Patient-Level Artificial Intelligence-Enhanced Electrocardiography in Hypertrophic Cardiomyopathy: Longitudinal Treatment and Clinical Biomarker Correlations.

BACKGROUND: Artificial intelligence (AI) applied to 12-lead electrocardiographs (ECGs) can detect hypertrophic cardiomyopathy (HCM). OBJECTIVES: The purpose of this study was to determine if AI-enhanced ECG (AI-ECG) can track longitudinal therapeutic response and changes in cardiac structure, function, or hemodynamics in obstructive HCM during mavacamten treatment. METHODS: We applied 2 independently developed AI-ECG algorithms (University of California-San Francisco and Mayo Clinic) to serial ECGs (n = 216) from the phase 2 PIONEER-OLE trial of mavacamten for symptomatic obstructive HCM (n = 13 patients, mean age 57.8 years, 69.2% male). Control ECGs from 2,600 age- and sex-matched individuals without HCM were obtained. AI-ECG output was correlated longitudinally to echocardiographic and laboratory metrics of mavacamten treatment response. RESULTS: In the validation cohorts, both algorithms exhibited similar performance for HCM diagnosis, and exhibited mean HCM score decreases during mavacamten treatment: patient-level score reduction ranged from approximately 0.80 to 0.45 for Mayo and 0.70 to 0.35 for USCF algorithms; 11 of 13 patients demonstrated absolute score reduction from start to end of follow-up for both algorithms. HCM scores were significantly associated with other HCM-relevant parameters, including left ventricular outflow tract gradient at rest, postexercise, and with Valsalva, and NT-proBNP level, independent of age and sex (all P < 0.01). For both algorithms, the strongest longitudinal correlation was between AI-ECG HCM score and left ventricular outflow tract gradient postexercise (slope estimate: University of California-San Francisco 0.70 [95% CI: 0.45-0.96], P < 0.0001; Mayo 0.40 [95% CI: 0.11-0.68], P = 0.007). CONCLUSIONS: AI-ECG analysis longitudinally correlated with changes in echocardiographic and laboratory markers during mavacamten treatment in obstructive HCM. These results provide early evidence for a potential paradigm for monitoring HCM therapeutic response.

Automated Assessment of Cardiac Systolic Function From Coronary Angiograms With Video-Based Artificial Intelligence Algorithms.

Importance: Understanding left ventricular ejection fraction (LVEF) during coronary angiography can assist in disease management. Objective: To develop an automated approach to predict LVEF from left coronary angiograms. Design, Setting, and Participants: This was a cross-sectional study with external validation using patient data from December 12, 2012, to December 31, 2019, from the University of California, San Francisco (UCSF). Data were randomly split into training, development, and test data sets. External validation data were obtained from the University of Ottawa Heart Institute. Included in the analysis were all patients 18 years or older who received a coronary angiogram and transthoracic echocardiogram (TTE) within 3 months before or 1 month after the angiogram. Exposure: A video-based deep neural network (DNN) called CathEF was used to discriminate (binary) reduced LVEF (≤40%) and to predict (continuous) LVEF percentage from standard angiogram videos of the left coronary artery. Guided class-discriminative gradient class activation mapping (GradCAM) was applied to visualize pixels in angiograms that contributed most to DNN LVEF prediction. Results: A total of 4042 adult angiograms with corresponding TTE LVEF from 3679 UCSF patients were included in the analysis. Mean (SD) patient age was 64.3 (13.3) years, and 2212 patients were male (65%). In the UCSF test data set (n = 813), the video-based DNN discriminated (binary) reduced LVEF (≤40%) with an area under the receiver operating characteristic curve (AUROC) of 0.911 (95% CI, 0.887-0.934); diagnostic odds ratio for reduced LVEF was 22.7 (95% CI, 14.0-37.0). DNN-predicted continuous LVEF had a mean absolute error (MAE) of 8.5% (95% CI, 8.1%-9.0%) compared with TTE LVEF. Although DNN-predicted continuous LVEF differed 5% or less compared with TTE LVEF in 38.0% (309 of 813) of test data set studies, differences greater than 15% were observed in 15.2% (124 of 813). In external validation (n = 776), video-based DNN discriminated (binary) reduced LVEF (≤40%) with an AUROC of 0.906 (95% CI, 0.881-0.931), and DNN-predicted continuous LVEF had an MAE of 7.0% (95% CI, 6.6%-7.4%). Video-based DNN tended to overestimate low LVEFs and underestimate high LVEFs. Video-based DNN performance was consistent across sex, body mass index, low estimated glomerular filtration rate (≤45), presence of acute coronary syndromes, obstructive coronary artery disease, and left ventricular hypertrophy. Conclusion and relevance: This cross-sectional study represents an early demonstration of estimating LVEF from standard angiogram videos of the left coronary artery using video-based DNNs. Further research can improve accuracy and reduce the variability of DNNs to maximize their clinical utility.

