Research Of Dynamic Blood Information Management System

Nowadays, with the development of the technology of medicine and theincreasing using of the blood for clinic, the requirement to the blood quality hasbeen enhanced step by step. The traditional management pattern can not meet theneeds of blood station. With the increasing number of the blood donor, computershave been more and more used in most of the blood stations. Since the presentblood information management system has some disadvantages in several problemssuch as the management of the blood information and the identification of blooddonor, in this paper, the specific technology and the whole progress of the dynamicblood information management system has been studied and the capability andperformance of the system has been presented. The paper focuses on the followingcontents.Dynamic blood information management system improves and strengthens themanagement of the blood station by using some techniques including RFID, dataencryption, fingerprint identification, image compression and wireless network.The application of this system can not only make the management of blood stationmore regularity and modernization but also ensure the security of bloodinformation. In this system, the donation card and barcode on the surface of theblood bag has been respectively replaced by smart card and tag which contains allthe blood information. Furthermore, the data has been encrypted by Rijndaelencryption 1algorithm to avoid accessing that was not authorized. We put the photoof blood donor into smart card through the image compression technique. In orderto avoid all kinds of illegal blood donation, we propose an improved fingerprintmatching algorithm to confirm the blood donor's identification. In addition, GPRSis applied in this system so that real-time data can be transmitted between thebloodmobile and blood center through wireless internet. RFID technique has been applied in our system to ensure the security andcredibility of the blood information that will be stored. On one hand, we give everyblood donor an electronic donation card instead of paper card to carry the requiredinformation. On the other hand, smart label was intended to replace barcodes thatwere widely used in the traditional blood information system. These two sectionscan make great efficiency and secure the data. Considering the shortage of RFIDtechnique, we store the blood data into smart card which are encrypted by Rijndaelencryption algorithm instead of put them into it directly. We did a lot of research onRijndael encryption algorithm;the process and the way of encryption are detaileddiscussed in our paper. In the process of encryption and decryption, we propose animprovement on Rijndael algorithm and a more complex method of key creationhas been used to encrypt and decrypt the blood information. Because of the strongability avoiding assault of Rijndael encryption 1algorithm, the information can notbe seen even if the key of smart card or tag lost. In addition, we provided eachsmart card and tag with a unique key to reduce the hidden danger of security.Besides the security of the blood and donor information, the identification ofthe blood donor is also a very important aspect in our system. Although thetraditional identification way is out of date, we use fingerprints to identify donor inorder to avoid the situation that someone who wants to take place of the other. Inthis paper, we have studied principle and characteristic of fingerprint imagerecognition method based on structure feature matching and analyzed the way ofblood donor identification by fingerprint in blood information management system.Aiming at the problem that general fingerprint image recognition algorithm gaveno support to the rotation and the displacement of the fingerprint image, themathematic model in proposed in which the distance between two points in a planeis independent of the rotation or displacement of this plane. The matchingalgorithm based on structure feature matching refers to the distance betweenspecial points, the types, and the numbers lines spanned by the special points. Then,according to the analysis of the structural feature minutia of the special points, weimproved the algorithm and proposed the second matching to some feature pointsthat were not matched. After then, we presented the flow chart of searchingfingerprint before phlebotomize, fast check and phlebotomize. We can find theFRR has been reduced from the experimentation. It is proved that the matchingalgorithm based on structure feature matching can meet the needs of therequirement of blood donor identification in the blood information managementsystem.In the rest part of the paper, we described the function and structure ofdynamic blood information management system, and then the design andimplementation of the system are given. The whole system includes two parts;oneis phlebotomy management in bloodmobile, the other is transaction in blood station.The former part accomplished the registration to the donor before and afterphlebotomize. It also provided the function of fingerprint searching in the localdatabase. The later part includes blood verifying, blood component separation,blood bag packing etc. The information about blood donor and blood productionhas been stored in the smart card and tag through card reader and electronicdonation card will be created by a card printer. Besides this, GPRS networktechnique has been applied in our blood information management system. Itincludes two parts. The first part includes the management of the verifying module,separation module, unfinished product module and finished produce modulethrough local area network. The management of registration module by GPRSwireless network and internet belongs to the second part. The system runningenvironment is made up of Windows 2000, Sybase and GPRS. At last, we testedthe system with many experiments and result shows that the system is credible andstable.As the development of the blood donation work, the management of bloodmust be more and more modernization, scientific and standardization. People haverealized that the quality of the blood should be ensured and real-time data shouldbe transmitted between the bloodmobile and blood center so that the efficiency ofoperation management can be increased. However, there are still lots of problemsto solve in our dynamic blood information management system including the speedof fingerprint image recognition method based on structure feature matching andGPRS wireless network. As a conclusion of the former work, the system alsoprovides us a platform for further study in the future.