醫學影像資訊系統在眼科醫學上的應用

A PACS Applied in Ophthalmology

羅文斌(Wen-Pin Lo)

指導教授 ： 婁世亮

中原大學/工學院/醫學工程研究所/碩士(2006年)

https://doi.org/10.6840/CYCU.2006.00482

摘要

中文摘要
影像歸檔/傳輸系統（Picture Archiving and Communication System, PACS）在世界各國已相當普遍地運用在日常醫療運作上。然而，環顧現今已建構PACS系統的醫院，鮮少包含眼科部門。根據我們文獻及網路搜尋的結果，至今唯一提供眼科（Ophthalmology） PACS的廠商僅僅只有美國眼科影像系統公司（Ophthalmic Imaging Systems, OIS），該公司的眼科 PACS還是與艾克發（AGFA）公司共同研發而成的。眼科 PACS的少有，其原因大致有二：一則是因為醫院使用數位化眼科造影儀器並不廣泛，再則是眼科的數位醫學影像傳輸（Digital Imaging and Communications in Medicine, DICOM）標準協定規範是最近幾年內才訂妥。因為眼科 PACS並不多見，它的功能效率如何是值得探討的。
本研究課題的目標是研發一眼科PACS，並探討其運作效率。系統的製作是以DICOM為基礎，以之設計製做檔案格式和網路傳輸。系統硬體包含資訊/影像擷取子系統(Information/Image Acquisition Subsystem)、資料管理子系統 (Data Management Subsystem)和影像顯示子系統(Information/Image Display Subsystem)；藉區域網路將上述子系統鏈結成具功能性眼科的PACS。在軟體上，發展有七大模組：(1)檔案格式化、(2)網路傳輸、(3)資料庫存取、(4)歸檔/儲存、(5)搜尋/調檔、(6)影像顯示和(7)工作佇列模組(Queuing Module)。
眼科影像照相術包含有裂隙燈照相、眼底造影術（含眼底攝影、血管螢光攝影）、超音波照相等。其中以裂隙燈、眼底攝影、血管螢光攝影之影像最為普遍。裂隙燈影像平均每張約11 Mbytes (2,272 x 1,704 x 3 bytes)，經影像壓縮成504 Kbytes；眼底攝影之影像約331 Kbytes (388 x 291 x 3 bytes)；血管螢光攝影之影像約1 Mbytes (1,024 x 1,024 x 1 bytes)。系統傳輸效率的測試是分別以裂隙燈影像10張與眼底造影術(含20張眼底攝影和36張血管螢光攝影)為基準。量測的時間包括(1)影像經檔案格式化、(2)兩次傳輸服務、(3)兩次DICOM檔案資訊解析存入資料庫和(4)影像顯示。其過程於有線網路環境中，完成裂隙燈需要花費57.07±12.64秒；眼底造影需要花費118.08±27.01秒。在無線網路環境下，裂隙燈需要花費58.65±13.12秒；眼底造影則需要花費114.34±24.78秒。以同樣造影類別之影像數量進行病例影像調閱測試，由病患資料查詢、Study調閱、一次傳輸服務、一次資料解析至影像顯示。在有線網路環境，裂隙燈影像花費22.50±7.89秒、眼底造影（包含眼底攝影與血管螢光攝影）則需要57.15±8.15秒；在無線網路環境，裂隙燈影像花費22.43±7.92秒、眼底造影則需要花費57.11±8.01秒。
根據本課題的研究結果，我們認為眼科PACS模組的運作效能可以與眼科部門的造像及相關人工作業比擬。期望他日眼科PACS能應用於臨床之運作，以確切掌握其實際效能。

