Machine Learning-Driven Transcriptome Analysis of Keratoconus for Predictive Biomarker Identification DOI Creative Commons

S.-H. Chang,

Lung‐Kun Yeh, Kuo-Hsuan Hung

et al.

Biomedicines, Journal Year: 2025, Volume and Issue: 13(5), P. 1032 - 1032

Published: April 24, 2025

Background: Keratoconus (KTCN) is a multifactorial disease characterized by progressive corneal degeneration. Recent studies suggest that gene expression analysis of corneas may uncover potential novel biomarkers involved in matrix remodeling. However, identifying reliable combinations are linked to risk or progression remains significant challenge. Objective: This study employed multiple machine learning algorithms analyze the transcriptomes keratoconus patients, feature and their functional associations, with aim enhancing understanding pathogenesis. Methods: We analyzed GSE77938 (PRJNA312169) dataset for differential (DGE) performed set enrichment (GSEA) using Kyoto Encyclopedia Genes Genomes (KEGG) pathways identify enriched versus controls. Machine were then used sets, SHapley Additive exPlanations (SHAP) applied assess contribution key genes model’s predictions. Selected further through Gene Ontology (GO) explore roles biological processes cellular functions. Results: models, including XGBoost, Random Forest, Logistic Regression, SVM, identified important associated keratoconus, 15 notable appearing across such as IL1R1, JUN, CYBB, CXCR4, KRT13, KRT14, S100A8, S100A9, others. The under-expressed KTCN mechanical resistance epidermis (KRT14, KRT15) inflammation (S100A8/A9, CXCR4), compared GO highlighted S100A8/A9 complex its primarily related cytoskeleton organization, inflammation, immune response. Furthermore, we expanded our incorporating additional datasets from PRJNA636666 PRJNA1184491, thereby offering broader representation features increasing generalizability results diverse cohorts. Conclusions: differing sets XGBoost SVM reflect distinct but complementary aspects pathophysiology. Meanwhile, captured chemotactic regulators (e.g., suggesting upstream inflammatory signaling pathways. structural epithelial differentiation markers S100A8/A9), possibly reflecting downstream tissue remodeling stress responses. Our findings provide research platform evaluation learning-based approaches, valuable insights into pathogenesis therapeutic targets.

Language: Английский

A Review of ARIMA vs. Machine Learning Approaches for Time Series Forecasting in Data Driven Networks DOI Creative Commons
Vaia I. Kontopoulou, Athanasios D. Panagopoulos, Iοannis Kakkos

et al.

Future Internet, Journal Year: 2023, Volume and Issue: 15(8), P. 255 - 255

Published: July 30, 2023

In the broad scientific field of time series forecasting, ARIMA models and their variants have been widely applied for half a century now due to mathematical simplicity flexibility in application. However, with recent advances development efficient deployment artificial intelligence techniques, view is rapidly changing, shift towards machine deep learning approaches becoming apparent, even without complete evaluation superiority new approach over classic statistical algorithms. Our work constitutes an extensive review published literature regarding comparison algorithms forecasting problems, as well combination these two hybrid statistical-AI wide variety data applications (finance, health, weather, utilities, network traffic prediction). has shown that AI display better prediction performance most applications, few notable exceptions analyzed our Discussion Conclusions sections, while steadily outperform individual parts, utilizing best algorithmic features both worlds.

Language: Английский

Citations

97
Data-Driven Modelling of Corrosion Behaviour in Coated Porous Transport Layers for PEM Water Electrolyzers DOI Creative Commons
Pramoth Varsan Madhavan,

Leila Moradizadeh,

Samaneh Shahgaldi

et al.

Artificial Intelligence Chemistry, Journal Year: 2025, Volume and Issue: unknown, P. 100086 - 100086

Published: Feb. 1, 2025

Language: Английский

Citations

2
Predicting Potato Crop Yield with Machine Learning and Deep Learning for Sustainable Agriculture DOI Creative Commons

El-Sayed M. El-kenawy,

Amel Ali Alhussan, Nima Khodadadi

et al.

