Information Technology Reference
In-Depth Information
which is not suitable for our application (ECG data are in byte format), we
override InputFormat and define a format whose keys are represented by
TextWritable and the values are the file contents, represented by BytesWritable.
Figure 8.10 shows a case of process results or original data and output data
on HDFS.
8.3.1.3 Summary Report of ECG Status
After analysis of ECG data, the graphic-processing clusters can generate the
graphic data, and the visualization of results would be provided to the end
user. Many can be stored in the cluster for later requests, and all temporary
pictures can be drawn at the client so that the computational load on the server
decreases. All in all, the HCloud system can store large heterogeneous ECG
data and compute the corresponding physiological indices within 5  seconds.
If  one of the physiological indices becomes abnormal, it generates disease
warnings on time and sends messages to physicians. At the same time, it mines
the deep regularity or characters from the long-term historical ECG data, which
can help physicians better monitor an individual's health state. The system also
provides detailed ECG physiological indices for medical experts, including
normal-to-normal (NN) intervals, standard deviation of NN intervals (SDNN),
standard deviation of the averages of NN intervals in all 5-minute segments of
the entire recording (SDANN), heart rate variability (HRV) triangular indices,
the triangular interpolation of the NN interval histogram (TINN), and so on.
Figure 8.11 is a simple summary text report of an ECG in the health care
cloud system and shows the measurement duration, in how many seconds
the abnormal ECG also can be obtained, as Figure 8.12a shows. On the other
hand, HCloud generated the Poincaré image of one's heart movement while
measuring ECG signals each time to represent the heart's chaos characteris-
tics, as shown in Figure 8.12b.
Through spectrum analysis, we can draw a conclusion that everyone's
heart movement has a chaos status. Chaos features represent the health
status of the heart. They can illustrate the capability of the heart to adapt to
different situations or body conditions at different times. The image shape is
always heart-like if one is in good health. The more irregular the heart move-
ment is, the more dangerous life is.
8.3.2 Other Physiological Diagnostic Data
Other important physiological information should be manually set into
the system to build a traceable case history for health status (e.g., medical
inspection data as an output of a blood glucose meter or urine analyzer from
authorities, e.g., a qualified administration, a professional institution, a hos-
pital, etc.). A follow-up survey of physiological data with diagnostic value
is also presented. For instance, the general body information and health
knowledge database are established and further instructions are given to
customers. These services are optional to the end users.
Search WWH ::




Custom Search