ObjectivesLung ultrasound (LUS) is a portable, low-cost respiratory imaging tool but is challenged by user dependence and lack of diagnostic specificity. It is unknown whether the advantages of LUS implementation could be paired with deep learning (DL) techniques to match or exceed human-level, diagnostic specificity among similar appearing, pathological LUS images.DesignA convolutional neural network (CNN) was trained on LUS images with B lines of different aetiologies. CNN diagnostic performance, as validated using a 10% data holdback set, was compared with surveyed LUS-competent physicians.SettingTwo tertiary Canadian hospitals.Participants612 LUS videos (121 381 frames) of B lines from 243 distinct patients with either (1) COVID-19 (COVID), non-COVID acute respiratory distress syndrome (NCOVID) or (3) hydrostatic pulmonary edema (HPE).ResultsThe trained CNN performance on the independent dataset showed an ability to discriminate between COVID (area under the receiver operating characteristic curve (AUC) 1.0), NCOVID (AUC 0.934) and HPE (AUC 1.0) pathologies. This was significantly better than physician ability (AUCs of 0.697, 0.704, 0.967 for the COVID, NCOVID and HPE classes, respectively), p<0.01.ConclusionsA DL model can distinguish similar appearing LUS pathology, including COVID-19, that cannot be distinguished by humans. The performance gap between humans and the model suggests that subvisible biomarkers within ultrasound images could exist and multicentre research is merited.

No abstract

ObjectivesLung ultrasound (LUS) is a portable, low cost respiratory imaging tool but is challenged by user dependence and lack of diagnostic specificity. It is unknown whether the advantages of LUS implementation could be paired with deep learning techniques to match or exceed human-level, diagnostic specificity among similar appearing, pathological LUS images.DesignA convolutional neural network was trained on LUS images with B lines of different etiologies. CNN diagnostic performance, as validated using a 10% data holdback set was compared to surveyed LUS-competent physicians.SettingTwo tertiary Canadian hospitals.Participants600 LUS videos (121,381 frames) of B lines from 243 distinct patients with either 1) COVID-19, Non-COVID acute respiratory distress syndrome (NCOVID) and 3) Hydrostatic pulmonary edema (HPE).ResultsThe trained CNN performance on the independent dataset showed an ability to discriminate between COVID (AUC 1.0), NCOVID (AUC 0.934) and HPE (AUC 1.0) pathologies. This was significantly better than physician ability (AUCs of 0.697, 0.704, 0.967 for the COVID, NCOVID and HPE classes, respectively), p < 0.01.ConclusionsA deep learning model can distinguish similar appearing LUS pathology, including COVID-19, that cannot be distinguished by humans. The performance gap between humans and the model suggests that subvisible biomarkers within ultrasound images could exist and multi-center research is merited.

Objective Sudden unexpected death in epilepsy (SUDEP) is a diagnosis of exclusion; the definition includes individuals with epilepsy who die suddenly without an identifiable toxicological or anatomical cause of death. Limited data suggest underidentification of SUDEP as the cause of death on death certificates. Here, we evaluate the autopsy‐reported cause of death in a population‐based cohort of SUDEP cases. Methods Case summaries of forensic autopsies conducted in Ontario, Canada between January 2014 and June 2016 were retrospectively screened using a language processing script for decedents with a history of epilepsy or seizures. After manual review for potential SUDEP cases, two neurologists independently examined the autopsy reports and classified deaths by Nashef criteria. Demographic characteristics and consideration by the forensic pathologist of the role of epilepsy, seizure, and SUDEP in death were summarized. Results One hundred and eight Definite, 34 Definite Plus, and 22 Possible SUDEP cases were identified. Seventy‐five percent of Definite/Definite Plus SUDEP cases identified by the neurologists were attributed to SUDEP, epilepsy, or seizure disorder in the autopsy report. There was a significant association between the proportion of cases listed in the autopsy report as SUDEP, epilepsy, or seizure disorder and neurologists' SUDEP classification (86% of Definite, 38% of Definite Plus, 0% of Possible). Age was significantly associated with SUDEP classification; Definite cases were younger than Definite Plus, which were younger than Possible SUDEP cases. Significance Most SUDEP cases identified by neurologists were classified concordantly by forensic pathologists in Ontario, Canada; however, concordance decreased with increased case complexity. Although the role of epilepsy/seizures was considered in most Definite/Definite Plus cases, this study highlights the need for autopsy report review of potential SUDEP cases in research studies and assessments of the public health burden of SUDEP. The relationship between age and SUDEP classification has important public health implications; SUDEP incidence may be underappreciated in older adults.

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.