaps acceptance improved healthcare results.

Language: Английский

---

GeneAI 3.0: powerful, novel, generalized hybrid and ensemble deep learning frameworks for miRNA species classification of stationary patterns from nucleotides

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: March 26, 2024

Abstract Due to the intricate relationship between small non-coding ribonucleic acid (miRNA) sequences, classification of miRNA species, namely Human, Gorilla, Rat, and Mouse is challenging. Previous methods are not robust accurate. In this study, we present AtheroPoint’s GeneAI 3.0, a powerful, novel, generalized method for extracting features from fixed patterns purines pyrimidines in each sequence ensemble paradigms machine learning (EML) convolutional neural network (CNN)-based deep (EDL) frameworks. 3.0 utilized five conventional (Entropy, Dissimilarity, Energy, Homogeneity, Contrast), three contemporary (Shannon entropy, Hurst exponent, Fractal dimension) features, generate composite feature set given sequences which were then passed into our ML DL framework. A 11 new classifiers was designed consisting 5 EML 6 EDL binary/multiclass classification. It benchmarked against 9 solo (SML), (SDL), 12 hybrid (HDL) models, resulting total + 27 = 38 models designed. Four hypotheses formulated validated using explainable AI (XAI) as well reliability/statistical tests. The order mean performance accuracy (ACC)/area-under-the-curve (AUC) 24 was: > HDL SDL. with CNN layers superior that without by 0.73%/0.92%. Mean SML improvements ACC/AUC 6.24%/6.46%. performed significantly better than increase 7.09%/6.96%. tool produced expected XAI plots, statistical tests showed significant p -values. Ensemble highly effective effectively classifying sequences.

Language: Английский

---

Polygenic Risk Score for Cardiovascular Diseases in Artificial Intelligence Paradigm: A Review

Journal of Korean Medical Science, Journal Year: 2023, Volume and Issue: 38(46)

Published: Jan. 1, 2023

Cardiovascular disease (CVD) related mortality and morbidity heavily strain society.The relationship between external risk factors our genetics have not been well established.It is widely acknowledged that environmental influence individual behaviours play a significant role in CVD vulnerability, leading to the development of polygenic scores (PRS).We employed PRISMA search method locate pertinent research literature extensively review artificial intelligence (AI)-based PRS models for prediction.Furthermore, we analyzed compared conventional vs. AI-based solutions PRS.We summarized recent advances understanding use prediction CVD.Our study proposes three hypotheses: i) Multiple genetic variations

Language: Английский

---

Deep learning approach for cardiovascular disease risk stratification and survival analysis on a Canadian cohort

The International Journal of Cardiovascular Imaging, Journal Year: 2024, Volume and Issue: 40(6), P. 1283 - 1303

Published: April 28, 2024

Language: Английский

---

HyEpiSeiD: a hybrid convolutional neural network and gated recurrent unit model for epileptic seizure detection from electroencephalogram signals

Brain Informatics, Journal Year: 2024, Volume and Issue: 11(1)

Published: Aug. 21, 2024

Abstract Epileptic seizure (ES) detection is an active research area, that aims at patient-specific ES with high accuracy from electroencephalogram (EEG) signals. The early of crucial for timely medical intervention and prevention further injuries the patients. This work proposes a robust deep learning framework called HyEpiSeiD extracts self-trained features pre-processed EEG signals using hybrid combination convolutional neural network followed by two gated recurrent unit layers performs prediction based on those extracted features. proposed evaluated public datasets, UCI Epilepsy Mendeley datasets. model achieved 99.01% 97.50% classification accuracy, respectively, outperforming most state-of-the-art methods in epilepsy domain.

Language: Английский

---

Cardiovascular Disease Risk Stratification Using Hybrid Deep Learning Paradigm: First of Its Kind on Canadian Trial Data

Diagnostics, Journal Year: 2024, Volume and Issue: 14(17), P. 1894 - 1894

Published: Aug. 28, 2024

The risk of cardiovascular disease (CVD) has traditionally been predicted via the assessment carotid plaques. In proposed study, AtheroEdge™ 3.0

Language: Английский

---

Attention-based hybrid deep learning models and its scientific validation for cardiovascular disease risk stratification

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 108, P. 107824 - 107824

Published: April 8, 2025

Language: Английский

---