Published: Oct. 11, 2024
Language: Английский
Chemical Engineering Journal, Journal Year: 2024, Volume and Issue: 495, P. 153382 - 153382
Published: June 21, 2024
Language: Английский
Citations
9
Cell Reports Physical Science, Journal Year: 2023, Volume and Issue: 4(8), P. 101535 - 101535
Published: Aug. 1, 2023
The regulation of most physiological processes relies on a state equilibrium called homeostasis, which is achieved through biological control loop involving sensors and actuators. However, disease aging can disrupt these loops, leading to impaired or slower homeostatic mechanisms. Bioelectronic devices offer the opportunity interface artificial technology with systems, enabling measurement specific using To effectively interact complex dynamics adapt changing environmental conditions, interfacing must be capable real-time sensing response. In this context, we propose that machine learning significantly enhance capabilities bioelectronics by facilitating processing sensor actuator data. By utilizing machine-learning-driven bioelectronics, maintain regulate system responses more compared traditional approaches. This advancement holds promising implications for bioelectronic medicine precision medicine, particularly in repairing
Language: Английский
Citations
13
BMC Bioinformatics, Journal Year: 2023, Volume and Issue: 24(1)
Published: April 25, 2023
Abstract Background Wound healing involves careful coordination among various cell types carrying out unique or even multifaceted functions. The abstraction of this complex dynamic process into four primary wound stages is essential to the study care for timing treatment and tracking progression. For example, a that may promote in inflammatory stage prove detrimental proliferative stage. Additionally, time scale individual responses varies widely across within same species. Therefore, robust method assess can help advance translational work from animals humans. Results In work, we present data-driven model robustly identifies dominant using transcriptomic data biopsies gathered mouse human wounds, both burn surgical. A training dataset composed publicly available arrays used derive 58 shared genes are commonly differentially expressed. They divided 5 clusters based on temporal gene expression dynamics. represent 5-dimensional parametric space containing trajectory. We then create mathematical classification algorithm demonstrate it distinguish between healing: hemostasis, inflammation, proliferation, remodeling. Conclusions an detection expression. This suggests there universal characteristics despite seeming disparities species wounds. Our performs well wounds surgical types. has potential serve as diagnostic tool precision by providing way progression with more accuracy finer resolution compared visual indicators. increases preventive action.
Language: Английский
Citations
12
Applied Energy, Journal Year: 2024, Volume and Issue: 359, P. 122684 - 122684
Published: Jan. 25, 2024
Language: Английский
Citations
4
World Journal of Clinical Cases, Journal Year: 2025, Volume and Issue: 13(15)
Published: Jan. 21, 2025
The integration of artificial intelligence (AI) and multiomics has transformed clinical life sciences, enabling precision medicine redefining disease understanding. Scientific publications grew significantly from 2.1 million in 2012 to 3.3 2022, with AI research tripling during this period. Multiomics fields, including genomics proteomics, also advanced, exemplified by the Human Proteome Project achieving a 90% complete blueprint 2021. This growth highlights opportunities challenges integrating into reporting. A review studies case reports was conducted evaluate integration. Key areas analyzed included diagnostic accuracy, predictive modeling, personalized treatment approaches driven tools. Case examples were studied assess impacts on decision-making. enhanced data integration, insights, personalization. Fields like radiomics, genomics, proteomics improved diagnostics guided therapy. For instance, “AI oncopathomics, surgomics project” combined radiomics for surgical decision-making, preoperative, intraoperative, postoperative interventions. applications predicted conditions delirium monitored cancer progression using genomic imaging data. enable standardized analysis, dynamic updates, modeling reports. Traditional often lack objectivity, but enhances reproducibility decision-making processing large datasets. Challenges include standardization, biases, ethical concerns. Overcoming these barriers is vital optimizing advancing medicine. revolutionizing practice. Standardizing reporting addressing ethics quality will unlock their full potential. Emphasizing collaboration transparency essential leveraging tools improve patient care scientific communication.
