Page 117 - Big Data Analytics for Intelligent Healthcare Management
P. 117

110     CHAPTER 5 CHRONIC TTH ANALYSIS BY EMG AND GSR BIOFEEDBACK




                                                Period                            Techniques, period
                                                                                    EMGa, BaseLine
                     15.11
                   Subjects: 27                                                     EMGa, 1 month
                14                                                                  EMGa, 3 months
                      12.31
                    Subjects: 26                                                    EMGa, 6 months
                12                                                                  EMGa, 12 months
                                  11.59                                             GSRa, BaseLine
                                Subjects: 27                                        GSRa, 1 month
                10                                                                  GSRa, 3 months
               Avg. duration  8  Subjects: 26   8.08                                GSRa, 6 months
                                 8.58
                                                                                    GSRa, 12 months
                                                                 5.32
                 6                            Subjects: 25     Subjects: 25
                                                5.63                        3.61
                                             Subjects: 24    5.00         Subjects: 23
                 4                                         Subjects: 23
                                                                            3.29
                                                                          Subjects: 21
                 2
                 0
                      BaseLine     1 month      3 months    6 months    12 months
             FIG. 5.9
             Variation of average duration with time.



             rate in the last period (6–12months) observed. Overall, the improvement observed in 1year was
             12.18h/week, which is much better than the GSRa.
                Hence we can conclude that the EMGa technique provided better improvement in the decrease
             of TTH duration per week. EMGa was 123.65 times more efficient than GSRa. In both the tech-
             niques/therapies, there was a slight decrease in the rate of improvement of TTH duration observed from
             6 to 12months that might be the result of longer intimation gap of patient with the feedback process
             (almost 6-month gap).
                Also, on analysis, one can add one more numerical value to this graph, showing the rate of improve-
             ment, such as RoI, is 12unit/month. This can be easily done by calculating the slope of the line for
             different periods, which will provide the same base of comparison between EMG and GSR.
                The performance of the EMG pattern recognition may vary depending for each subject, experimen-
             tal protocol, acquisition setup, and differences in preprocessing. For instance,Dey, Bhatt, and Ashour
             [29] discovered that there was no consequence on the minimum value of the threshold parameter of two
             of the most commonly used EMG features: zero crossings (ZC) and slope sign changes (SSC).




             5.8.3 THE TREND OF AVERAGE OF INTENSITY
             GSRa: The rate of average intensity reduction over a period varied alternatively in alternate periods. In
             the first duration of 1month, the rate of decrease in intensity of TTH was steep but became nearly stag-
             nant in the later period (1–3months) (Fig. 5.10).
   112   113   114   115   116   117   118   119   120   121   122