+ Reply to Thread
Results 1 to 3 of 3

Thread: SP1: Pullman et al,. (1998) spiral analysis using MovAlyzeR or MovAlyzeRx

  1. Join Date
    Nov 2003
    Posts
    3,353

    SP1: Pullman et al,. (1998) spiral analysis using MovAlyzeR or MovAlyzeRx

    Pullman SL.
    Spiral analysis: a new technique for measuring tremor with a digitizing tablet.
    Mov Disord. 1998;13 Suppl 3:85-9.
    http://www.ncbi.nlm.nih.gov/pubmed/16647772

    NOTE: You will need the Movalyzer6.1 Processmod.dll patch elsewhere in this forum: http://www.neuroscript.net/forum/sho...ght=processmod

    (1) Please download SP1.MEF and save the MovAlyzeR Export File of Experiment SP1 (SP1.MEF). (2010-12-01)
    (Do not change the name)
    In MovAlyzeR: Click File
    In MovAlyzeRx click in the top left circle
    then: Import >Select SP1.MEF

    (2) Type in the box in the Patients window: Test.
    You will see the results recorded before.

    (3) To record your own patient, type in the Patients window a new patient code.
    If not already existing click the blue + (New Patient) and complete public and private data.
    Then click in the Tests window on the green arrow (Run test).

    (4) Immediately after a test all Pullman et al. (1998) parameters, including the Degree Of Severity are calculated together with basic movement features.

    (5) In the Tests window click on View Analysis Chart. Customize the chart to your liking.
    Click Actions/Settings and Xmean data
    Click Excel to get output data in MS Excel format

    You can start viewing data and testing new data now.

    If you test a patient a next time, by default this patient will be grouped with the next Visit number.
    Last edited by Ameya; 2010-12-02 at 12:42.
    Company http://www.neuroscript.net and work http://www.neuroscript.net/hans_leo_teulings.php on handwriting and drawing movement recording and processing.

  2. Join Date
    Nov 2003
    Posts
    3,353
    Pullman.exe is an external function to MovAlyzeR that calculates several Pullman features of spirals drawn on a pen tablet (or on paper using image processing).

    Examples are simulated spirals in 3 Groups or Tests (1=bad, 2=better, 3=best, resp). Line color represents speed (blue is high speed). Red circles represent the segmentation points between up and down strokes:
    Name:  g0101.png
Views: 813
Size:  20.0 KB Name:  g0201.png
Views: 742
Size:  17.1 KB Name:  g0301.png
Views: 779
Size:  15.0 KB

    Smooth1 = i1 = 1stOrderSmoothness ** -- FIRST ORDER SMOOTHNESS First order smoothness characterizes imperfections in spiral drawing. Although the smoothness value for a given spiral will capture the overall deviation from ideal spiral shape, it does not differentiate between irregularities that would be clinically perceived as tremor, and other irregularities such as lopsidedness or overlapping spiral loops.

    The Smooth1 feature per spiral (and 2 more spirals) are shown for Tests 1, 2, 3:
    Name:  smooth1.png
Views: 716
Size:  6.4 KB

    Smooth2 = i2 = 2ndOrderSmoothness ** -- SECOND ORDER SMOOTHNESS Second order smoothness is the rate of change (first derivative) of first order smoothness. In other words, it is the rate at which the rate of irregularity changes. It is similar to first order smoothness in that it quantifies variation from ideal spiral shape. Second order smoothness can be useful when considering whether a spiral is “regularly irregular” or “irregularly irregular”. For example, a patient with essential tremor who has a pronounced 5 Hz drawing tremor and a patient with a dystonic disorder resulting in unpredictable movements could both have similar first order smoothness values, but the dystonia patient should have a higher second order smoothness value, reflecting lack of consistency in his/her deviations from the ideal spiral.