Potato Research, Journal Year: 2024, Volume and Issue: unknown

Published: July 13, 2024

Abstract Potatoes are an important crop in the world; they main source of food for a large number people globally and also provide income many people. The true forecasting potato yields is determining factor rational use maximization agricultural practices, responsible management resources, wider regions’ security. latest discoveries machine learning deep new directions to yield prediction models more accurately sparingly. From study, we evaluated different types predictive models, including K-nearest neighbors (KNN), gradient boosting, XGBoost, multilayer perceptron that learning, as well graph neural networks (GNNs), gated recurrent units (GRUs), long short-term memory (LSTM), which popular models. These on basis some performance measures like mean squared error (MSE), root (RMSE), absolute (MAE) know how much predict yields. terminal results show although boosting XGBoost algorithms good at prediction, GNNs LSTMs not only have advantage high accuracy but capture complex spatial temporal patterns data. Gradient resulted MSE 0.03438 R 2 0.49168, while had 0.03583 0.35106. Out all displayed 0.02363 0.51719, excelling overall performance. GRUs were reported be very promising well, with comprehending 0.03177 grabbing 0.03150. findings underscore potential advanced support sustainable practices informed decision-making context farming.

Language: Английский

Citations

10
Modelling Anti-Corrosion Coating Performance of Metallic Bipolar Plates for PEM Fuel Cells: A Machine Learning Approach DOI Creative Commons
Pramoth Varsan Madhavan, Samaneh Shahgaldi, Xianguo Li

et al.

Energy and AI, Journal Year: 2024, Volume and Issue: 17, P. 100391 - 100391

Published: June 26, 2024

Proton exchange membrane (PEM) fuel cells have significant potential for clean power generation, yet challenges remain in enhancing their performance, durability, and cost-effectiveness, particularly concerning metallic bipolar plates, which are pivotal lightweight compact cell stacks. Protective coatings commonly employed to combat plate corrosion enhance water management within Conventional methods predicting coating performance terms of resistance involve complex physical-electrochemical modelling extensive experimentation, with time cost. In this study machine learning techniques model diamond-like-carbon varying thicknesses deposited on SS316L considered, is evaluated using potentiodynamic polarization electrochemical impedance spectroscopy. The obtained experimental data split into two datasets modelling: one current density another parameters. Machine models, including extreme gradient boosting (XGB) artificial neural networks (ANN), developed, optimized predict attributes. Data preprocessing hyperparameter tuning carried out accuracy. Results show that ANN outperforms XGB density, achieving an R2 > 0.98, accurately parameters 0.99, indicating the models developed very promising accurate prediction coated plates PEM cells.

Language: Английский

Citations

9
Machine learning and spatio-temporal analysis of meteorological factors on waterborne diseases in Bangladesh DOI Creative Commons
Arman Hossain Chowdhury, Md. Siddikur Rahman

PLoS neglected tropical diseases, Journal Year: 2025, Volume and Issue: 19(1), P. e0012800 - e0012800

Published: Jan. 16, 2025

Background Bangladesh is facing a formidable challenge in mitigating waterborne diseases risk exacerbated by climate change. However, comprehensive understanding of the spatio-temporal dynamics these at district level remains elusive. Therefore, this study aimed to fill gap investigating pattern and identifying best tree-based ML models for determining meteorological factors associated with Bangladesh. Methods This used district-level reported cases (cholera, amoebiasis, typhoid hepatitis A) obtained from Bureau Statistics (BBS) data (temperature, relative humidity, wind speed, precipitation) sourced NASA period spanning 2017 2020. Exploratory spatial analysis, regression machine learning were utilized analyze data. Results From 2020, 73, 606 cholera, 38, 472 typhoid, 2, 510 A 1, 643 amoebiasis disease cases. Among cholera showed higher incidence rates Chapai-Nawabganj (456.23), Brahmanbaria (417.44), Faridpur (225.07), Nilphamari (188.62) Pirojpur (171.62) districts. The model identified mean temperature (β = 12.16, s.e: 3.91) as significant factor diseases. optimal XGBoost highlighted minimum temperature, humidity precipitation determinants Conclusions findings study, incorporating One Health perspective, provide insights planning early warning, prevention, control strategies combat similar endemic countries. Precautionary measures intensified surveillance need be implemented certain high-risk districts across country.

