The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
Centre Latitude 
24.928055 
22.692492 
Radius 
1.204827 
0.481931 
Population Number 
1669635 
1901586 
Observed Cases 
49 
52 
Expected Cases 
16.766081 
19.095279 
LLR 
27.824853 
9.668875 
Relative Risk 
3.341264 
3.140871 
P 
0.001 
0.001 
Table 2. b) Spatial scan statistics results (scale: province) 
Hotspot Cluster: 1 
Cluster type 
Circle 
Centre Name 
Shuikou Town 
Centre Code 
45142309 
Centre Longitude 
106.586283 
Centre Latitude 
22.478791 
Radius 
0.156412 
Population Number 
41485 
Observed Cases 
2 
Expected Cases 
0.704587 
LLR 
2.967778 
Relative Risk 
5.676992 
P 
0.001 
^ 22:29 
riiH&KPÜ 
É 
m 
Si ¡¡«Stift*« 
ê m È&SliMîêl'lift*« 
M &ÊiE'tift)«0A 
$~gi 
? m xwm 
s-ii«# 
è m ft#** 
*£&»«**& 
Jfh 
Figure 5. Fieldwork response module of PDA client 
In the experiment of prototype system, the surveillance 
computing program could deal with multi-scale geographical 
and disease data, and provide corresponding computing results. 
Intelligent disease diagnosis program of PDA client could 
accomplish the intelligent diagnosis of several hundred kinds of 
diseases. Meanwhile, on the condition of nice internet and 
mobile network, collaborative server, PC client and PDA client 
could implement the one-to-one, one-to-many, many-to-one, 
and many-to-many collaborative work rapidly and accurately. 
Table 2. c) Spatial scan statistics results (scale: county) 
7. CONCLUSION 
a) b) 
c) 
d) 
Figure 4. Intelligent disease diagnosis module of PDA client: a) 
choose the symptoms; b) choose the districts and the 
exposed time; c) reconfirm the input items; d) 
display the results. 
The structure design and implementation of a collaborative 
epidemical surveillance and response system was discussed in 
this paper. This system combined epidemical surveillance 
system and response system using collaborative working mode, 
and accomplished the surveillance computing program and 
intelligent disease diagnosis program based on scan statistics 
and Bayesian analysis methods. The prototype system had 
completed the architecture design, basic program developing, 
equipment testing and practical experiment, and reached a 
satisfying result. Before applied in practical applications, the 
system has some possible problems to solve or improve, such as: 
a) Due to the full-time and long-period system running 
demand, tests for the architecture mode and the carrying 
capacity of the server are needed. 
b) Further improvement to the architecture mode is 
required for the optimization of the P2P (Person to Person), 
P2C (Person to Computer) and C2C (Computer to Computer) 
collaborations. 
c) The disease models in the surveillance and response 
modules are possibly required to be improved and optimized. 
d) A great deal of real disease data are needed for the 
precision and validity tests to the system and models. 
ACKNOWLEDGEMENTS 
This research is partially supported by the National High-tech 
R&D Program (863 Program) 2007AA12Z240, the Beijing’s 
Natural Science Key Foundation Project No.7061005, and the 
National High-tech R&D Program (863 Program) 
2006AA12Z109.