Heart failure alters diagnostic yield for pulmonary embolism in patients undergoing computed tomography pulmonary angiogram.

INTRODUCTION: There is limited evidence regarding the effects of a pre-existing heart failure (HF) on the diagnostic yield of pulmonary embolism (PE) evaluation in the Emergency Department (ED). METHODS: Electronic medical record of consecutive adults who underwent a computed tomography pulmonary angiogram (CTPA) in the ED at Loma Linda University Medical Center between June 1, 2019 and March 25, 2022 were reviewed. Repeat studies for the same patient and patients with unspecified HF diagnoses or isolated right ventricular HF were excluded. Key demographics, lab values and vital signs, relevant medications were collected. Primary outcome was the incidence of PE on CTPA compared between patients with and without pre-existing HF. RESULTS: A total of 2846 patients were included in the study (602 patients with HF and 2244 without). In total cohort, 11.7% (n = 334) of patients had PE found on CTPA. The incidence of PE on CTPA was lower among patients with a history of HF than patients without a history of HF (12.5% vs 9%). A history of pre-existing HF was associated with a lower odds ratio for a positive PE study (OR 0.13, 95%CI: 0.03-0.57) in multivariable analyses. CONCLUSIONS: In this study, we observed that the incidence of PE among patients who undergo CTPA was lower among patients with pre-existing HF compared to those without. Further studies should determine if HF is an important mitigating factor when risk stratifying patients for PE.

Generalizable fully automated multi-label segmentation of four-chamber view echocardiograms based on deep convolutional adversarial networks.

A major issue in translation of the artificial intelligence platforms for automatic segmentation of echocardiograms to clinics is their generalizability. The present study introduces and verifies a novel generalizable and efficient fully automatic multi-label segmentation method for four-chamber view echocardiograms based on deep fully convolutional networks (FCNs) and adversarial training. For the first time, we used generative adversarial networks for pixel classification training, a novel method in machine learning not currently used for cardiac imaging, to overcome the generalization problem. The method's performance was validated against manual segmentations as the ground-truth. Furthermore, to verify our method's generalizability in comparison with other existing techniques, we compared our method's performance with a state-of-the-art method on our dataset in addition to an independent dataset of 450 patients from the CAMUS (cardiac acquisitions for multi-structure ultrasound segmentation) challenge. On our test dataset, automatic segmentation of all four chambers achieved a dice metric of 92.1%, 86.3%, 89.6% and 91.4% for LV, RV, LA and RA, respectively. LV volumes' correlation between automatic and manual segmentation were 0.94 and 0.93 for end-diastolic volume and end-systolic volume, respectively. Excellent agreement with chambers' reference contours and significant improvement over previous FCN-based methods suggest that generative adversarial networks for pixel classification training can effectively design generalizable fully automatic FCN-based networks for four-chamber segmentation of echocardiograms even with limited number of training data.