關鍵詞：影像歸檔/傳輸系統、眼科影像、醫學數位影像與通訊標準

關鍵字

影像歸檔/傳輸系統 ； 醫學數位影像與通訊標準 ； 眼科影像

並列摘要

Abstract
Picture archiving and communication system (PACS) has been a daily practice facility in hospitals worldwide. However very few ophthalmology PACS modules are available in the global. According to our literature and internet survey, the only ophthalmology PACS available is developed by the join effort of Ophthalmic Imaging Systems and AGFA. The main causes are two: (1) lacking digital imaging systems equipped in ophthalmology and (2) newly available of the DICOM protocols of digital ophthalmic images. The PACS operation performance in ophthalmology departments is thus not known.
The objectives of this thesis are (1) to implement an ophthalmology PACS module and (2) to study its operational efficiency by simulations. The system implementation was based on the DICOM standard including those newly addressed protocols specifically for ophthalmic images. This is to assure that the image file format and the network communication are fully DICOM compatible. The system includes an acquisition subsystem, a management subsystem, and a display subsystem which are linked by wire or wireless digital networks. Seven software modules were developed in this thesis. These include (1) file formation, (2) network transmission, (3) database transactions, (4) file archival, (5) query/retrieval, (6) image display, and (7) job queue.
Ophthalmic images may be obtained from several systems. Examples are slit lamps, fluorescein angiography, fundus cameras, sonography, etc. Particularly slit lamps, fluorescein angiography, and fundus cameras are commonly used. For these reasons, these types of images were chosen in the studies of the system operation performance. In the studies, ten slit-lamp images, twenty fundus images, and thirty six fluorescein angiograms were used to assess the performance. It must be noted that on average a slit-lamp image contains 12 Mbytes (2,288 x 1,732 x 3 bytes) which is generally compressed to 504 Kbytes. Fundus images and fluorescein angiograms are 331 Kbytes (388 x 291 x 3 bytes) and 1 Mbytes (1,024 x 1,024 x 1 bytes) per image on average, respectively.
The system operation performance was evaluated in two ways: (1) image delivery, and (2) image retrieval. The image delivery studies measured the elapsed time of all the developed software modules operated in the three subsystems and the network. These software modules include image formatting, image transmitting through the network, information extraction, database transactions, and image display. The image retrieval studies aimed on Study Query/Retrieve which summed the time spent in the database query transactions, image transmission over the network, information extraction, and image display. The study results show when the system linked with wire (wireless) networks on average it takes 57.07 (58.65) and 118.08 (114.34) seconds to deliver 10 slit-lamp images and 20 fundus images plus 36 fluorescein angiograms from an acquisition system to an image display system, respectively. For the same number of images, the PACS system is able to retrieve the images in about 22.50 (22.43) and 57.15 (57.11) seconds for the slit-lamp images and fundus/fluorescein angiograms, respectively.
Based upon the study results, we conclude that the operational efficiency of the PACS module developed in this thesis is comparable to the current practice in the ophthalmology department. However, to fully understand the operational performance of the PACS module, it is essential to integrate the system into the ophthalmologic practices.

Key words: PACS, Ophthalmic image, DICOM

並列關鍵字

Ophthalmic image ； PACS ； DICOM

參考文獻

[13] 李三剛、黃樹棍、溫嘉憲、楊晴雯，“醫院電腦化影像儲取與傳輸系統台中榮民總醫院的經驗談”，資訊與電腦，pp.39-46, 1994.

[1] J. K. Portello, “Chloroquine retinopathy and secondary trauma”, Clinical Eye and Vision Care, pp.161-165, 2000.

[2] D. N. Tomasini, “Varix of the vortex ampulla: a dynamic phenomenon”, Clinical Eye and Vision Care, pp.151-154, 2000.

[7] A. Tuulonen, A. Ohinmaa, H. I. Alanko, P. Hyytinen, A. Juutinen, and E. Toppinen, “The application of teleophthalmology in examining patients with glaucoma: A pilot study”, Journal of Glaucoma, pp.367-373, 1999.

[8] K. Yogesan, I. J. Constable, W. Morgan, and D. Y. Soebadi, “International transmission of tele-ophthalmology images”, Journal of Telemedicine and Telecare, pp. 41-44, 2000.