Prediction and Factor Extraction of Drug Function by Analyzing Medical Records in Developing Countries

研究成果: ジャーナルへの寄稿記事

抄録

The World Health Organization has declared Bangladesh one of 58 countries facing acute Human Resources for Health (HRH) crisis. Artificial intelligence in healthcare has been shown to be successful for diagnostics. Using machine learning to predict pharmaceutical prescriptions may solve HRH crises. In this study, we investigate a predictive model by analyzing prescription data of 4,543 subjects in Bangladesh. We predict the function of prescribed drugs, comparing three machine-learning approaches. The approaches compare whether a subject shall be prescribed medicine from the 21 most frequently prescribed drug functions. Receiver Operating Characteristics (ROC) were selected as a way to evaluate and assess prediction models. The results show the drug function with the best prediction performance was oral hypoglycemic drugs, which has an average AUC of 0.962. To understand how the variables affect prediction, we conducted factor analysis based on tree-based algorithms and natural language processing techniques.

元の言語英語
ページ(範囲)403-407
ページ数5
ジャーナルStudies in Health Technology and Informatics
245
出版物ステータス出版済み - 1 2018

Fingerprint

Developing countries
Developing Countries
Medical Records
Health
Bangladesh
Learning systems
Health Resources
Personnel
Pharmaceutical Preparations
Prescriptions
Trees (mathematics)
Natural Language Processing
Factor analysis
Drug products
Artificial intelligence
Artificial Intelligence
Hypoglycemic Agents
ROC Curve
Statistical Factor Analysis
Area Under Curve

これを引用

@article{f4b22dfc3cb349f1900a6f79e8d3b2f7,
title = "Prediction and Factor Extraction of Drug Function by Analyzing Medical Records in Developing Countries",
abstract = "The World Health Organization has declared Bangladesh one of 58 countries facing acute Human Resources for Health (HRH) crisis. Artificial intelligence in healthcare has been shown to be successful for diagnostics. Using machine learning to predict pharmaceutical prescriptions may solve HRH crises. In this study, we investigate a predictive model by analyzing prescription data of 4,543 subjects in Bangladesh. We predict the function of prescribed drugs, comparing three machine-learning approaches. The approaches compare whether a subject shall be prescribed medicine from the 21 most frequently prescribed drug functions. Receiver Operating Characteristics (ROC) were selected as a way to evaluate and assess prediction models. The results show the drug function with the best prediction performance was oral hypoglycemic drugs, which has an average AUC of 0.962. To understand how the variables affect prediction, we conducted factor analysis based on tree-based algorithms and natural language processing techniques.",
author = "Min Hu and Yasunobu Nohara and Masafumi Nakamura and Naoki Nakashima",
year = "2018",
month = "1",
language = "English",
volume = "245",
pages = "403--407",
journal = "Studies in Health Technology and Informatics",
issn = "0926-9630",
publisher = "IOS Press",

}

TY - JOUR

T1 - Prediction and Factor Extraction of Drug Function by Analyzing Medical Records in Developing Countries

AU - Hu, Min

AU - Nohara, Yasunobu

AU - Nakamura, Masafumi

AU - Nakashima, Naoki

PY - 2018/1

Y1 - 2018/1

N2 - The World Health Organization has declared Bangladesh one of 58 countries facing acute Human Resources for Health (HRH) crisis. Artificial intelligence in healthcare has been shown to be successful for diagnostics. Using machine learning to predict pharmaceutical prescriptions may solve HRH crises. In this study, we investigate a predictive model by analyzing prescription data of 4,543 subjects in Bangladesh. We predict the function of prescribed drugs, comparing three machine-learning approaches. The approaches compare whether a subject shall be prescribed medicine from the 21 most frequently prescribed drug functions. Receiver Operating Characteristics (ROC) were selected as a way to evaluate and assess prediction models. The results show the drug function with the best prediction performance was oral hypoglycemic drugs, which has an average AUC of 0.962. To understand how the variables affect prediction, we conducted factor analysis based on tree-based algorithms and natural language processing techniques.

AB - The World Health Organization has declared Bangladesh one of 58 countries facing acute Human Resources for Health (HRH) crisis. Artificial intelligence in healthcare has been shown to be successful for diagnostics. Using machine learning to predict pharmaceutical prescriptions may solve HRH crises. In this study, we investigate a predictive model by analyzing prescription data of 4,543 subjects in Bangladesh. We predict the function of prescribed drugs, comparing three machine-learning approaches. The approaches compare whether a subject shall be prescribed medicine from the 21 most frequently prescribed drug functions. Receiver Operating Characteristics (ROC) were selected as a way to evaluate and assess prediction models. The results show the drug function with the best prediction performance was oral hypoglycemic drugs, which has an average AUC of 0.962. To understand how the variables affect prediction, we conducted factor analysis based on tree-based algorithms and natural language processing techniques.

M3 - Article

VL - 245

SP - 403

EP - 407

JO - Studies in Health Technology and Informatics

JF - Studies in Health Technology and Informatics

SN - 0926-9630

ER -