Language: Английский
Citations
0
Medicina, Journal Year: 2025, Volume and Issue: 61(2), P. 188 - 188
Published: Jan. 22, 2025
Background and Objectives: Recent research has focused on exploring the relationships between various factors associated with headaches understanding their impact individuals’ psychological states. Utilizing statistical methods machine learning models, these studies aim to analyze predict develop effective approaches for headache management prevention. Materials Methods: Analyzing data from 398 patients (train set = 318 test 80), we investigated influence of features outcomes such as depression, anxiety, intensity using linear regression. The study employed a mixed-methods approach, combining medical records, interviews, surveys gather comprehensive participants’ experiences effects. Results: Machine including Random Forest (utilized Headache Impact Test-6, Patient Health Questionnaire-9, Generalized Anxiety Disorder-7) Support Vector Regression (applied Migraine Disability Assessment), revealed key contributing each outcome through Shapley values, while regression provided additional insights. Frequent analgesic medication emerged significant predictor poorer life quality (Headache root mean squared error 7.656) increased depression (Patient 5.07) anxiety (Generalized Disorder-7, 4.899) in model. However, interpreting importance complex models like supportive vector poses challenges, determining causality usage pain severity was not feasible. Conclusions: Our underscores considering individual characteristics optimizing treatment strategies patients.
Language: Английский
Citations
0
Lecture notes in networks and systems, Journal Year: 2025, Volume and Issue: unknown, P. 401 - 412
Published: Jan. 1, 2025
Language: Английский
Citations
0
International Research Journal of Multidisciplinary Technovation, Journal Year: 2025, Volume and Issue: unknown, P. 66 - 76
Published: April 16, 2025
The objective of this research is to explore and compare the performance several Deep Learning (DL) models identify most accurate classification model for predicting brain tumors using MRI images. utilizes dataset 450 images which include healthy cases, cases with grade 1 & 2 benign tumors, 3 4 malignant tumor cases. further divided into training, validation, testing sets. Each then trained validated on training validation sets tested set overall assessed compared. results demonstrated unique trends among models, where CNN ResNet50 have consistently performed best highest accuracy least data loss. VGG16 VGG19 also exemplified great results, although they utilised more epochs achieve similar accuracy. Based study, it concluded that appropriate DL architecture should be selected especially in medical fields. In general, residual networks showed chosen when important requirement. potential application outcomes can applied field medicine mainly identification, classification, detection, prediction various diseases.
Language: Английский
Citations
0
Frontiers in Animal Science, Journal Year: 2025, Volume and Issue: 6
Published: April 28, 2025
Current achievements in omics technologies have modernized livestock biotechnology, offering extraordinary comprehension of animal productivity, health, and reproduction. This extensive study examines the integration implementation approaches, genomics, transcriptomics, proteomics, metabolomics, epigenomics production systems. We reconnoitered how genomic novelties redesign breeding strategies with marker-assisted selection CRISPR-based gene editing. Together, transcriptomic analyses indicate key insights into expression patterns governing economically essential traits such as muscle growth milk production. also shows role proteomics identifying biomarkers for health surveillance product quality improvement along which contributes to understanding feed efficiency disease resistance. Particular attention is given studies exploring DNA methylation histone modifications reproductive efficacy, underlining their importance fertility embryonic development. Integrating multi-omics data through systems biology approaches discussed, demonstrating its perspective evolving precision observed improve assisted (ART) by better molecular mechanisms underlying embryo While acknowledging potential these technologies, we discuss critical challenges, complications, ethical respect genetic modification. review outlines prospect directions highlighting crucial addressing global food security contests productivity efficiency. suggests that continuous might be cause determination future sustainable
Language: Английский
Citations
0
Drug Healthcare and Patient Safety, Journal Year: 2025, Volume and Issue: Volume 17, P. 135 - 143
Published: May 1, 2025
Language: Английский
Citations
0