Deep learning augmented ECG analysis to identify biomarker-defined myocardial injury.

Chest pain is a common clinical complaint for which myocardial injury is the primary concern and is associated with significant morbidity and mortality. To aid providers' decision-making, we aimed to analyze the electrocardiogram (ECG) using a deep convolutional neural network (CNN) to predict serum troponin I (TnI) from ECGs. We developed a CNN using 64,728 ECGs from 32,479 patients who underwent ECG within 2 h prior to a serum TnI laboratory result at the University of California, San Francisco (UCSF). In our primary analysis, we classified patients into groups of TnI < 0.02 or ≥ 0.02 µg/L using 12-lead ECGs. This was repeated with an alternative threshold of 1.0 µg/L and with single-lead ECG inputs. We also performed multiclass prediction for a set of serum troponin ranges. Finally, we tested the CNN in a cohort of patients selected for coronary angiography, including 3038 ECGs from 672 patients. Cohort patients were 49.0% female, 42.8% white, and 59.3% (19,283) never had a positive TnI value (≥ 0.02 µg/L). CNNs accurately predicted elevated TnI, both at a threshold of 0.02 µg/L (AUC = 0.783, 95% CI 0.780-0.786) and at a threshold of 1.0 µg/L (AUC = 0.802, 0.795-0.809). Models using single-lead ECG data achieved significantly lower accuracy, with AUCs ranging from 0.740 to 0.773 with variation by lead. Accuracy of the multi-class model was lower for intermediate TnI value-ranges. Our models performed similarly on the cohort of patients who underwent coronary angiography. Biomarker-defined myocardial injury can be predicted by CNNs from 12-lead and single-lead ECGs.

Advancing cardiovascular medicine with machine learning: Progress, potential, and perspective.

Recent advances in machine learning (ML) have made it possible to analyze high-dimensional and complex data-such as free text, images, waveforms, videos, and sound-in an automated manner by successfully learning complex associations within these data. Cardiovascular medicine is particularly well poised to take advantage of these ML advances, due to the widespread digitization of medical data and the large number of diagnostic tests used to evaluate cardiovascular disease. Various ML approaches have successfully been applied to cardiovascular tests and diseases to automate interpretation, accurately perform measurements, and, in some cases, predict novel diagnoses from less invasive tests, effectively expanding the utility of more widely accessible diagnostic tests. Here, we present examples of some impactful advances in cardiovascular medicine using ML across a variety of modalities, with a focus on deep learning applications.

Regenerated interneurons integrate into locomotor circuitry following spinal cord injury.

Whereas humans and other adult mammals lack the ability to regain locomotor function after spinal cord injury, zebrafish are able to recover swimming behavior even after complete spinal cord transection. We have previously shown that zebrafish larvae regenerate lost spinal cord neurons within 9 days post-injury (dpi), but it is unknown whether these neurons are physiologically active or integrate into functional circuitry. Here we show that genetically defined premotor interneurons are regenerated in injured spinal cord segments as functional recovery begins. Further, we show that these newly-generated interneurons receive excitatory input and fire synchronously with motor output by 9 dpi. Taken together, our data indicate that regenerative neurogenesis in the zebrafish spinal cord produces interneurons with the ability to integrate into existing locomotor circuitry.

Hypothalamic Dopamine Neurons Control Sensorimotor Behavior by Modulating Brainstem Premotor Nuclei in Zebrafish.

Dopamine (DA)-producing neurons are critically involved in the production of motor behaviors in multiple circuits that are conserved from basal vertebrates to mammals. Although there is increasing evidence that DA neurons in the hypothalamus play a locomotor role, their precise contributions to behavior and the circuit mechanisms by which they are achieved remain unclear. Here, we demonstrate that tyrosine-hydroxylase-2-expressing (th2+) DA neurons in the zebrafish hypothalamus fire phasic bursts of activity to acutely promote swimming and modulate audiomotor behaviors on fast timescales. Their anatomy and physiology reveal two distinct functional DA modules within the hypothalamus. The first comprises an interconnected set of cerebrospinal-fluid-contacting DA nuclei surrounding the 3rd ventricle, which lack distal projections outside of the hypothalamus and influence locomotion through unknown means. The second includes neurons in the preoptic nucleus, which send long-range projections to targets throughout the brain, including the mid- and hindbrain, where they activate premotor circuits involved in swimming and sensorimotor integration. These data suggest a broad regulation of motor behavior by DA neurons within multiple hypothalamic nuclei and elucidate a novel functional mechanism for the preoptic DA neurons in the initiation of movement.

