USING ARTIFICIAL INTELLIGENCE FOR BIOMARKER ANALYSIS IN CLINICAL DIAGNOSTICS DOI
П. В. Селиверстов, V. Kutsenko,

V. G. Gorelova

et al.

Molekulyarnaya Meditsina (Molecular medicine), Journal Year: 2024, Volume and Issue: unknown, P. 31 - 40

Published: Nov. 6, 2024

Introduction. Artificial intelligence (AI) technologies are becoming crucial in clinical diagnostics due to their ability process and interpret large volumes of data. The implementation AI for biomarker analysis opens new opportunities personalized medicine, offering more accurate individualized approaches disease diagnosis treatment. relevance this review stems from the need systematize recent advances application analysis, which is critical early prediction chronic non-communicable diseases (NCDs). Material methods. peer-reviewed scientific publications reports leading research centers over past five years was conducted. Studies on algorithms analyzing genomic, proteomic, metabolomic biomarkers were reviewed, including machine learning methods deep neural networks. Special attention paid integration multi-marker panels improving accuracy cardiovascular, digestive, respiratory, endocrine system diseases, as well oncological neurodegenerative pathologies. Results. has significantly increased sensitivity specificity diagnostics, especially complex cases requiring multiple parameters. effectiveness been demonstrated lung, breast, colorectal cancer, cardiovascular complications NCDs progression, diabetes mellitus Alzheimer’s disease. AI’s significant contribution discovery biomarkers, optimization treatment, improvement therapeutic strategies noted. Conclusion. use become a breakthrough medical particularly oncology, cardiology, diseases. technology allows data about various contributes creating models prediction. Further development associated with advancement overcoming ethical regulatory barriers, will expand capabilities practice.

Language: Английский

Impact of Metabolites from Foodborne Pathogens on Cancer DOI Creative Commons
Alice Njolke Mafe, Dietrich Büsselberg

Foods, Journal Year: 2024, Volume and Issue: 13(23), P. 3886 - 3886

Published: Dec. 1, 2024

Foodborne pathogens are microorganisms that cause illness through contamination, presenting significant risks to public health and food safety. This review explores the metabolites produced by these pathogens, including toxins secondary metabolites, their implications for human health, particularly concerning cancer risk. We examine various such as Salmonella sp., Campylobacter Escherichia coli, Listeria monocytogenes, detailing specific of concern carcinogenic mechanisms. study discusses analytical techniques detecting chromatography, spectrometry, immunoassays, along with challenges associated detection. covers effective control strategies, processing techniques, sanitation practices, regulatory measures, emerging technologies in pathogen control. manuscript considers broader highlighting importance robust policies, awareness, education. identifies research gaps innovative approaches, recommending advancements detection methods, preventive policy improvements better manage foodborne metabolites.

Language: Английский

Citations

11
Recent Developments in Monitoring of Organophosphorus Pesticides in Food Samples DOI Creative Commons
Kokob Teshome Wondimu, Abiyot Kelecha Geletu, Welela Meka Kedir

et al.

Journal of Agriculture and Food Research, Journal Year: 2025, Volume and Issue: 19, P. 101709 - 101709

Published: Feb. 7, 2025

Language: Английский

Citations

1
Residue analysis, dissipation dynamics, and risk assessment of ipconazole in paddy environment by a modified QuEChERS/HPLC-MS method DOI
Zhonggui Hu, Yuqi Li, Wenwen Zhou

et al.

Journal of Food Composition and Analysis, Journal Year: 2025, Volume and Issue: unknown, P. 107516 - 107516

Published: March 1, 2025

Language: Английский

Citations

0
Innovative applications and future perspectives of chromatography-mass spectrometry in drug research DOI Creative Commons
Hong Cai,

Xue Xing,

Ying Su

et al.

Frontiers in Pharmacology, Journal Year: 2025, Volume and Issue: 16

Published: March 26, 2025

Chromatography coupled with mass spectrometry (MS) has emerged as a cornerstone analytical technique in drug research. Over the years, advancements chromatography-MS have significantly enhanced its capabilities, leading to improved sensitivity, specificity, and throughput. This review explores innovative applications of research, particularly focusing on role absorption, distribution, metabolism, excretion (ADME), toxicity evaluation, personalized medicine. It also addresses future perspectives this powerful technique, including challenges potential solutions, highlights how emerging trends such high spatial resolution imaging multimodal integration could revolutionize discovery development. Through these innovations, promises contribute substantially development more effective, safer, therapeutic interventions.

Language: Английский

Citations

0
Application of LLMs/Transformer-Based Models for Metabolite Annotation in Metabolomics DOI
Yijiang Liu, Feifan Zhang, Ying Ge

et al.

