Migration of Digital Medical Image Data Stored through Mini-PACS to Full-PACS.
- Author:
Haijo JUNG
1
;
Hee Joung KIM
;
Won Suk KANG
;
Sang Ho LEE
;
Sae Rome KIM
;
Young Mok SON
;
Chang Lyong JI
;
Jung Han KIM
;
Sun Kook YOO
;
Ki Hwang KIM
;
Hyung Sik YOO
Author Information
1. Department of Diagnostic Radiology, Yonsei University College of Medicine. hjkim@yumc.yonsei.ac.kr
- Publication Type:Original Article
- Keywords:
PACS;
Migration;
Medical Image Data;
DLT;
CD
- MeSH:
Academic Medical Centers
- From:Journal of Korean Society of Medical Informatics
2003;9(3):275-284
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
The purpose of this study was to evaluate the migration of medical image data stored thorough mini-PACS to full-PACS at Yonsei University Medical Center (YUMC). The image data to be migrated were 2.7 TB stored through approximately 4,500 CD archives at Yongdong Severance hospital and 4.7 TB (2:1 compression) stored through 196 digital linear tape (DLT) archives at Severance hospital. Prior to carrying out the migration, principles, methods and expected practical affairs for the migration were discussed and planned to optimize the migration work by considering the cost and the effectiveness of migration work. Migration gateway workstations were set up and a migration software tool was developed. Real migration works were performed based on the results of several migration simulations. Severance hospital decided to migrate all stored image data. The CD image data of 2.7 TB were estimated total 2,250 hr (about 94 days) migration time, but the practical migration work was completed within 3 months by using maximum 5 workstations. The DLT data of 4.7 TB were estimated total 100 days migration time by applying 16 hr working time per day with single workstation, however, the practical migration work was taken 5 months. Meanwhile, 20% of the DLT image data were not able to migrate because the DLT were partially damaged due to frequent access. In conclusion, a migration plan should be carefully prepared by considering the individual hospital environments because the server system, archival media, the network, and the policy of data management may be unique.