Zebrafish oxytocin neurons drive nocifensive behavior via brainstem premotor targets.

Animals have evolved specialized neural circuits to defend themselves from pain- and injury-causing stimuli. Using a combination of optical, behavioral and genetic approaches in the larval zebrafish, we describe a novel role for hypothalamic oxytocin (OXT) neurons in the processing of noxious stimuli. In vivo imaging revealed that a large and distributed fraction of zebrafish OXT neurons respond strongly to noxious inputs, including the activation of damage-sensing TRPA1 receptors. OXT population activity reflects the sensorimotor transformation of the noxious stimulus, with some neurons encoding sensory information and others correlating more strongly with large-angle swims. Notably, OXT neuron activation is sufficient to generate this defensive behavior via the recruitment of brainstem premotor targets, whereas ablation of OXT neurons or loss of the peptide attenuates behavioral responses to TRPA1 activation. These data highlight a crucial role for OXT neurons in the generation of appropriate defensive responses to noxious input.

The Midline Axon Crossing Decision Is Regulated through an Activity-Dependent Mechanism by the NMDA Receptor.

Axon guidance in vertebrates is controlled by genetic cascades as well as by intrinsic activity-dependent refinement of connections. Midline axon crossing is one of the best studied pathfinding models and is fundamental to the establishment of bilaterally symmetric nervous systems. However, it is not known whether crossing requires intrinsic activity in axons, and what controls that activity. Further, a mechanism linking neuronal activity and gene expression has not been identified for axon pathfinding. Using embryonic zebrafish, we found that the NMDA receptor (NMDAR) NR1.1 subunit (grin1a) is expressed in commissural axons. Pharmacological inhibition of grin1a, hypoxia exposure reduction of grin1a expression, or CRISPR knock-down of grin1a leads to defects in midline crossing. Inhibition of neuronal activity phenocopies the effects of grin1a loss on midline crossing. By combining pharmacological inhibition of the NMDAR with optogenetic stimulation to precisely restore neuronal activity, we observed rescue of midline crossing. This suggests that the NMDAR controls pathfinding by an activity-dependent mechanism. We further show that the NMDAR may act, via modulating activity, on the transcription factor arxa (mammalian Arx), a known regulator of midline pathfinding. These findings uncover a novel role for the NMDAR in controlling activity to regulate commissural pathfinding and identify arxa as a key link between the genetic and activity-dependent regulation of midline axon guidance.

Motor Behavior Mediated by Continuously Generated Dopaminergic Neurons in the Zebrafish Hypothalamus Recovers after Cell Ablation.

Postembryonic neurogenesis has been observed in several regions of the vertebrate brain, including the dentate gyrus and rostral migratory stream in mammals, and is required for normal behavior [1-3]. Recently, the hypothalamus has also been shown to undergo continuous neurogenesis as a way to mediate energy balance [4-10]. As the hypothalamus regulates multiple functional outputs, it is likely that additional behaviors may be affected by postembryonic neurogenesis in this brain structure. Here, we have identified a progenitor population in the zebrafish hypothalamus that continuously generates neurons that express tyrosine hydroxylase 2 (th2). We develop and use novel transgenic tools to characterize the lineage of th2(+) cells and demonstrate that they are dopaminergic. Through genetic ablation and optogenetic activation, we then show that th2(+) neurons modulate the initiation of swimming behavior in zebrafish larvae. Finally, we find that the generation of new th2(+) neurons following ablation correlates with restoration of normal behavior. This work thus identifies for the first time a population of dopaminergic neurons that regulates motor behavior capable of functional recovery.

Transgenic FingRs for Live Mapping of Synaptic Dynamics in Genetically-Defined Neurons.

Tools for genetically-determined visualization of synaptic circuits and interactions are necessary to build connectomics of the vertebrate brain and to screen synaptic properties in neurological disease models. Here we develop a transgenic FingR (fibronectin intrabodies generated by mRNA display) technology for monitoring synapses in live zebrafish. We demonstrate FingR labeling of defined excitatory and inhibitory synapses, and show FingR applicability for dissecting synapse dynamics in normal and disease states. Using our system we show that chronic hypoxia, associated with neurological defects in preterm birth, affects dopaminergic neuron synapse number depending on the developmental timing of hypoxia.