Development of an automated patient recognition system in the PACS environment
| Project/Area Number |
14570894
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Radiation science
|
|---|
| Research Institution | Kyoto College of Medical Technology |
|---|
Principal Investigator |
MORISHITA Junji Kyoto College of Medical Technology, Department of Radiological technology, Associate Professor, 診療放射線技術学科, 助教授 (40271473)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
WATANABE Hideyuki University of Occupational and Environmental Health, Department of Radiology, Associate Professor, 放射線科, 助教授 (10210931)
KATSURAGAWA Shigehiko Kumamoto University, School of Health Sciences, Professor, 医学部・保健学科, 教授 (60021630)
TOYAMA Keiko Kyoto College of Medical Technology, Department of Radiological technology, Instructor, 診療放射線技術学科, 講師 (20342157)
|
|---|
| Project Period (FY) |
2002 – 2003
|
| Project Status |
Completed (Fiscal Year 2003)
|
| Budget Amount *help |
¥2,500,000 (Direct Cost: ¥2,500,000)
Fiscal Year 2003: ¥1,200,000 (Direct Cost: ¥1,200,000)
Fiscal Year 2002: ¥1,300,000 (Direct Cost: ¥1,300,000)
|
|---|
| Keywords | chest radiography / PACS / image recognition / template matching |
|---|
| Research Abstract |
1. We have developed two different methods for automated patient recognition of chest radiographs by using "biological fingerprints" and edge-enhanced images. The database used in this study was the same as used in the previous study, which consisted of 2000 PA chest radiographs that included 1000 current and 1000 previous images from 1000 patients.
2. Developed system of an automated patient recognition and identification was installed for a prospective study in the department of radiology, University of Occupational and Environmental Health.
3. The performance of the automated patient recognition and identification system in the prospective study indicated a high performance of 86.4% correct warning rate for different patients. On the other hand, incorrect warning rate for the same patients was 1.5%. These results were promising to prevent filing errors in the PACS environment by use of the developed method in this study.
3. Motivation of this study was based on unavoidable human errors in the PACS environment. If patient information associated with an image is entered incorrectly or accidental image acquisitions for a wrong patient occur, the images will be assigned to a different patient and will not be stored in the proper folder. Misfiled cases in the PACS environment may create serious medical accidents in hospitals. It is, therefore, desirable to discover a misfiled case as soon as an acquired image is transferred to the PACS server, and to develop an automated warning system for radiology personnel to prevent subsequent medical accidents. We also surveyed and analyzed the misfiled cases for various modalities at another university. We found that the misfiled cases were actually happened in all modalities and main reason of the misfiling cases was due to human errors. These results were presented at RSNA2003 and awarded as certificate of merit.
|
Report
(3 results)
Research Products
(11 results)
