Future frame prediction in videos is a promising avenue for unsupervised video representation learning. Video frames are naturally generated by the inherent pixel flows from preceding frames based on the appearance and motion dynamics in the video. However, existing methods focus on directly hallucinating pixel values, resulting in blurry predictions. In this paper, we develop a dual motion Generative Adversarial Net (GAN) architecture, which learns to explicitly enforce future-frame predictions to be consistent with the pixel-wise flows in the video through a duallearning mechanism. The primal future-frame prediction and dual future-flow prediction form a closed loop, generating informative feedback signals to each other for better video prediction. To make both synthesized future frames and flows indistinguishable from reality, a dual adversarial training method is proposed to ensure that the futureflow prediction is able to help infer realistic future-frames, while the future-frame prediction in turn leads to realistic optical flows. Our dual motion GAN also handles natural motion uncertainty in different pixel locations with a new probabilistic motion encoder, which is based on variational autoencoders. Extensive experiments demonstrate that the proposed dual motion GAN significantly outperforms stateof-the-art approaches on synthesizing new video frames and predicting future flows. Our model generalizes well across diverse visual scenes and shows superiority in unsupervised video representation learning.

No abstract

The severe acute respiratory syndrome coronavirus 2 (coronavirus disease 2019 [COVID-19]) pandemic has challenged medical systems and clinicians globally to unforeseen levels. Rapid spread of COVID-19 has forced clinicians to care for patients with a highly contagious disease without evidence-based guidelines. Using a virtual modified nominal group technique, the Pediatric Difficult Intubation Collaborative (PeDI-C), which currently includes 35 hospitals from 6 countries, generated consensus guidelines on airway management in pediatric anesthesia based on expert opinion and early data about the disease. PeDI-C identified overarching goals during care, including minimizing aerosolized respiratory secretions, minimizing the number of clinicians in contact with a patient, and recognizing that undiagnosed asymptomatic patients may shed the virus and infect health care workers. Recommendations include administering anxiolytic medications, intravenous anesthetic inductions, tracheal intubation using video laryngoscopes and cuffed tracheal tubes, use of in-line suction catheters, and modifying workflow to recover patients from anesthesia in the operating room. Importantly, PeDI-C recommends that anesthesiologists consider using appropriate personal protective equipment when performing aerosol-generating medical procedures in asymptomatic children, in addition to known or suspected children with COVID-19. Airway procedures should be done in negative pressure rooms when available. Adequate time should be allowed for operating room cleaning and air filtration between surgical cases. Research using rigorous study designs is urgently needed to inform safe practices during the COVID-19 pandemic. Until further information is available, PeDI-C advises that clinicians consider these guidelines to enhance the safety of health care workers during airway management when performing aerosol-generating medical procedures. These guidelines have been endorsed by the Society for Pediatric Anesthesia and the Canadian Pediatric Anesthesia Society.

No abstract

Assessing formal and informal workarounds employed by users should be part of routine organisational implementation strategies of major health information technology initiatives. Workarounds can create new risks and present new opportunities for improvement in system design and integration.