Machine learning assisted optical diagnostics on a cylindrical surface dielectric barrier discharge DOI
Dimitrios Stefas, Konstantinos Giotis, Laurent Invernizzi

и другие.

Journal of Physics D Applied Physics, Год журнала: 2024, Номер 57(45), С. 455206 - 455206

Опубликована: Авг. 6, 2024

Abstract The present study explores combining machine learning (ML) algorithms with standard optical diagnostics (such as time-integrated emission spectroscopy and imaging) to accurately predict operating conditions assess the uniformity of a cylindrical surface dielectric barrier discharge (SDBD). It is demonstrated that these can provide input data for ML which identifies peculiarities associated pattern at different high voltage waveforms (AC pulsed) amplitudes. By employing unsupervised (principal component analysis (PCA)) supervised (multilayer perceptron (MLP) neural networks) algorithms, applied waveform amplitude are predicted based on correlations/differences identified within large amounts corresponding data. PCA allowed us effectively visualise patterns related amplitudes SDBD through transformation spectroscopic/imaging into principal components (PCs) their projection two-dimensional PCs vector space. Furthermore, an accurate prediction achieved using MLP trained PCs. A particularly interesting aspect this concept involves examining discharge. This was by analysing spectroscopic recorded four regions around two algorithms. These discoveries instrumental in enhancing plasma-induced processes. They open avenues real-time control, monitoring, optimization plasma-based applications across diverse fields such flow control SDBD.

Язык: Английский

Progressive Approaches in Oncological Diagnosis and Surveillance: Real‐Time Impedance‐Based Techniques and Advanced Algorithms DOI
Viswambari Devi Ramaswamy, Michael Keidar

Bioelectromagnetics, Год журнала: 2025, Номер 46(1)

Опубликована: Янв. 1, 2025

ABSTRACT Cancer remains a formidable global health challenge, necessitating the development of innovative diagnostic techniques capable early detection and differentiation tumor/cancerous cells from their healthy counterparts. This review focuses on confluence advanced computational algorithms with noninvasive, label‐free impedance‐based biophysical methodologies—techniques that assess biological processes directly without need for external markers or dyes. elucidates diverse array state‐of‐the‐art technologies, illuminating distinct electrical signatures inherent to cancer vs tissues. Additionally, study probes transformative potential these modalities in recalibrating personalized treatment paradigms. These offer real‐time insights into tumor dynamics, paving way precision‐guided therapeutic interventions. By emphasizing quest continuous vivo monitoring, herald pivotal advancement overarching endeavor combat globally.

Язык: Английский

Процитировано

1
Advanced Data Processing of Pancreatic Cancer Data Integrating Ontologies and Machine Learning Techniques to Create Holistic Health Records DOI Creative Commons
George Manias, Ainhoa Azqueta-Alzúaz, Athanasios Dalianis

и другие.

Sensors, Год журнала: 2024, Номер 24(6), С. 1739 - 1739

Опубликована: Март 7, 2024

The modern healthcare landscape is overwhelmed by data derived from heterogeneous IoT sources and Electronic Health Record (EHR) systems. Based on the advancements in science Machine Learning (ML), an improved ability to integrate process so-called primary secondary fosters provision of real-time personalized decisions. In that direction, innovative mechanism for processing integrating health-related introduced this article. It describes details its internal subcomponents workflows, together with results utilization, validation, evaluation a real-world scenario. also highlights potential integration into Holistic Records (HHRs) utilization advanced ML-based Semantic Web techniques improve quality, reliability, interoperability examined data. viability approach evaluated through datasets pertaining risk identification monitoring related pancreatic cancer. key outcomes innovations are introduction HHRs, which facilitate capturing all health determinants harmonized way, holistic ingestion analysis.

Язык: Английский

Процитировано

6
Unveiling the interaction mechanisms of cold atmospheric plasma and amino acids by machine learning DOI

Zhao‐Nan Chai,

Xucheng Wang, Maksudbek Yusupov

и другие.

Plasma Processes and Polymers, Год журнала: 2024, Номер 21(7)

Опубликована: Апрель 15, 2024

Abstract Plasma medicine has attracted tremendous interest in a variety of medical conditions, ranging from wound healing to antimicrobial applications, even cancer treatment, through the interactions cold atmospheric plasma (CAP) and various biological tissues directly or indirectly. The underlying mechanisms CAP treatment are still poorly understood although oxidative effects with amino acids, peptides, proteins have been explored experimentally. In this study, machine learning (ML) technology is introduced efficiently unveil interaction acids reactive oxygen species (ROS) seconds based on data obtained molecular dynamics (MD) simulations, which performed probe five types ROS timescale hundreds picoseconds but huge computational load several days. reactions typically start H‐abstraction, details breaking formation chemical bonds revealed; modification types, such as nitrosylation, hydroxylation, carbonylation, can be observed. dose also investigated by varying number simulation box, indicating agreement experimental observation. To overcome limits timescales size systems MD deep neural network (DNN) hidden layers constructed according reaction employed predict type probability occurrence only varies. well‐trained DNN effectively accurately processes productions, greatly improves efficiency almost ten orders magnitude compared simulation. This study shows great potential ML underpinning simulations measurements.

