This research explores novel methods for extracting relevant information from EEG data to characterize individual differences in cognitive processing. Our approach combines expertise in machine learning, statistics, and cognitive science, advancing the state-of-the art in all three domains. Specifically, by using cognitive science expertise to interpret results and inform algorithm development, we have developed a generalizable and interpretable machine learning method that can accurately predict individual differences in cognition. The output of the machine learning method revealed surprising features of the EEG data that, when interpreted by the cognitive science experts, provided novel insights to the underlying cognitive task. Additionally, the outputs of the statistical methods show promise as a principled approach to quickly find regions within the EEG data where individual differences lie, thereby supporting cognitive science analysis and informing machine learning models. This work lays methodological ground work for applying the large body of cognitive science literature on individual differences to high consequence mission applications.
Pothoulakis, Ioannis P.; Paragomi, Pedram P.; Tuft, Marie T.; Lahooti, Ali L.; Archibugi, Livia A.; Capurso, Gabriele C.; Papachristou, Georgios I.
Hypertriglyceridemia (HTG) is considered within the top 5 etiologies in acute pancreatitis (AP), but the association of serum triglyceride (TG) levels with the clinical course of AP remains controversial. This study aims to examine the effect of TG levels on severity of AP. Patients were enrolled prospectively through APPRENTICE. High TG levels were defined based on the Endocrine Society Clinical Practice Guidelines. HTG was categorized as mild (serum TG levels 150–199 mg/dL), moderate (200–999 mg/dL), severe (1,000–1,999 mg/dL), and very severe (≥2,000 mg/dL). Severity of AP was based on the revised Atlanta classification criteria. Additionally, early TG levels were measured in 764 subjects and found elevated in 342 (120 with mild; 176, moderate; and 46, severe/very severe HTG). Patients with increased TG levels were younger (age ≥60, 16.7 vs. 30.3%), more likely to be male (66.1 vs. 51.2%), with more frequent alcohol use (62.8 vs. 50.7%), and diabetes mellitus (30.2 vs. 12.3%; all p ≤ 0.005). Severe AP (24.9 vs. 10.0%), ICU admission (32.5 vs. 19.7%), and mortality (5.3 vs. 1.7%; all p ≤ 0.005) were more frequently seen in patients with elevated TG levels. Based on multivariable analysis, elevated TG levels were independently associated with severe AP (p < 0.05). This large multicenter study confirms that elevated TG levels are associated with severe disease regardless of AP etiology.
Gandley, Robin G.; Brands, Judith B.; Tuft, Marie T.; Jeyabalan, Arun J.; Santillan, Mark K.; Pierce, Gary L.; Tang, Gong T.; Hubel, Carl A.; Catov, Janet M.