Language: Английский

Citations

1
Modern computational approaches for rice yield prediction: A systematic review of statistical and machine learning-based methods DOI
Djavan De Clercq, Adam Mahdi

Computers and Electronics in Agriculture, Journal Year: 2025, Volume and Issue: 231, P. 109852 - 109852

Published: Feb. 5, 2025

Language: Английский

Citations

1
Cadmium accumulation in tropical island paddy soils: From environment and health risk assessment to model prediction DOI
Yan Guo, Yi Yang,

Ruxia Li

et al.

Journal of Hazardous Materials, Journal Year: 2023, Volume and Issue: 465, P. 133212 - 133212

Published: Dec. 12, 2023

Language: Английский

Citations

21
Decoding Potato Power: A Global Forecast of Production with Machine Learning and State-of-the-Art Techniques DOI
Shikha Yadav, Abdullah Mohammad Ghazi Al Khatib, Bayan Mohamad Alshaib

et al.

Potato Research, Journal Year: 2024, Volume and Issue: 67(4), P. 1581 - 1602

Published: Feb. 26, 2024

Language: Английский

Citations

7
Predicting anxiety, depression, and insomnia among Bangladeshi university students using tree‐based machine learning models DOI Creative Commons
Arman Hossain Chowdhury, Dana Rad, Md. Siddikur Rahman

et al.

Health Science Reports, Journal Year: 2024, Volume and Issue: 7(4)

Published: April 1, 2024

Abstract Background and Aims Mental health problem is a rising public concern. People of all ages, specially Bangladeshi university students, are more affected by this burden. Thus, the objective study was to use tree‐based machine learning (ML) models identify major risk factors predict anxiety, depression, insomnia in students. Methods A social media‐based cross‐sectional survey employed for data collection. We used Generalized Anxiety Disorder (GAD‐7), Patient Health Questionnaire (PHQ‐9) Insomnia Severity Index (ISI‐7) scale measuring students' depression problems. The supervised decision tree (DT), random forest (RF) robust eXtreme Gradient Boosting (XGBoost) ML algorithms were build prediction their predictive performance evaluated using confusion matrix receiver operating characteristic (ROC) curves. Results Of 1250 students surveyed, 64.7% male 35.3% female. ages ranged from 18 26 years old, with an average age 22.24 (SD = 1.30). Majority (72.6%) rural areas media addicted (56.6%). Almost 83.3% had moderate severe 84.7% 76.5% Students' addiction, age, academic performance, smoking status, monthly family income morningness‐eveningness main insomnia. highest observed XGBoost model Conclusion findings offer valuable insights stakeholders, families policymakers enabling profound comprehension pressing mental disorders. This understanding can guide formulation improved policy strategies, initiatives promotion, development effective counseling services within campus. Additionally, our proposed might play critical role diagnosing predicting problems among similar settings.

Language: Английский

Citations

7
A Machine Learning Approach to Monitor the Physiological and Water Status of an Irrigated Peach Orchard under Semi-Arid Conditions by Using Multispectral Satellite Data DOI Open Access
Pasquale Campi, Anna Francesca Modugno, Gabriele De Carolis

et al.

Water, Journal Year: 2024, Volume and Issue: 16(16), P. 2224 - 2224

Published: Aug. 6, 2024

Climate change is making water management increasingly difficult due to rising temperatures and unpredictable rainfall patterns, impacting crop availability irrigation needs. This study investigated the ability of machine learning satellite remote sensing monitor status physiology. The research focused on predicting different eco-physiological parameters in an irrigated peach orchard under Mediterranean conditions, utilizing multispectral reflectance data algorithms (extreme gradient boosting, random forest, support vector regressor); ground were acquired from 2021 2023 south Italy. forest model outperformed net assimilation (R2 = 0.61), while performed best electron transport rate 0.57), Fv/Fm ratio 0.66) stomatal conductance 0.56). Random also proved be most effective stem potential 0.62). These findings highlighted integrating techniques with high-resolution imagery assist farmers monitoring health optimizing practices, thereby addressing challenges determined by climate change.

Language: Английский

Citations

7