Язык: Английский

Процитировано

5
Assessing the role of model choice in parameter identifiability of cancer treatment efficacy DOI Creative Commons

Nadine Kuehle Genannt Botmann,

Hana M. Dobrovolny

Frontiers in Applied Mathematics and Statistics, Год журнала: 2025, Номер 11

Опубликована: Март 24, 2025

Several mathematical models are commonly used to describe cancer growth dynamics. Fitting of these experimental data has not yet determined which particular model best describes growth. Unfortunately, choice is known drastically alter the predictions both future tumor and effectiveness applied treatment. Since there growing interest in using help predict chemotherapy, we need determine if affects estimates chemotherapy efficacy. Here, simulate an vitro study by creating synthetic treatment each seven fit sets other (“wrong”) models. We estimate ε max (the maximum efficacy drug) IC 50 drug concentration at half effect achieved) effort whether use incorrect changes parameters. find that largely weakly practically identifiable no matter generate or data. The more likely be identifiable, but sensitive model, showing poor identifiability when Bertalanffy either

Язык: Английский

Процитировано

0
Harnessing Machine Learning Potential for Personalised Drug Design and Overcoming Drug Resistance DOI
Mohammed Ageeli Hakami

Journal of drug targeting, Год журнала: 2024, Номер 32(8), С. 918 - 930

Опубликована: Июнь 6, 2024

Drug resistance in cancer treatment presents a significant challenge, necessitating innovative approaches to improve therapeutic efficacy. Integrating machine learning (ML) research is promising as ML algorithms outrival analysing complex datasets, identifying patterns, and predicting outcomes. Leveraging diverse data sources such genomic profiles, clinical records, drug response assays, uncovers molecular mechanisms of resistance, enabling personalised treatment, maximising efficacy minimising adverse effects. Various contribute the discovery process— Random Forest Decision Trees predict drug-target interactions aid virtual screening, SVM classify leads on bioactivity data. Neural Networks model QSAR optimise lead compounds K-means clustering group with similar chemical properties aiding compound selection. Gaussian Processes responses, Bayesian infer causal relationships, Autoencoders generate novel compounds, Genetic Algorithms structures. These collectively enhance efficiency success rates design endeavours, from identification optimisation are cost-effective, empowering clinicians real-time monitoring improving patient This review highlights immense potential revolutionising care through effective reduce we have also discussed various limitations gaps understand better.

Язык: Английский

Процитировано

2
Unveiling the Potential: Can Machine Learning Cluster Colorimetric Images of Cold Atmospheric Plasma Treatment? DOI Creative Commons
Gizem Dilara Özdemir, Mehmet Akif Özdemir, Mustafa Şen

и другие.

Advanced Intelligent Systems, Год журнала: 2024, Номер 6(9)

Опубликована: Июнь 27, 2024

In this transformative study, machine learning (ML) and t‐distributed stochastic neighbor embedding (t‐SNE) are employed to interpret intricate patterns in colorimetric images of cold atmospheric plasma (CAP)‐treated water. The focus is on CAP's therapeutic potential, particularly its ability generate reactive oxygen nitrogen species (RONS) that play a crucial role antimicrobial activity. RGB, HSV, LAB, YCrCb, grayscale color spaces extracted from the expression oxidative stress induced by RONS, these features used for unsupervised ML, employing density‐based spatial clustering applications with noise (DBSCAN). DBSCAN model's performance evaluated using homogeneity, completeness, adjusted rand index predictive data distribution graph. best results achieved 3,3′,5,5′‐tetramethylbenzidine–potassium iodide assay solution immediately after treatment, values 0.894, 0.996, 0.826. t‐SNE further conducted best‐case scenario evaluate efficacy find combination better present results. Correspondingly, enhances adeptly handles challenging points. approach pioneers dynamic comprehensive solutions, showcasing ML's precision t‐SNE's visualization. Through innovative fusion, complex relationships unraveled, marking paradigm shift biomedical analytical methodologies.

Язык: Английский

Процитировано

1
Machine learning assisted optical diagnostics on a cylindrical surface dielectric barrier discharge DOI
Dimitrios Stefas, Konstantinos Giotis, Laurent Invernizzi

и другие.

Journal of Physics D Applied Physics, Год журнала: 2024, Номер 57(45), С. 455206 - 455206

Опубликована: Авг. 6, 2024

Abstract The present study explores combining machine learning (ML) algorithms with standard optical diagnostics (such as time-integrated emission spectroscopy and imaging) to accurately predict operating conditions assess the uniformity of a cylindrical surface dielectric barrier discharge (SDBD). It is demonstrated that these can provide input data for ML which identifies peculiarities associated pattern at different high voltage waveforms (AC pulsed) amplitudes. By employing unsupervised (principal component analysis (PCA)) supervised (multilayer perceptron (MLP) neural networks) algorithms, applied waveform amplitude are predicted based on correlations/differences identified within large amounts corresponding data. PCA allowed us effectively visualise patterns related amplitudes SDBD through transformation spectroscopic/imaging into principal components (PCs) their projection two-dimensional PCs vector space. Furthermore, an accurate prediction achieved using MLP trained PCs. A particularly interesting aspect this concept involves examining discharge. This was by analysing spectroscopic recorded four regions around two algorithms. These discoveries instrumental in enhancing plasma-induced processes. They open avenues real-time control, monitoring, optimization plasma-based applications across diverse fields such flow control SDBD.

Язык: Английский

Процитировано

1