Search

Abstract This study aimed to develop a deep learning model to assess the quality of fetal echocardiography and to perform prospective clinical validation. The model was trained on data from the 18–22-week anomaly scan conducted in seven hospitals from 2008 to 2018. Prospective validation involved 100 patients from two hospitals. A total of 5363 images from 2551 pregnancies were used for training and validation. The model's segmentation accuracy depended on image quality measured by a quality score (QS). It achieved an overall average accuracy of 0.91 (SD 0.09) across the test set, with images having above-average QS scoring 0.97 (SD 0.03). During prospective validation of 192 images, clinicians rated 44.8% (SD 9.8) of images as equal in quality, 18.69% (SD 5.7) favoring auto-captured images and 36.51% (SD 9.0) preferring manually captured ones. Images with above average QS showed better agreement on segmentations (p

Proper specimen collection is the most important step to ensure accurate testing for the coronavirus disease 2019 (COVID-19) and other infectious diseases. Assessment of healthcare workers’ upper respiratory tract specimen collection skills is needed to ensure samples of high-quality clinical specimens for COVID-19 testing. This study explored the validity evidence for a theoretical MCQ-test and checklists developed for nasopharyngeal (NPS) and oropharyngeal (OPS) specimen collection skills assessment. We found good inter-item reliability (Cronbach’s alpha = 0.76) for the items of the MCQ-test and high inter-rater reliability using the checklist for the assessment of OPS and NPS skills on 0.86 and 0.87, respectively. The MCQ scores were significantly different between experts (mean 98%) and novices (mean 66%), p < 0.001, and a pass/fail score of 91% was established. We found a significant discrimination between checklist scores of experts (mean 95% score for OPS and 89% for NPS) and novices (mean 50% score for OPS and 36% for NPS), p < 0.001, and a pass/fail score was established of 76% for OPS and 61% for NPS. Further, the results also demonstrated that a group of non-healthcare educated workers can perform upper respiratory tract specimen collection comparably to experts after a short and focused simulation-based training session. This study, therefore, provides validity evidence for the use of a theoretical and practical test for upper respiratory specimens’ collection skills that can be used for competency-based training of the workers in the COVID-19 test centers.

Billfishes are considered to be among the fastest swimmers in the oceans. Previous studies have estimated maximum speed of sailfish and black marlin at around 35 m s−1 but theoretical work on cavitation predicts that such extreme speed is unlikely. Here we investigated maximum speed of sailfish, and three other large marine pelagic predatory fish species, by measuring the twitch contraction time of anaerobic swimming muscle. The highest estimated maximum swimming speeds were found in sailfish (8.3±1.4 m s−1), followed by barracuda (6.2±1.0 m s−1), little tunny (5.6±0.2 m s−1) and dorado (4.0±0.9 m s−1); although size-corrected performance was highest in little tunny and lowest in sailfish. Contrary to previously reported estimates, our results suggest that sailfish are incapable of exceeding swimming speeds of 10-15 m s−1, which corresponds to the speed at which cavitation is predicted to occur, with destructive consequences for fin tissues.

Fast-starts are brief accelerations commonly observed in fish within the context of predator–prey interactions. In typical C-start escape responses, fish react to a threatening stimulus by bending their body into a C-shape during the first muscle contraction (i.e. stage 1) which provides a sudden acceleration away from the stimulus. Recently, similar C-starts have been recorded in fish aiming at a prey. Little is known about C-starts outside the context of predator–prey interactions, though recent work has shown that escape response can also be induced by high temperature. Here, we test the hypothesis that air-breathing fish may use C-starts in the context of gulping air at the surface. Hoplosternum littorale is an air-breathing freshwater catfish found in South America. Field video observations reveal that their air-breathing behaviour consists of air-gulping at the surface, followed by a fast turn which re-directs the fish towards the bottom. Using high-speed video in the laboratory, we compared the kinematics of the turn immediately following air-gulping performed by H. littorale in normoxia with those of mechanically-triggered C-start escape responses and with routine (i.e. spontaneous) turns. Our results show that air-breathing events overlap considerably with escape responses with a large stage 1 angle in terms of turning rates, distance covered and the relationship between these rates. Therefore, these two behaviours can be considered kinematically comparable, suggesting that air-breathing in this species is followed by escape-like C-start motions, presumably to minimise time at the surface and exposure to avian predators. These findings show that C-starts can occur in a variety of contexts in which fish may need to get away from areas of potential danger.