Hardware-Efficient 1D CNN for Patient-Specific Early Seizure Detection.

Closed-loop brain-implantable neuromodulation devices are a new treatment option for patients with refractory epilepsy. Seizure detection algorithms implemented on such devices are subject to strict power and area constraints. Deep learning methods, though very powerful, tend to have high computational complexity and thus are typically impractical for resource-constrained neuromodulation devices. In this paper, we propose a compact and hardware-efficient one-dimensional convolutional neural network (1D CNN) structure for patient-specific early seizure detection. Feature extraction techniques and a novel initialization method based on the forward-chaining training and testing scheme are used to improve model performance. Our compact model achieves similar accuracy to that of support vector machines, the state-of-the-art method for seizure detection, while consuming over 20x less power.

Machine Learning with Imbalanced EEG Datasets using Outlier-based Sampling.

Epilepsy is a neurological disorder which causes seizures in over 65 million people worldwide. Recently developed implantable therapeutic devices aim to prevent symptoms by applying acute electrical stimulation to the seizure-generating brain region in response to activity detected by on-device machine learning hardware. Many training algorithms require an equal number of examples for each target class (e.g. normal activity and seizures), and performance can suffer if this condition is not satisfied. In the case of epilepsy, poor performance can cause seizures to be missed, or stimulation to be applied erroneously. As there is an abundance of normal (interictal) data in clinical EEG recordings, but seizures are rare events (less than 1% of the dataset), the data available for training is severely imbalanced. There are several conventional pre-processing methods used to address imbalanced class learning, such as down-sampling of the majority class and up-sampling of the minority class, but each have performance drawbacks. This paper presents an improved method which involves reducing the majority class down to the most effective interictal outlier samples. Outliers are determined by using Exponentially Decaying Memory Signal Energy (EDMSE) features with Isolation Forests and an ANOVA-based method, which involves comparing a moving feature window to a baseline reference window. Outlier-based sampling is tested with two classifiers (KNN and Logistic Regression) and achieves higher accuracy (∼2% increase) and fewer false positives (∼38% decrease), along with a lower latency (∼3 seconds shorter) compared to conventional training set pre-processing methods.

Smoking Cessation Following Text Message Intervention in Pregnant Women.

BACKGROUND: Smoking during pregnancy has detrimental effects on mother and fetus. Text messaging has been utilized to improve patient care. OBJECTIVE: To evaluate the impact of text messaging on smoking cessation rates among pregnant women in addition to standard of care (SOC) smoking cessation services. Our SOC includes pharmacist-driven education with or without nicotine patch or bupropion. METHODS: This randomized, open-label, prospective trial was conducted at a maternal fetal care center from May 2014 to January 2016. Pregnant patients in the preparation stage of change were randomized to text messaging or SOC. The primary outcome was smoking cessation verified with exhaled carbon monoxide levels (eCO) 2 weeks from quit date. All received clinical pharmacist weekly calls for 3 weeks and biweekly visits until pharmacotherapy completion. The text messaging group also received predetermined motivational messages. RESULTS: Of 49 randomized patients, 13 withdrew, and 6 were lost to follow-up. The remaining included 14 texting and 16 SOC patients. eCO-verified cessation was achieved by 57.1% in the texting group versus 31.3% in the control ( P = 0.153). Overall, 64.3% of the texting group achieved an eCO below 8 ppm at ≥1 visit versus 37.5% in the control group ( P = 0.143). No difference was found in birth outcomes. The study was underpowered because of slow enrollment and high drop-out rates. CONCLUSIONS AND RELEVANCE: Text messaging had minimal impact on improving smoking cessation rates in the obstetric population. However, further research is warranted because of the underpowered nature of this trial. Given the detrimental effects of smoking in pregnancy, more comprehensive cessation strategies are warranted.

Administration Rates of the Tdap Vaccine in Obstetric Patients.

BACKGROUND: Infants younger than 6 months of age are at high risk for contracting pertussis because of not being fully vaccinated. The Advisory Committee on Immunization Practices (ACIP) recommends vaccinating all pregnant women with tetanus toxoid, reduced diphtheria toxoid, and acellular pertussis vaccine (Tdap) between 27 and 36 weeks to offer passive immunity to the infant to help protect them until they are able to receive the full pertussis series. OBJECTIVE: To assess and compare compliance with the 2013 ACIP recommendation of vaccinating pregnant women with Tdap at 27 to 36 weeks' gestation in 2 obstetric clinics. METHODS: This cross-sectional, retrospective chart review evaluated Tdap vaccine compliance in a random sample of obstetric patients from October 2013 to September 2014. The primary outcome evaluated the proportion of patients who received Tdap between 27 and 36 weeks' gestation. Secondary outcomes included the proportion of patients who received Tdap at any point in pregnancy and within 30 days postpartum. RESULTS: The charts of 573 patients were reviewed, and 237 met inclusion criteria. For the primary outcome, 142 patients (59.9%) received the Tdap vaccine. Overall, 156 patients (65.8%) received Tdap at some point during the pregnancy. Factors associated with receiving the Tdap vaccination were insurance status, prenatal care risk level and site of prenatal care, receipt of the influenza vaccine, and preterm labor in the current pregnancy. CONCLUSION: The Tdap vaccine rate was 65.8%, with 59.9% of patients receiving the vaccine within the recommended ACIP timeframe. Further education, improvements in documentation, and chart reminders are needed to enhance administration.