[Impact of baseline pulse pressure on new-onset cardio-and cerebrovascular events in diabetic population].

OBJECTIVE: To observe the association between baseline pulse pressure (PP) level and new-onset cardio-cerebrovascular events in diabetic population. METHODS: Physical examination data between July 2006 to October 2007 from a total of 101 510 employees of Kailuan Group were reviewed, 8306 subjects with a fasting plasma glucose level of ≥ 7.0 mmol/L or with confirmed diabetes diagnosis and were enrolled in this prospective cohort study. Subjects were followed up for 38-53 (48.1 ± 3.1) months and the cardio-cerebrovascular events were obtained every six months, association between baseline PP and new-onset cardio-cerebrovascular events in the diabetic population were analyzed. RESULTS: (1) Incidences of total cardio-cerebrovascular events in the PP groups were 3.4%, 2.8%, 4.5%, 6.4%, respectively. Incidences of cerebral infarction events and myocardial infarction were 2.1%, 1.6%, 2.9%, 3.9% and 1.1%, 0.7%, 1.0%, 1.7%, respectively. (2) Multivariate Cox's proportional hazards regression analysis indicated that baseline PP group was the risk factor for total cardio-cerebrovascular events, cerebral infarction events and myocardial infarction, and the risk for all the events of the PP ≥ 60 mm Hg (1 mm Hg = 0.133 kPa) group was increasing. The values of RR(95%CI) were 1.88 (95%CI 1.34-2.65, P < 0.01), 1.92 (95%CI 1.23-2.99, P < 0.01) and 1.52 (95%CI 0.82-2.81, P > 0.05) after adjust the other factors.(3) In line with increasing level of baseline PP, age, BMI, SBP, DBP, HDL-C, and hs-CRP levels significantly increased in this diabetic population (P < 0.01 or P < 0.05). CONCLUSION: The level of high baseline PP is a risk factor for new-onset cardio-cerebrovascular events in diabetic population.

Machine Learning to Classify Relative Seizure Frequency From Chronic Electrocorticography.

PURPOSE: Brain responsive neurostimulation (NeuroPace) treats patients with refractory focal epilepsy and provides chronic electrocorticography (ECoG). We explored how machine learning algorithms applied to interictal ECoG could assess clinical response to changes in neurostimulation parameters. METHODS: We identified five responsive neurostimulation patients each with ≥200 continuous days of stable medication and detection settings (median, 358 days per patient). For each patient, interictal ECoG segments for each month were labeled as "high" or "low" to represent relatively high or low long-episode (i.e., seizure) count compared with the median monthly long-episode count. Power from six conventional frequency bands from four responsive neurostimulation channels were extracted as features. For each patient, five machine learning algorithms were trained on 80% of ECoG, then tested on the remaining 20%. Classifiers were scored by the area-under-the-receiver-operating-characteristic curve. We explored how individual circadian cycles of seizure activity could inform classifier building. RESULTS: Support vector machine or gradient boosting models achieved the best performance, ranging from 0.705 (fair) to 0.892 (excellent) across patients. High gamma power was the most important feature, tending to decrease during low-seizure-frequency epochs. For two subjects, training on ECoG recorded during the circadian ictal peak resulted in comparable model performance, despite less data used. CONCLUSIONS: Machine learning analysis on retrospective background ECoG can classify relative seizure frequency for an individual patient. High gamma power was the most informative, whereas individual circadian patterns of seizure activity can guide model building. Machine learning classifiers built on interictal ECoG may guide stimulation programming.