Passive Nocturnal Physiologic Monitoring Enables Early Detection of Exacerbations in Children with Asthma. A Proof-of-Concept Study.

RATIONALE: Asthma management depends on prompt identification of symptoms, which challenges both patients and providers. In asthma, a misapprehension of health between exacerbations can compromise compliance. Thus, there is a need for a tool that permits objective longitudinal monitoring without increasing the burden of patient compliance. OBJECTIVES: We sought to determine whether changes in nocturnal physiology are associated with asthma symptoms in pediatric patients. METHODS: Using a contactless bed sensor, nocturnal heart rate (HR), respiratory rate, relative stroke volume, and movement in children with asthma 5-18 years of age (n = 16) were recorded. Asthma symptoms and asthma control test (ACT) score were reported every 2 weeks. Random forest model was used to identify physiologic parameters associated with asthma symptoms. Elastic net regression was used to identify variables associated with ACT score. MEASUREMENTS AND MAIN RESULTS: The model on the full cohort performed with sensitivity of 47.2%, specificity of 96.3%, and accuracy of 87.4%; HR and respiratory parameters were the most important variables in this model. The model predicted asthma symptoms 35% of the time on the day before perception of symptoms, and 100% of the time for a select subject for which the model performed with greater sensitivity. Multivariable and bivariable analyses demonstrated significant association between HR and respiratory rate parameters and ACT score. CONCLUSIONS: Nocturnal physiologic changes correlate with asthma symptoms, supporting the notion that nocturnal physiologic monitoring represents an objective diagnostic tool capable of longitudinally assessing disease control and predicting asthma exacerbations in children with asthma at home.

Automatic detection and segmentation of robot-assisted surgical motions.

Robotic surgical systems such as Intuitive Surgical's da Vinci system provide a rich source of motion and video data from surgical procedures. In principle, this data can be used to evaluate surgical skill, provide surgical training feedback, or document essential aspects of a procedure. If processed online, the data can be used to provide context-specific information or motion enhancements to the surgeon. However, in every case, the key step is to relate recorded motion data to a model of the procedure being performed. This paper examines our progress at developing techniques for "parsing" raw motion data from a surgical task into a labelled sequence of surgical gestures. Our current techniques have achieved >90% fully automated recognition rates on 15 datasets.