System Performance

Puli Christian Hospital has in place a high speed internet and intranet communication system that has been in operation for a number of years. Recently, the 10/100/1000 Base-T Ethernet was installed. With this high speed communication capability, image retrieval and display processes can be carried out very smoothly at each viewing terminal. The average response time is 0.2 seconds for retrievalof a CT image and 0.08 seconds for retrieval of an ultrasound or endoscopic image.

There are currently over 40 terminals that can access the medical database at the same time without notable lag.

Since this PACS is an in-house developed software, the cost of the entire system is proportional to the hardware (mainly, the viewing terminals). For each single unit, the hardware is equipped with a PC (Cerelon CPU, 512 MB RAM) and one 19-inch LCD monitor with 1280 x 1024 resolution, with a cost of approximately $900 USD.

The HP ProLiant, which is being used as the image server, is equipped with a 200 Gigabyte hard drive storage capacity. For data backup purposes, the RAID (Redundant Array of Independent Disks) system with four 300 Gigabyte hard drives is used.

As it was mentioned previously, the image quality is of particular importance for the PACS system. For those images that are available in digital format (such as the Computed Tomography), there is no need to do any conversion. However, for those non-digital sources, there is a need to capture those images with a special graphical device. An “ASUS Extreme AX 700 PRO”

acquisition card (also known as Frame Grabber) is used to obtain high quality images. The price for this card is around $250 USD each, and this device is currently installed for the following five procedures: the Cerebrovascular Ultrasound, Cardiac Ultrasound, Abdomen Ultrasound, Kidney Ultrasound, and the Upper and Lower alimentary endoscope. Because these five different departments have different hardware equipments, the organic interface necessary for these machines to capture images is very important. A user-friendly interface was designed for these non-DICOM apparatus to capture these images and store them in the medical database. When images from these sources are acquired, the system can convert them to JPG, BMP, DICOM or any animated formats (AVI or WMV).

Table 1 shows the image quality and the hardware specification used, and Table 2 shows the number of patients and number of images acquired and processed in one month (March 01 to March 31, 2006).

Most commercial PACS systems are sold based on either the number of desired licenses or the “KeyPro”

purchase which locks the view terminals. Hence, increasing the number of view station not only increases the hardware cost but also the software expense, resulting in a heavy financial load to the medical

facilities. For government-supported clinics or metropolitan hospitals, the financial pressure may not be a major concern. However, because regional medical facilities handles not only local residents’ medical problems but also has the responsibility to provide healthcare services for the mountainous area or remote communities, the in-house self-made PACS means a lot since it can be freely duplicated without incurring excessive hospital expenses.

Table 1. Image quality and the hardware specification.

Ultrasound HP 8500 Frame Grabbing

720×480 8 Ultrasound HP 4500 Frame

Grabbing

Table 2. Images acquired during March, 2006 Imaging

Ultrasound 460 2886 193 Mbyte

Endoscope 120 1080 74 Mbyte

Total 801 10970 4087 Mbyte

5.

Conclusion

When PACS was first announced in the 80s, many people, especially the radiologists, expressed high interest and expected that it could help the hospital achieve a filmless environment. Even though there were controversies during the development period, PACS has become an indispensable tool in hospital operations.

However, the main obstacles preventing wide scale implementation of PACS among all the hospitals and clinics are its cost and lack of flexibility. Most PACSs were developed by vendors and software companies, and can generally be afforded only by metropolitan hospitals.

In addition, because they were developed by software companies who were anticipating that they would be used by a full-service hospital, these PACSs may not completely fit the needs of diverse or smaller hospitals.

For regional hospitals with limited budget, it has been even more difficult to promote the system in serving the needs of their patients. According to statistics reports from the Department of Health, Executive Yuan, R.O.C., only around 7% of hospitals in Taiwan utilize the PACS.

