QtWS17 - Building a Human Machine Interface for the medical IoT, Jereme Givens-Lamothe, DocBox
Healthcare is held back by lack of access to quality clinical data. Access to quality clinical data is difficult due to lack of interoperability between medical devices. Periodic EMR documentation is not adequate for fueling research, Artificial Intelligence (AI), or Machine Learning (ML). The ASTM F2761 standard describes the Integrated Clinical Environment (ICE), an infrastructure for supporting interoperability between medical devices and user access facilitated through apps. Building on the ICE architecture, DocBox has created a platform that enables the development and use of clinical apps. These clinical apps support advances in operational and clinical automation, and improvements in patient safety. DocBox has chosen Qt as the application framework to build the bedside touchscreen HMI (Human Machine Interface), and clinical apps built on the platform. We wanted to create an HMI with a modern, intuitive UI. Qt Quick allows us to rapidly build apps made up of re-usable UI components, while taking advantage of gestures unique to a touch screen interface. This talk will demonstrate how we’ve repurposed the Qt Application Manager from the Automotive Suite. The DocBox at the patient’s bedside can install different apps suited to the task at hand. This enables the DocBox to scale to a large number of use cases in a clinical context, a general purpose device for clinician use. This talk will also demonstrate how we’ve leveraged components like Qt Charts and the Qt Virtual Keyboard to create a clinical app. Data from connected medical devices is automatically added to a patient’s flowsheet, and the remainder completed with manual entry using the Qt Virtual Keyboard. We have heavily customized the Qt Virtual Keyboard to allow convenient entry of measurements. Qt Charts is leveraged to graphically display an 18 hour history of patient data in a line chart, as well as system events. User interaction is possible through tapping specific events to reveal detailed information, and pinch / swipe gestures to explore patient data on a time series. Behind these clinical applications is DICES, a patient-centric data model. Leveraging DDS (Data Distribution Service), data from connected medical devices, hospital information systems, and apps we maintain a living model of all information pertinent to the patient’s care. DDS is a data-centric, publish / subscribe, powerful middleware for the Industrial Internet of Things. Use cases extend beyond patient care, and into the realms of operations and logistics, such as easy inventory management of medical devices. This talk will demonstrate how we’ve integrated RTI DDS Connext into our apps. Looking to the future, apps will be built to assist the clinician in caring for patients, making intelligent decisions, and documenting their progress. The data is what’s important; how to discern quality data, how to surface and present it to the user. The DocBox ICE platform, powered by the Qt Application framework, will enable data driven healthcare.