Novel approaches are needed to address elevated tobacco use among people with schizophrenia. This exploratory study examined the frequency, timing, and type of communication about tobacco-related content on Twitter among users who self-identify as having schizophrenia compared with users from the general population. Over a 200-day period from January to July 2016, Twitter users who self-identify as having a schizophrenia spectrum disorder (n = 203) and a randomly selected group of general population control users (n = 173) posted 1,544,122 tweets. Communication frequency did not differ between groups. Tweets containing tobacco-related keywords were extracted. Twitter users with schizophrenia posted significantly more tweets containing any tobacco-related terms (mean = 3.74; SD = 16.3) compared with control users (mean = 0.82; SD = 1.8). A significantly greater proportion of Twitter users with schizophrenia (45%; n = 92) posted tweets containing any tobacco terms compared with control users (30%; n = 52). Schizophrenia users showed significantly greater odds of tweeting about tobacco compared with control users (OR = 1.99; 95% CI 1.29-3.07). These findings suggest that online communication about tobacco may parallel real world trends of elevated tobacco use observed among people with schizophrenia. By showing that Twitter users who self-identify as having schizophrenia discuss tobacco-related content online, popular social media could inform smoking cessation efforts targeting this at-risk group.

Atrial fibrillation (AF) often escapes detection, given its frequent paroxysmal and asymptomatic presentation. Deep learning of transthoracic echocardiograms (TTEs), which have structural information, could help identify occult AF. We created a two-stage deep learning algorithm using a video-based convolutional neural network model that (1) distinguished whether TTEs were in sinus rhythm or AF and then (2) predicted which of the TTEs in sinus rhythm were in patients who had experienced AF within 90 days. Our model, trained on 111,319 TTE videos, distinguished TTEs in AF from those in sinus rhythm with high accuracy in a held-out test cohort (AUC 0.96 (0.95-0.96), AUPRC 0.91 (0.90-0.92)). Among TTEs in sinus rhythm, the model predicted the presence of concurrent paroxysmal AF (AUC 0.74 (0.71-0.77), AUPRC 0.19 (0.16-0.23)). Model discrimination remained similar in an external cohort of 10,203 TTEs (AUC of 0.69 (0.67-0.70), AUPRC 0.34 (0.31-0.36)). Performance held across patients who were women (AUC 0.76 (0.72-0.81)), older than 65 years (0.73 (0.69-0.76)), or had a CHA2DS2VASc ≥2 (0.73 (0.79-0.77)). The model performed better than using clinical risk factors (AUC 0.64 (0.62-0.67)), TTE measurements (0.64 (0.62-0.67)), left atrial size (0.63 (0.62-0.64)), or CHA2DS2VASc (0.61 (0.60-0.62)). An ensemble model in a cohort subset combining the TTE model with an electrocardiogram (ECGs) deep learning model performed better than using the ECG model alone (AUC 0.81 vs. 0.79, p = 0.01). Deep learning using TTEs can predict patients with active or occult AF and could be used for opportunistic AF screening that could lead to earlier treatment.

Artificial intelligence (AI) has been developed for echocardiography^1-3, although it has not yet been tested with blinding and randomization. Here we designed a blinded, randomized non-inferiority clinical trial (ClinicalTrials.gov ID: NCT05140642; no outside funding) of AI versus sonographer initial assessment of left ventricular ejection fraction (LVEF) to evaluate the impact of AI in the interpretation workflow. The primary end point was the change in the LVEF between initial AI or sonographer assessment and final cardiologist assessment, evaluated by the proportion of studies with substantial change (more than 5% change). From 3,769 echocardiographic studies screened, 274 studies were excluded owing to poor image quality. The proportion of studies substantially changed was 16.8% in the AI group and 27.2% in the sonographer group (difference of -10.4%, 95% confidence interval: -13.2% to -7.7%, P < 0.001 for non-inferiority, P < 0.001 for superiority). The mean absolute difference between final cardiologist assessment and independent previous cardiologist assessment was 6.29% in the AI group and 7.23% in the sonographer group (difference of -0.96%, 95% confidence interval: -1.34% to -0.54%, P < 0.001 for superiority). The AI-guided workflow saved time for both sonographers and cardiologists, and cardiologists were not able to distinguish between the initial assessments by AI versus the sonographer (blinding index of 0.088). For patients undergoing echocardiographic quantification of cardiac function, initial assessment of LVEF by AI was non-inferior to assessment by sonographers.

Apolipoprotein AV (APOA5) lowers plasma triglyceride (TG) levels by binding to the angiopoietin-like protein 3/8 complex (ANGPTL3/8) and suppressing its capacity to inhibit lipoprotein lipase (LPL) catalytic activity and its ability to detach LPL from binding sites within capillaries. However, the sequences in APOA5 that are required for suppressing ANGPTL3/8 activity have never been defined. A clue to the identity of those sequences was the presence of severe hypertriglyceridemia in two patients harboring an APOA5 mutation that truncates APOA5 by 35 residues (APOA5Δ35). We found that wild-type (WT) human APOA5, but not APOA5Δ35, suppressed ANGPTL3/8s ability to inhibit LPL catalytic activity. To pursue that finding, we prepared a mutant mouse APOA5 protein lacking 40 C-terminal amino acids (APOA5Δ40). Mouse WT-APOA5, but not APOA5Δ40, suppressed ANGPTL3/8s capacity to inhibit LPL catalytic activity and sharply reduced plasma TG levels in mice. WT-APOA5, but not APOA5Δ40, increased intracapillary LPL levels and reduced plasma TG levels in Apoa5-/- mice (where TG levels are high and intravascular LPL levels are low). Also, WT-APOA5, but not APOA5Δ40, blocked the ability of ANGPTL3/8 to detach LPL from cultured cells. Finally, an antibody against a synthetic peptide corresponding to the last 26 amino acids of mouse APOA5 reduced intracapillary LPL levels and increased plasma TG levels in WT mice. We conclude that C-terminal sequences in APOA5 are crucial for suppressing ANGPTL3/8 activity in vitro and for regulating intracapillary LPL levels and plasma TG levels in vivo.