Elizabeth Comitz

Classification of EEG Features for Prediction of Working Memory Load

The objective of this research was to compare classification methods aimed at predicting working memory (WM) load. Electroencephalogram (EEG) data was collected from physicians while performing basic WM tasks and simulated medical scenarios. Data processing was performed to remove noise from the signal used for analysis (e.g., muscle activity, eye-blinks). The data from basic WM tasks was used to develop and test the four classification models (LASSO regression, support vector machines (SVM), nearest shrunken centroids (NSC), and iterated supervised principal components (ISPC) to predict a WM state indicative of physicians’ optimal performance. The naive misclassification rate was 19.74 %; LASSO and SVM outperformed this threshold: 18.10 and 12.21 % respectively). Both classification models had relatively high-specificity (LASSO: 97.2 %; SVM: 99.8 %); but relatively low-sensitivity LASSO: 20.7 %; SVM: 39.6 %). Results from simulated medical scenarios suggest that physicians were approximately 83 % of the time in the WM state that is likely indicative of optimal performance.

Assessing the Quality of the A3 Thinking Tool for Problem Solving

The objective of this pilot study was to assess the inter-rater reliability of a newly developed A3 Quality Assessment (QA) rubric to evaluate the quality of completed Plan-Do-Study-Act (PDSA) projects that used an A3 Thinking Tool (A3) for problem solving. One A3 was independently reviewed by 7 PDSA experts using 5 main levels and 22 sublevels. Evaluations were compared and coded for agreement and used for statistical analysis. Fleiss’ kappa statistics was performed to test for inter-rater reliability between experts across 5 main and 22 sublevels. Preliminary results suggest that the A3 QA rubric meets reliability criteria with a moderate level of agreement beyond chance alone (κ = 0.44) and it is applicable to measure progress on problem solving abilities spearheaded via PDSA cycles. Additional verification testing is needed across multiple A3 improvement projects completed in multiple A3 Thinking templates.