Week 3 - Charlie Heinke

This week I made major progress in my project. Using Python, I succeeded in producing an easy-to-use program for assessing the precision of transcranial magnetic stimulation. While this calculation has been achieved before with other neuronavigation software, the system used by my mentor, Localite, does not include this capability.

First, my program asks the user to provide some information about the data it is analyzing: the type of session (diagnostic or therapeutic), the assumed scalp-cortex distance (more on that later), and the tolerated distance from the target site in mm. The user is then prompted to navigate to the patient folder generated by Localite's neuronavigation system. Once this has been selected, my program outputs the average distance from the target, the standard deviation of this distance, the total number of applied stimuli, total number of accurate stimuli (within the tolerated range), and the percent accuracy of stimulation. Along with this info, the following figures are generated:

Figure 1 shows the deviation from the target as the TMS session progresses. In this particular session, the coil drifted from the target position. Figure 2 is a histogram showing the spread of distances. Average for this session was about 1-2mm. Figure 3 is a 3D plot of stimulation coordinates. The big cyan dot in the center is the target location. Dark blue dots are simuli within the user-defined "tolerated range", while red are outside. The blue arrows extending from each dot show the direction of the magnetic field.

Now that this system is established, I am going to compare my method of analyzing precision with an open-source program I found on the internet. This program cannot directly use the data generated by Localite's system because this program only takes coordinates in text format, while the data from Localite is contained in .xml files. On Monday I will edit my code to output stimuli coordinates into a format compatible with this other program and see how it compares.