Published: April 15, 2025

Review Application of LLMs/Transformer-Based Models for Metabolite Annotation in Metabolomics Yijiang Liu 1,†, Feifan Zhang 2,†, Yifei Ge 2, Qiao 3, Siyu He 4, and Xiaotao Shen 1,2,5,* 1 School Chemistry, Chemical Engineering Biotechnology, Nanyang Technological University, Singapore 637459, 2 Lee Kong Chian Medicine, 308232, 3 Department Statistics, Stanford University Palo Alto, CA 94304, USA 4 Biomedical Data Science, 5 Phenome Center, 636921, * Correspondence: [email protected] † These authors contributed equally to this work. Received: 20 December 2024; Revised: 6 January 2025; Accepted: March Published: 15 April 2025 Abstract: Liquid Chromatography-Mass Spectrometry (LC-MS) untargeted metabolomics has become a cornerstone modern biomedical research, enabling the analysis complex metabolite profiles biological systems. However, annotation, key step LC-MS metabolomics, remains major challenge due limited coverage existing reference libraries vast diversity natural metabolites. Recent advancements large language models (LLMs) powered by Transformer architecture have shown significant promise addressing challenges data-intensive fields, including metabolomics. LLMs, which when fine-tuned with domain-specific datasets such as mass spectrometry (MS) spectra chemical property databases, together other Transformer-based models, excel at capturing relationships processing large-scale data significantly enhance annotation. Various tasks include retention time prediction, theoretical MS2 generation. For example, methods LipiDetective MS2Mol potential machine learning lipid species prediction de novo molecular structure annotation directly from spectra. tools leverage transformer principles their integration LLM frameworks could further expand utility Moreover, ability LLMs integrate multi-modal datasets—spanning genomics, transcriptomics, metabolomics—positions them powerful systems-level analysis. This review highlights application future perspectives incorporating multiomics. Such transformative paves way enhanced accuracy, expanded coverage, deeper insights into metabolic processes, ultimately driving precision medicine systems biology.

Language: Английский

Citations

0
Polymeric sensors for blood analysis: current and future scope of research DOI

Dipak Thikar,

Gaurav Gopal Naik,

Sarojini Verma

et al.

Polymer International, Journal Year: 2025, Volume and Issue: unknown

Published: April 28, 2025

Abstract This review explores the advancements in polymer‐based sensors for blood analysis, emphasizing online detection of key components such as uric acid, creatinine, urea, bilirubin, cholesterol, total proteins, amino acids and hormones. It categorizes polymer into electrochemical, optical molecularly imprinted polymers, providing insights their working mechanisms advantages biomarker identification. Recent innovations are highlighted to evaluate current state sensor technology terms selectivity, sensitivity real‐time monitoring capabilities. Challenges stability issues, biofouling compliance also addressed. The underscores transformative potential these diagnostics, role enhancing patient care through convenient point‐of‐care healthcare testing. © 2025 Society Chemical Industry.

Language: Английский

Citations

0
Metabolomics as a tool for understanding and treating triple-negative breast cancer DOI
Gyas Khan, Md Sadique Hussain, Sarfaraz Ahmad

et al.

Naunyn-Schmiedeberg s Archives of Pharmacology, Journal Year: 2025, Volume and Issue: unknown

Published: May 2, 2025

Language: Английский

Citations

0
Advances in Metabolomics: A Comprehensive Review of Type 2 Diabetes and Cardiovascular Disease Interactions DOI Open Access
Lilian Fernandes Silva, Markku Laakso

International Journal of Molecular Sciences, Journal Year: 2025, Volume and Issue: 26(8), P. 3572 - 3572

Published: April 10, 2025

Type 2 diabetes (T2D) and cardiovascular diseases (CVDs) are major public health challenges worldwide. Metabolomics, the exhaustive assessment of metabolites in biological systems, offers important insights regarding metabolic disturbances related to these disorders. Recent advances toward integration metabolomics into clinical practice facilitate discovery novel biomarkers that can improve diagnosis, prognosis, treatment T2D CVDs discussed this review. Metabolomics potential characterize key alterations associated with disease pathophysiology treatment. is a heterogeneous develops through diverse pathophysiological processes molecular mechanisms; therefore, disease-causing pathways not completely understood. studies have identified several robust clusters variants representing biologically meaningful, distinct pathways, such as beta cell proinsulin cluster pancreatic insulin secretion, obesity, lipodystrophy, liver/lipid cluster, glycemia, blood pressure, syndrome different causing resistance. Regarding CVDs, recent allowed metabolomic profile delineate contribute atherosclerosis heart failure, well development targeted therapy. This review also covers role integrated genomics other omics platforms better understand mechanisms, along transition precision medicine. further investigates use multi-metabolite modeling enhance risk prediction models for predicting first occurrence adverse events among individuals T2D, highlighting value approaches optimizing preventive therapeutic used practice.

Language: Английский

Citations

0
USING ARTIFICIAL INTELLIGENCE FOR BIOMARKER ANALYSIS IN CLINICAL DIAGNOSTICS DOI
П. В. Селиверстов, V. Kutsenko,

V. G. Gorelova

et al.

Molekulyarnaya Meditsina (Molecular medicine), Journal Year: 2024, Volume and Issue: unknown, P. 31 - 40

Published: Nov. 6, 2024

Introduction. Artificial intelligence (AI) technologies are becoming crucial in clinical diagnostics due to their ability process and interpret large volumes of data. The implementation AI for biomarker analysis opens new opportunities personalized medicine, offering more accurate individualized approaches disease diagnosis treatment. relevance this review stems from the need systematize recent advances application analysis, which is critical early prediction chronic non-communicable diseases (NCDs). Material methods. peer-reviewed scientific publications reports leading research centers over past five years was conducted. Studies on algorithms analyzing genomic, proteomic, metabolomic biomarkers were reviewed, including machine learning methods deep neural networks. Special attention paid integration multi-marker panels improving accuracy cardiovascular, digestive, respiratory, endocrine system diseases, as well oncological neurodegenerative pathologies. Results. has significantly increased sensitivity specificity diagnostics, especially complex cases requiring multiple parameters. effectiveness been demonstrated lung, breast, colorectal cancer, cardiovascular complications NCDs progression, diabetes mellitus Alzheimer’s disease. AI’s significant contribution discovery biomarkers, optimization treatment, improvement therapeutic strategies noted. Conclusion. use become a breakthrough medical particularly oncology, cardiology, diseases. technology allows data about various contributes creating models prediction. Further development associated with advancement overcoming ethical regulatory barriers, will expand capabilities practice.

Language: Английский

Citations

0