To overcome this restriction in Puli Christian Hospital, a customized PACS system has been developed by their in-hospital physicians and technicians from the Information Department. It is well known that in order to develop productive software, a good and thoughtful system analysis needs to be performed prior to building a PACS that will fit the needs of the hospital. In order to achieve this, the on-site physicians were deemed to be the best system analysts. This full-function PACS was initiated three years ago, and has achieved a relatively mature state in providing hospital-wide service. Because this system was developed and implemented by the PCH physicians, its maintenance and upgrade capabilities are much stronger than any commercially-developed PACS software. With regular feedbacks from the users (the clinicians, nurses, and technologist), modifications can be made quickly and easily. For the Puli Christian Hospital, this PACS not only saved considerable expense compared with purchasing commercial software, it has also provided the following contributions and

benefits to PCH:

z Instead of manually handling medical results, this PACS provides systematic access to the patient’s information on a hospital-wide basis. This has sped up the administrative process and medical treatments considerably.

z Medical information, examination results, and treatment histories have been stored electronically, thereby greatly reducing required physical storage space while increasing the capability of data preservation.

z Integration of the original Healthcare Information System (HIS) and the Radiology Information System (RIS) has increased hospital administrative efficiency; management abilities have improved enormously while reducing operation costs.

z This easily manipulated system has helped physicians/clinicians to make more accurate diagnosis, resulting in enhanced medical service.

The ability to show medical results though the PACS has also earned the trust of patients.

z Simplification of the referral process for patients to transfer in or out of the hospital. Sharing the medical resources has also facilitated network treatments.

z The majority of PACS systems in Taiwan are commercially-developed products. In stark contrast, in developing this customized system, considerable valuable experience has been gained through designing the system analysis, software architecture design, debug process, system maintenance and developing updates, … etc. It is strongly believed these experiences can be used to

showcase and promote further hospital software development in Taiwan.

References

[1] U.S. Project of the National Health Information Infrastructure. http://aspe.hhs.gov/sp/NHII/

[2] Homepage of National Health Service of the Great Britian. http://www.nhs.uk/

[3] Homepage of Health of Canada. http://www.hc-

sc.gc.ca/hcs-sss/ehealth-esante/ehr-dse/biblio_can_e.html

[4] Homepage of Department of Health, Executive Yuan, Taiwan, ROC. http://www.doh.gov.tw/cht/index.aspx [5] Technical Report, National Science Council (NSC), Taiwan, ROC. http://www.nsc.gov.tw/pub/yearbook/

yearbook93/main/4/4-2.htm

[6] Fu-Zhong Wang, “The Study of National Health Information Infrastructure in Taiwan.”, Technical Report, Bureau of National Health Insurance, Taiwan, R.O.C., Dec, 2005.

[7] Cheng-Hua Lee, “Health Care and Medical Insurance: The National Health Information Infrastructure.”, Technical Report, Bureau of National Health Insurance, Taiwan, R.O.C., Sep 20, 2003.

[8] Steckel RJ, “The Current Application of PACS to Radiology Practice.”, Radiology 1994; vol. 190(3), pp 50A-52A.

[9] Friedenberg RM, “PACS in the Clinical Setting:

Revisited.”, Radiology 1994; Vol 190(3): 53A-54A.

[10] Davide Caramella, “Is PACS Research and Development Still Necessary?” International Congress Series 1281, 2005, pp. 11-14.

[11] Tain-Chen Wu, San-Kan Lee, Chen-Hsing Peng, Chia-Hsien Wen and Shu-Kun Huang, “An Economical, Personal Computer-based Picture Archiving and Communication System”, RadioGraphics, 1999, Vol.

19(2) , pp 523-530.

[12] I-Chiu Chang, Hsin-Ginn Hwang, David C. Yen, J.W. Lian, “Critical Factors for Adopting PACS in Taiwan: Views of Radiology Department Directors”, Journal of Decision Support System.

[13] Mervyn D. Cohen, MB, ChB, MD, Lori L.

Rumreich, MBA, Kimberley M. Garriot, S. Gregory Jennings, MD “Planning for PACS: A Comprehensive Guide to Non-technical Considerations.”, Journal of the

American College of Radiology, April 2005, Vol. 2 No.

4, pp 327-337.

[14] William J. Schroeder, Kenneth M. Martin, and William E. Lorenson. “The Design and Implementation of an Object-Oriented Toolkit for 3D Graphics and Visualization”. In IEEE Visualization, pp. 93-100, 1996.

[15] vtk homepage, http://www.vtk.org/

[16] itk homepage, http://www.itk.org/