Lung ultrasound (LUS) is an accurate thoracic imaging technique distinguished by its handheld size, low-cost, and lack of radiation. User dependence and poor access to training have limited the impact and dissemination of LUS outside of acute care hospital environments. Automated interpretation of LUS using deep learning can overcome these barriers by increasing accuracy while allowing point-of-care use by non-experts. In this multicenter study, we seek to automate the clinically vital distinction between A line (normal parenchyma) and B line (abnormal parenchyma) on LUS by training a customized neural network using 272,891 labelled LUS images. After external validation on 23,393 frames, pragmatic clinical application at the clip level was performed on 1162 videos. The trained classifier demonstrated an area under the receiver operating curve (AUC) of 0.96 (±0.02) through 10-fold cross-validation on local frames and an AUC of 0.93 on the external validation dataset. Clip-level inference yielded sensitivities and specificities of 90% and 92% (local) and 83% and 82% (external), respectively, for detecting the B line pattern. This study demonstrates accurate deep-learning-enabled LUS interpretation between normal and abnormal lung parenchyma on ultrasound frames while rendering diagnostically important sensitivity and specificity at the video clip level.

Background: Annotating large medical imaging datasets is an arduous and expensive task, especially when the datasets in question are not organized according to deep learning goals. Here, we propose a method that exploits the hierarchical organization of annotating tasks to optimize efficiency. Methods: We trained a machine learning model to accurately distinguish between one of two classes of lung ultrasound (LUS) views using 2908 clips from a larger dataset. Partitioning the remaining dataset by view would reduce downstream labelling efforts by enabling annotators to focus on annotating pathological features specific to each view. Results: In a sample view-specific annotation task, we found that automatically partitioning a 780-clip dataset by view saved 42 min of manual annotation time and resulted in $55\pm 6$ additional relevant labels per hour. Conclusions: Automatic partitioning of a LUS dataset by view significantly increases annotator efficiency, resulting in higher throughput relevant to the annotating task at hand. The strategy described in this work can be applied to other hierarchical annotation schemes.

Background: Currently, the early pre-clinical detection of left ventricular (LV) dysfunction is difficult as biomarkers are not specific for the cardiomyopathic process. The underlying molecular mechanisms leading to heart failure remain elusive, highlighting the need for identification of cardiac-specific markers. The growth hormone secretagogue receptor (GHSR) and its ligand ghrelin are present in cardiac tissue and are known to contribute to myocardial energetics. Here, we examined tissue ghrelin-GHSR levels as specific markers of cardiac dysfunction in patients who underwent cardiac transplantation. Methods and Results: Samples of cardiac tissue were obtained from 10 cardiac transplant patients at the time of organ harvesting, and during serial post-transplant biopsies. Quantitative fluorescence microscopy using a novel fluorescent ghrelin analog was used to measure levels of GHSR, and immunofluorescence was used to measure levels of ghrelin, b-type natriuretic peptide (BNP) and tissue markers of cardiomyocyte contractility and growth. GHSR and ghrelin expression levels were highly variable in the explanted heart, less in the grafted heart biopsies. GHSR and ghrelin were strongly positively correlated, and both markers were negatively correlated with LV ejection fraction. Ghrelin had stronger positive correlations than BNP with the signaling markers for contractility and growth. Conclusions: These data suggest that GHSR-ghrelin have potential use as an integrated marker of cardiac dysfunction. Interestingly, tissue ghrelin appeared to be a more sensitive indicator than BNP to the biochemical processes that are characteristic of heart failure. This work allows for further use of ghrelin-GHSR to interrogate cardiac-specific biochemical mechanisms in pre-clinical stages of HF. Precis: This study shows the relationships between GHSR, ghrelin, and signaling molecules with relation to heart function in human heart failure with tissue from diseased heart and healthy heart biopsies.