Construction of an Automatic Audit System for electronic medical record by the machine learning for Informed Consent
| Project/Area Number |
26460869
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Medical and hospital managemen
|
|---|
| Research Institution | University of Hyogo |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
桑田 成規 国立研究開発法人国立循環器病研究センター, その他, 総長付 (40379631)
岡本 和也 京都大学, 医学研究科, 講師 (60565018)
山田 ひとみ 国立研究開発法人国立循環器病研究センター, 病院, 室長 (70634913)
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2016: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
Fiscal Year 2015: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2014: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
|
|---|
| Keywords | 病院情報システム / インフォームドコンセント / 機械学習 / カルテ監査 / 質判定 / 自動監査システム / インフォームド・コンセント / 電子カルテ / 自動監査 / 自然言語処理 / 診療録監査 |
|---|
| Outline of Final Research Achievements |
Informed consent (IC) is very important because IC record is only one of evidences of consent with patient or patient family and health care provider. Therefore, we developed automatic audit system worked on hospital information system (HIS) that is able to evaluate using machine learning automatically. Concretely, first, we extracted 298 electronic IC charts from HIS. And a health information manager evaluated these IC charts in five levels from the point of experts. Next, we used support vector machines (SVMs) were supervised learning models and be able to marked as belonging to one of two categories. On this time, this system determined whether IC chart is lower level 2 or not.
As a result, we evaluated this system using leave-one-out validation (LOOCV). Consequently, this system could evaluate them 89.4% (261/292) correctly. And false negative rate was 29% (16/56) and false positive rate is 6.4% (15/236) on determine lower level 2 IC charts.
|
Report
(4 results)
Research Products
(11 results)
