Towards an Explainable AI-based Tool to Predict the Presence of Obstructive Coronary Artery Disease DOI Open Access
Ilias Kyparissidis Kokkinidis, Emmanouil S. Rigas, Evangelos Logaras

et al.

Published: Nov. 25, 2022

Obstructive coronary artery disease (CAD) is characterized as significant upon detection of stenosis diameter. In this paper, we adapt Artificial Intelligence (AI)-based predictive models to accurately estimate the pretest likelihood obstructive CAD on computed tomography angiography (CCTA) in patients with suspected CAD. doing so, use patients' objective results and variables extracted from screening procedure combination demographics, medical history, social other data. We a dataset consisting 77 apply number alternative Machine Learning (ML) algorithms predict severity . The ensemble voting model showed best across all performance metrics an area under curve (AUC) approximately 0.88. also attempt provide clinicians explanation prediction make it more trustworthy.

Language: Английский

Predict Diabetes Using Voting Classifier and Hyper Tuning Technique DOI Creative Commons

Chra Ali Kamal,

Manal Ali Atiyah

Kurdistan Journal of Applied Research, Journal Year: 2023, Volume and Issue: unknown, P. 115 - 130

Published: Jan. 15, 2023

Today, diabetes is one of the most common chronic diseases in world due to people’s sedentary lifestyle which led many health issues like heart attack, kidney frailer and blindness. Additionally, people are unrealizable about early-stage symptoms prevent it. The above reasons were encouraging develop a prediction system using machine learning techniques. Pima Indian Diabetes Dataset (PIDD) was utilized for this framework as it appropriate dataset .CSV format. While there not any duplicate or null values, however, some zero values replaced, four outlier records removed data standardization performed dataset. In addition, project methodology divided into two phases model selection. first phase, different hyper parameter techniques (Randomized Search TPOT(autoML)) used increase accuracy level each algorithm. Then six algorithms (Logistic Regression, Decision Tree, Random Forest, K-nearest neighbor, Support Vector Machine Naïve Bayes) applied. second best (with estimated parameters them) chosen an input voting classifier, because applies find algorithm between group multiple options. result satisfying, Forest achieved 98.69% stage, while its 81.04% previous predict via simple graphic user interface.

Language: Английский

Citations

2
Toward personalized care for insomnia in the US Army: development of a machine-learning model to predict response to pharmacotherapy DOI
Frances H. Gabbay, Gary H. Wynn,

Matthew W. Georg

et al.

Journal of Clinical Sleep Medicine, Journal Year: 2023, Volume and Issue: 19(8), P. 1399 - 1410

Published: April 20, 2023

Although many military personnel with insomnia are treated prescription medication, little reliable guidance exists to identify patients most likely respond. As a first step toward personalized care for insomnia, we present results of machine-learning model predict response medication.

Language: Английский

Citations

2
Data collaboration analysis in predicting diabetes from a small amount of health checkup data DOI Creative Commons

Go Uchitachimoto,

Noriyoshi Sukegawa,

Masayuki Kojima

et al.

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: July 21, 2023

Recent studies showed that machine learning models such as gradient-boosting decision tree (GBDT) can predict diabetes with high accuracy from big data. In this study, we asked whether highly accurate prediction of is possible even small data by expanding the amount through collaboration (DC) analysis, a modern framework for integrating and analyzing accumulated at multiple institutions while ensuring confidentiality. To end, focused on two institutions: health checkup 1502 citizens in Tsukuba City history 1399 patients collected University Hospital. When using only data, ROC-AUC Recall logistic regression (LR) were 0.858 ± 0.014 0.970 0.019, respectively, those GBDT 0.856 0.983 0.016, respectively. also DC these values LR improved to 0.875 0.013 0.993 0.009, deteriorated because low compatibility method used confidential sharing (although analysis brought improvements). Even situation where 324 are available, 0.767 0.025 0.867 0.04, thanks indicating an 11% 12% improvement. Thus, concluded answer above question was "Yes" but "No" set tested study.

Language: Английский

Citations

2
Analysis of blood glucose monitoring – a review on recent advancements and future prospects DOI

Gayathri Priyadarshini R,

Sathiya Narayanan

Multimedia Tools and Applications, Journal Year: 2023, Volume and Issue: 83(20), P. 58375 - 58419

Published: Dec. 21, 2023

Language: Английский

Citations

2
Towards an Explainable AI-based Tool to Predict the Presence of Obstructive Coronary Artery Disease DOI Open Access
Ilias Kyparissidis Kokkinidis, Emmanouil S. Rigas, Evangelos Logaras

et al.

Published: Nov. 25, 2022

Obstructive coronary artery disease (CAD) is characterized as significant upon detection of stenosis diameter. In this paper, we adapt Artificial Intelligence (AI)-based predictive models to accurately estimate the pretest likelihood obstructive CAD on computed tomography angiography (CCTA) in patients with suspected CAD. doing so, use patients' objective results and variables extracted from screening procedure combination demographics, medical history, social other data. We a dataset consisting 77 apply number alternative Machine Learning (ML) algorithms predict severity . The ensemble voting model showed best across all performance metrics an area under curve (AUC) approximately 0.88. also attempt provide clinicians explanation prediction make it more trustworthy.

Language: Английский

Citations

3