    ZeroCross1 = i4 = 0CrossingRate -- FIRST ORDER ZERO CROSSING RATE Zero crossing is another measure of irregularity. The first order zero crossing rate measures how frequently !r !" values cross their own rms value on the graph of !r !" vs. !" . At the exact point of crossing, the rate of increase of the spiral is equal to the rms rate of increase. The more frequently crossing occurs, the more the change rate is fluctuating about its rms value, indicating greater irregularity. Compared with smoothness indices, the zero crossing indices are more sensitive to small or frequent fluctuations.

    ZeroCross2 = i5 = 2ndOrderSecond0CrossingRate ** -- SECOND ORDER ZERO CROSSING RATE The second order zero crossing rate is the rate of change (first derivative) of the first order zero crossing rate. This index provides valuable additional information concerning how “regularly” or “irregularly” irregular a given spiral is. If the second order zero crossing rate is high, it means a large proportion of the spiral not only contains irregularities, but that these irregularities are different from place to place on the spiral. For example, a single frequency tremor along one axis throughout a spiral would result in a low second order zero crossing rate, whereas unevenly timed jerking motions along many axes with differing amplitudes would result in a high second order zero crossing rate.

    The indices marked with ** are used:

    DOS = DegreeOfSeverity = 0.4615 * i1 + 0.0544 * i5 - 0.2331 * i1**2 - 0.0726 * i2**2 - 0.001 * i5**2 + 0.2539*i1*i2 + 1.3688
    DEGREE OF SEVERITY Degree of severity provides an overall evaluation of a spiral that has been clinically validated to reflect a meaningful and comparative assessment of spiral execution.
    This important calculation is based on spiral data originally obtained from normal controls and patients with movement disorders that were used to “train” the spiral analysis program to rate the severity of handwritten spirals on a 5-point scale from 0 to 4 as
    follows:
    0 to 1 = normal,
    1 to 2 = mildly,
    2 to 3 = moderately
    3 to 4 = severely
    abnormal spirals.

    Program “training” was achieved through multiple regressions of indices against average spiral ratings from experienced movement disorder neurologists. Simple linear and second order polynomial regressions (to detect interactions between indices) were performed for each index against expert physician averages. As a result, three indices (first order smoothness, second order smoothness and second order zero crossing) with the highest significance, highest r2 and lowest residuals created a degree of severity or DOS equation, the output of which now is a clinical score from 0 to 4 inclusive:

    In addition this External App calculates
    ResThetaR -- RESIDUAL OF THETA-R The “residual” of the graph of the radius (r) vs. the angle (theta, or ! ) reflects the overall distance between the transformed spiral data points and the line of best fit on the graph of radius length vs. angle change. The larger this index, the more the spiral is changing its radius in an irregular fashion.

    CenterOfSpiralX

    CenterOfSpiralY

    OutwardClockwise

    Many more parameters can be calculated in the future:

    i3 = Tightness -- TIGHTNESS Tightness is a measure of average spacing between consecutive loops across the spiral. When data is acquired, patients are asked to draw each spiral is within a 10 x 10 cm box. Tightness is a relative measure, based on the definition that a spiral with five loops in the 10 x 10 cm box has a tightness value of 1. Spirals with more than five loops within 10 x 10 cm have tightness values > 1, and spirals with fewer than five loops within 10 x 10 cm have tightness values < 1. High tightness is the mathematical equivalent of micrographia. Normalizing scores relative to a value of 5 loops/10cm is based on healthy control data showing values in this range.
    Company http://www.neuroscript.net and work http://www.neuroscript.net/hans_leo_teulings.php on handwriting and drawing movement recording and processing.

  3. Join Date
    Aug 2009
    Posts
    244
    Name:  SPI.PNG
Views: 752
Size:  11.5 KB
    - Please download the new experiment SPI.MEF SPI.MEF
    - Please save the MovAlyzeR Export File of Experiment SPI (SPI.MEF). (2010-12-06)
    (Do not change the name)
    In MovAlyzeR: Click File
    In MovAlyzeRx click in the top left circle
    then: Import >Select SPI.MEF
    Last edited by Ameya; 2010-12-10 at 12:11.

+ Reply to Thread

Thread Information

Users Browsing this Thread

There are currently 2 users browsing this thread. (0 members and 2 guests)

     

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts