This week, I am working to get the dynamic time warping functionality into my program. The process of doing so includes re-processing the features to include the time series, putting each series back together when we construct sequences, and then performing the DTW to generate a number that will be used to compute the kNN of each sequence which can then be used for predictions with the models. The processing time of these activities has gone up significantly since we have been using five different metrics with each of the F phase datasets. I am returning to school next week, and once I've completed the DTW processing all that will remain before we put together our second paper (The date for the reach journal we would like to submit it to is October 1), I am hoping I will have time to look again into the Agglomerative Hierarchical Clustering concept, which I did not successfully complete when we explored it earlier in the summer and then changed focus to the paper. We heard back
This week, I continued to look at measures of daytime activity and how they may relate to sleep the night before. Upon our realizations about the difference in actigraph sensitivity between subjects, I wanted to look for a way to tailor standards of low (sedentary-like) activity counts to individual patients. I decided to look at the first, second and third quartile thresholds in order to see how counts were distributed across the daytime (lights on to lights off). I also decided to look at the quartiles for nighttime activity. This provided an interesting look at the way the two "phases" of actigraphs were different from one another since there were no patients in the first phase who had median counts of 0 during the night, a characteristic of almost all second phase subjects.
Going forward, I want to plot out minutes per day that subjects spent below their baseline Q1 threshold and see if there is correlation between this number and the number of sleep minutes the night before, and also the night after. Baseline days will not be involved in these calculations.
This exploration lead me to think more about the way that we've defined sleep so far, as it has not taken the individual's behavior into account. However, I am hesitant to apply the exact same approach I want to use for sedentary-like behavior because of the results I've gotten on nighttime quartiles.
Dr. Sprint, Alexa and I are going to meet with Dr. Weeks and Dr. Skornyakov next week to discuss the next steps in the project.
Going forward, I want to plot out minutes per day that subjects spent below their baseline Q1 threshold and see if there is correlation between this number and the number of sleep minutes the night before, and also the night after. Baseline days will not be involved in these calculations.
This exploration lead me to think more about the way that we've defined sleep so far, as it has not taken the individual's behavior into account. However, I am hesitant to apply the exact same approach I want to use for sedentary-like behavior because of the results I've gotten on nighttime quartiles.
Dr. Sprint, Alexa and I are going to meet with Dr. Weeks and Dr. Skornyakov next week to discuss the next steps in the project.
Comments
Post a Comment