Convolution Neural Network Algorithm for Shockable Arrhythmia Classification Within a Digitally Connected Automated External Defibrillator.

Background Diagnosis of shockable rhythms leading to defibrillation remains integral to improving out-of-hospital cardiac arrest outcomes. New machine learning techniques have emerged to diagnose arrhythmias on ECGs. In out-of-hospital cardiac arrest, an algorithm within an automated external defibrillator is the major determinant to deliver defibrillation. This study developed and validated the performance of a convolution neural network (CNN) to diagnose shockable arrhythmias within a novel, miniaturized automated external defibrillator. Methods and Results There were 26 464 single-lead ECGs that comprised the study data set. ECGs of 7-s duration were retrospectively adjudicated by 3 physician readers (N=18 total readers). After exclusions (N=1582), ECGs were divided into training (N=23 156), validation (N=721), and test data sets (N=1005). CNN performance to diagnose shockable and nonshockable rhythms was reported with area under the receiver operating characteristic curve analysis, F1, and sensitivity and specificity calculations. The duration for the CNN to output was reported with the algorithm running within the automated external defibrillator. Internal and external validation analyses included CNN performance among arrhythmias, often mistaken for shockable rhythms, and performance among ECGs modified with noise to mimic artifacts. The CNN algorithm achieved an area under the receiver operating characteristic curve of 0.995 (95% CI, 0.990-1.0), sensitivity of 98%, and specificity of 100% to diagnose shockable rhythms. The F1 scores were 0.990 and 0.995 for shockable and nonshockable rhythms, respectively. After input of a 7-s ECG, the CNN generated an output in 383±29 ms (total time of 7.383 s). The CNN outperformed adjudicators in classifying atrial arrhythmias as nonshockable (specificity of 99.3%-98.1%) and was robust against noise artifacts (area under the receiver operating characteristic curve range, 0.871-0.999). Conclusions We demonstrate high diagnostic performance of a CNN algorithm for shockable and nonshockable rhythm arrhythmia classifications within a digitally connected automated external defibrillator. Registration URL: https://clinicaltrials.gov/ct2/show/NCT03662802; Unique identifier: NCT03662802.

High-Affinity Interaction of the K-Ras4B Hypervariable Region with the Ras Active Site.

Ras proteins are small GTPases that act as signal transducers between cell surface receptors and several intracellular signaling cascades. They contain highly homologous catalytic domains and flexible C-terminal hypervariable regions (HVRs) that differ across Ras isoforms. KRAS is among the most frequently mutated oncogenes in human tumors. Surprisingly, we found that the C-terminal HVR of K-Ras4B, thought to minimally impact the catalytic domain, directly interacts with the active site of the protein. The interaction is almost 100-fold tighter with the GDP-bound than the GTP-bound protein. HVR binding interferes with Ras-Raf interaction, modulates binding to phospholipids, and slightly slows down nucleotide exchange. The data indicate that contrary to previously suggested models of K-Ras4B signaling, HVR plays essential roles in regulation of signaling. High affinity binding of short peptide analogs of HVR to K-Ras active site suggests that targeting this surface with inhibitory synthetic molecules for the therapy of KRAS-dependent tumors is feasible.

GTP-Dependent K-Ras Dimerization.

Ras proteins recruit and activate effectors, including Raf, that transmit receptor-initiated signals. Monomeric Ras can bind Raf; however, activation of Raf requires its dimerization. It has been suspected that dimeric Ras may promote dimerization and activation of Raf. Here, we show that the GTP-bound catalytic domain of K-Ras4B, a highly oncogenic splice variant of the K-Ras isoform, forms stable homodimers. We observe two major dimer interfaces. The first, highly populated ß-sheet dimer interface is at the Switch I and effector binding regions, overlapping the binding surfaces of Raf, PI3K, RalGDS, and additional effectors. This interface has to be inhibitory to such effectors. The second, helical interface also overlaps the binding sites of some effectors. This interface may promote activation of Raf. Our data reveal how Ras self-association can regulate effector binding and activity, and suggest that disruption of the helical dimer interface by drugs may abate Raf signaling in cancer.