Aims Clinical scoring systems for pulmonary embolism (PE) screening have low specificity and contribute to CT pulmonary angiogram (CTPA) overuse. We assessed whether deep learning models using an existing and routinely collected data modality, electrocardiogram (ECG) waveforms, can increase specificity for PE detection. Methods and Results We create a retrospective cohort of 21,183 patients at moderate- to high-suspicion of PE and associate 23,793 CTPAs (10.0% PE-positive) with 320,746 ECGs and encounter-level clinical data (demographics, comorbidities, vital signs, and labs). We develop three machine learning models to predict PE likelihood: an ECG model using only ECG waveform data, an EHR model using tabular clinical data, and a Fusion model integrating clinical data and an embedded representation of the ECG waveform. We find that a Fusion model (area under receiver-operating characteristic [AUROC] 0.81 ± 0.01) outperforms both the ECG model (AUROC 0.59 ± 0.01) and EHR model (AUROC 0.65 ± 0.01). On a sample of 100 patients from the test set, the Fusion model also achieves greater specificity (0.18) and performance (AUROC 0.84 ± 0.01) than four commonly evaluated clinical scores: Wells' Criteria, Revised Geneva Score, Pulmonary Embolism Rule-Out Criteria, and 4-Level Pulmonary Embolism Clinical Probability Score (AUROC 0.50-0.58, specificity 0.00-0.05). The model is superior to these scores on feature sensitivity analyses (AUROC 0.66 to 0.84) and achieves comparable performance across sex (AUROC 0.81) and racial/ethnic (AUROC 0.77 to 0.84) subgroups. Conclusion Synergistic deep learning of electrocardiogram waveforms with traditional clinical variables can increase the specificity of PE detection in patients at least at moderate suspicion for PE.

No abstract

No abstract

Background Pre-(PRE) and post-ejection (POE) velocities by mitral annular tissue Doppler (TD) are biphasic and may be related to myocardial deformations. We investigated the predominance and concordance of TD-PRE and POE velocities and their effect on myocardial functions in controls and in heart failure (HF) patients. Methods Retrospectively, 84 HF patients [57.6 years, 28(33%) females, NYHA: 2.3 ± 0.6, EF: 55 ± 15%, 52(62%) preserved EF, and 32(38%) reduced EF], 42 normal young controls, and 26 asymptomatic age matched controls were included. Echocardiography was done and from mitral annular tissue Doppler recordings, the biphasic PRE and POE velocity signals were identified and compared between groups. Results While controls had almost always predominantly positive PRE and negative POE, HF had more negative PRE and positive POE. Moreover, almost all controls exhibited normal concordance (positive PRE and negative POE). HF exhibited more abnormal concordance which was significantly associated with worse NYHA, and parameters of diastolic and systolic functions. Opposite PRE and POE velocities correlated significantly in all groups (PREp vs POEn: young:r = 0.52, p < 0.001, age controls:r = 0.79, p < 0.001, HFpEF: r = 0.56, p < 0.001, HFrEF: r = 0.42, p = 0.018; PREn vs POEp: young: r = 0.25,p = 0.1, age controls: r = 0.42, p = 0.04, HFpEF: r = 0.43, p = 0.004, HFrEF: r = 0.61, p < 0.001) and the ratios PRE-P/N and POE-N/P correlated significantly with E/e’ in HF only. Conclusions In physiological state, TD signals are predominantly positive during PRE and negative during POE. Opposite PRE and POE velocities corelate, representing the PRE-generation and POE-reversal of shortening–stretch relationships, the attenuation of which in HF may be related to elevated LV filling pressures. In HF, partially or completely reversed concordance of PRE and POE is associated with progressive worsening of clinical and hemodynamic profiles.

No abstract

No abstract

No abstract

Introduction: Pre-(PRE) and post-ejection (POE) velocities by mitral annular tissue Doppler (TD) are biphasic and may be related to myocardial deformations. We investigated the predominance and concordance of TD-PRE and POE velocities and their effect on myocardial functions in controls and in HF patients. Methods: Retrospectively, 84 HF patients [57.6±8 years, 28(33%) females, NYHA: 2.3±0.6, EF: 55±15%, 52(62%) preserved EF, and 32(38%) reduced EF], 42 normal young controls [27±4 years, 5(12%) females], and 26 asymptomatic age matched controls without cardiac abnormalities [53±12 years, 28(65%) females] were included. Results: While controls had almost always predominantly positive PRE and negative POE, HF had more negative PRE (Controls: 2%, HFpEF: 10%, HFrEF: 35%, p<0.001) and positive POE (Controls: 4%, HFpEF: 33%, HFrEF: 74%, p<0.001). Moreover, almost all controls exhibited normal concordance (positive PRE and negative POE). HF exhibited more abnormal concordance (Figure 1A,C) which was significantly associated with worse NYHA, and parameters of diastolic and systolic functions. Opposite PRE and POE velocities correlated significantly in all groups (PREp vs POEn: young:r=0.52, p<0.001, age controls:r=0.79, p<0.001, HFpEF: r=0.56, p<0.001, HFrEF: r=0.42, p=0.018; PREn vs POEp: young: r=0.25,p=0.1, age controls: r=0.42, p=0.04, HFpEF: r=0.43, p=0.004, HFrEF: r=0.61, p<0.001) and the ratios PRE-P/N and POE-N/P correlated significantly with E/e’ in HF only (Figure 1B). Conclusions: In physiological state, TD signals are predominantly positive during PRE and negative during POE. Opposite PRE and POE velocities corelate, representing the PRE-generation and POE-reversal of shortening-stretch relationships, the attenuation of which in HF may be related to elevated LV filling pressures. In HF, partially or completely reversed concordance of PRE and POE is associated with progressive worsening of clinical and hemodynamic profiles.