Background-Cardiac resynchronization therapy (CRT) has significant non-response rates. We assessed whether machine learning could predict CRT response beyond current guidelines. Methods-We analyzed CRT patients from Cleveland Clinic and Johns Hopkins. A training cohort was created from all Johns Hopkins patients and an equal number of randomly sampled Cleveland Clinic patients. All remaining patients comprised the testing cohort. Response was defined as ≥10% increase in left ventricular (LV) ejection fraction. Machine learning models were developed to predict CRT response using different combinations of classification algorithms and clinical variable sets on the training cohort. The model with the highest area under curve (AUC) was evaluated on the testing cohort. Probability of response was used to predict survival free from a composite endpoint of death, heart transplant, or placement of LV assist device. Predictions were compared to current guidelines.

Despite the promising potential of microfluidic artificial lungs, current designs suffer from short functional lifetimes due to surface chemistry and blood flow patterns that act to reduce hemocompatibility. Here, we present the first microfluidic artificial lung featuring a hemocompatible surface coating and a biomimetic blood path. The polyethylene-glycol (PEG) coated microfluidic lung exhibited a significantly improved in vitro lifetime compared to uncoated controls as well as consistent and significantly improved gas exchange over the entire testing period. Enabled by our hemocompatible PEG coating, we additionally describe the first extended (3 h) in vivo demonstration of a microfluidic artificial lung.

Background - We hypothesized that computerized morphologic analysis of the LA and pulmonary veins (PVs) via fractal measurements of shape and texture features of the LA myocardial wall could predict AF recurrence after ablation. Methods - Pre-ablation contrast CT scans were collected for 203 patients who underwent AF ablation. The LA body, PVs, and myocardial wall were segmented using a semi-automated region growing method. Twenty-eight fractal-based shape and texture-based features were extracted from resulting segments. The top features most associated with post-ablation recurrence were identified using feature selection and subsequently evaluated with a Random Forest classifier. Feature selection and classifier construction were performed on a discovery cohort (D 1 ) of 137 patients; classifiers were subsequently validated on an independent set (D 2 ) of 66 patients. Dedicated classifiers to capture the fractal and morphologic properties of LA body (C LA ), PVs (C PV ), and LA myocardial (C LAM ) tissue were constructed, as well as a model (C All ) capturing properties of all segmented compartments. Fractal-based models were also compared against a model employing machine estimation of LA volume. To assess the effect of clinical parameters, such as AF type and catheter technique, a clinical model (C clin ) was also compared against C All . Results - Statistically significant differences were observed for fractal features of C LA , C LAM and C All in distinguishing AF recurrence (p<0.001) on D 1 . Using the five top features, C All had the best prediction performance (AUC=0.81 [95% Confidence Interval (CI): 0.78-0.85]), followed by C PV (AUC=0.78 [95% CI: 0.74-0.80]) and C LA (AUC=0.70 [95% CI: 0.63-0.78]) on D 2 . The clinical parameter model C clin yielded an AUC=0.70 [95% CI: 0.65-0.77], while the atrial volume model yielded an AUC=0.59. Combining C All and C clin on D 2 improved the AUC to 0.87 [95% CI: 0.82-0.93]. Conclusions - Fractal measurements of the LA, PVs, and atrial myocardium on CT scans were associated with likelihood of post-ablation AF recurrence.