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.

Background. Cryptococcus infection is an opportunistic infection that occurs primarily among immunocompromised patients, and the morbidity and mortality of this infection is high if left unrecognized and untreated. There are no clinical or radiographic characteristics typical of cryptococcal pneumonia, and its clinical and radiological presentations often overlap with other diagnoses. Case Presentation. We present a case of a 25-year-old man from Ghana admitted for an altered mental state, weight loss, neck pain, fever, and photophobia. He was diagnosed with Cryptococcus neoformans meningitis by cerebrospinal fluid culture and with disseminated cryptococcal infection by a positive Cryptococcus blood test. Diffuse micronodular opacities were found in a miliary pattern in the upper portions of both lungs upon imaging, which suggested miliary tuberculosis; thus, the patient was started on antituberculosis therapy. The patient underwent flexible fiber optic bronchoscopy, and transbronchial biopsy of the right lung showed bronchopneumonia with fungal spores consistent with filamentous Cryptococcus neoformans, which grew in tissue culture of the right lung. Interferon-gamma release assay, Mycobacterium tuberculosis PCR, and acid-fast bacilli staining of the bronchoalveolar lavage were negative for the M. tuberculosis complex. Conclusion. The similarities in clinical and imaging findings among patients with acute immunodeficiency syndrome with coinfections make diagnoses difficult; thus image-guided biopsies are essential to confirm diagnoses.

A localized left atrial tamponade caused by left side pleural effusion is a rare finding that leads to hemodynamic instability. Here, we describe left atrial systolic and diastolic collapse resulting from left pleural effusion. An increase in intrapleural pressure by a pleural effusion can compress the pericardial space and lead to impaired cardiac filling and tamponade physiology. Here, we present a case of a 79-year old African American female who presented with shortness of breath and dry cough for a duration of one week. Chest radiograph and CT scan of the chest showed left pleural effusion. The echocardiogram revealed left atrial systolic and diastolic collapse due to pleural effusion, which triggered cardiac tamponade physiology. With the guidance of a bedside thoracic ultrasound, she underwent a diagnostic and therapeutic thoracentesis which resolved her symptoms. Repeat echocardiogram revealed resolution of the cardiac tamponade with no further indication of left atrial diastolic collapse. In conclusion, pleural effusions can cause tamponade physiology and can be resolved by thoracentesis. Early recognition by a bedside point-ofcare ultrasound may help provide prompt relief of tamponade.

No abstract