Future healthcare systems will rely heavily on clinical decision support systems (CDSS) to improve the decision-making processes of clinicians. To explore the design of future CDSS, we developed a research-focused CDSS for the management of patients in the intensive care unit that leverages Internet of Things devices capable of collecting streaming physiologic data from ventilators and other medical devices. We then created machine learning models that could analyze the collected physiologic data to determine if the ventilator was delivering potentially harmful therapy and if a deadly respiratory condition, acute respiratory distress syndrome (ARDS), was present. We also present work to aggregate these models into a mobile application that can provide

Summary Background: As healthcare increasingly digitizes, streaming waveform data is being made available from an variety of sources, but there still remains a paucity of performant clinical decision support systems. For example, in the intensive care unit (ICU) existing automated alarm systems typically rely on simple thresholding that result in frequent false positives. Recurrent false positive alerts create distrust of alarm mechanisms that can be directly detrimental to patient health. To improve patient care in the ICU, we need alert systems that are both pervasive, and accurate so as to be informative and trusted by providers. Objective: We aimed to develop a machine learning-based classifier to detect abnormal waveform events using the use case of mechanical ventilation waveform analysis, and the detection of harmful forms of ventilation delivery to patients. We specifically focused on detecting injurious subtypes of patient-ventilator asynchrony (PVA). Methods: Using a dataset of breaths recorded from 35 different patients, we used machine learning to create computational models to automatically detect, and classify two types of injurious PVA, double trigger asynchrony (DTA), breath stacking asynchrony (BSA). We examined the use of synthetic minority over-sampling technique (SMOTE) to overcome class imbalance problems, varied methods for feature selection, and use of ensemble methods to optimize the performance of our model. Results: We created an ensemble classifier that is able to accurately detect DTA at a sensitivity/specificity of 0.960/0.975, BSA at sensitivity/specificity of 0.944/0.987, and non-PVA events at sensitivity/specificity of .967/.980. Conclusions: Our results suggest that it is possible to create a high-performing machine learning-based model for detecting PVA in mechanical ventilator waveform data in spite of both intra-patient, and inter-patient variability in waveform patterns, and the presence of clinical artifacts like cough and suction procedures. Our work highlights the importance of addressing class imbalance in clinical data sets, and the combined use of statistical methods and expert knowledge in feature selection.

Healthcare-specific analytic software is needed to process the large volumes of streaming physiologic waveform data increasingly available from life support devices such as mechanical ventilators. Detection of clinically relevant events from these data streams will advance understanding of critical illness, enable real-time clinical decision support, and improve both clinical outcomes and patient experience. We used mechanical ventilation waveform data (VWD) as a use case to address broader issues of data access and analysis including discrimination between true events and waveform artifacts. We developed an open source data acquisition platform to acquire VWD, and a modular, multi-algorithm analytic platform (ventMAP) to enable automated detection of off-target ventilation (OTV) delivery in critically-ill patients. We tested the hypothesis that use of artifact correction logic would improve the specificity of clinical event detection without compromising sensitivity. We showed that ventMAP could accurately detect harmful forms of OTV including excessive tidal volumes and common forms of patient-ventilator asynchrony, and that artifact correction significantly improved the specificity of event detection without decreasing sensitivity. Our multi-disciplinary approach has enabled automated analysis of high-volume streaming patient waveform data for clinical and translational research, and will advance the study and management of critically ill patients requiring mechanical ventilation.

Lack of access to high-frequency, high-volume patient-derived data, such as mechanical ventilator waveform data, has limited the secondary use of these data for research, quality improvement, and decision support. Existing methods for collecting these data are obtrusive, require high levels of technical expertise, and are often cost-prohibitive, limiting their use and scalability for research applications. We describe here the development of an unobtrusive, open-source, scalable, and user-friendly architecture for collecting, transmitting, and storing mechanical ventilator waveform data that is generalizable to other patient care devices. The system implements a software framework that automates and enforces end-to-end data collection and transmission. A web-based data management application facilitates nontechnical end users' abilities to manage data acquisition devices, mitigates data loss and misattribution, and automates data storage. Using this integrated system, we have been able to collect ventilator waveform data from >450 patients as part of an ongoing clinical study.

Algorithmic classifiers are crucial components of clinical decision support (CDS) systems needed to advance healthcare delivery. Robust CDS systems must be derived and validated via creation of multi-reviewer adjudicated gold standard datasets. Manual annotation of physiologic data such as mechanical ventilator waveform data (VWD) can be time-consuming, and lacks methodological consistency in dataset development. To address these issues, we have created a system for annotating and adjudicating VWD called the Annotation PipeLine (APL) to optimize VWD annotation by expert reviewers. APL combines visual assessment of waveform characteristics with metadata display, enabling inclusion of quantitative thresholds into annotation decisions by reviewers. APL also includes specific features for resolving multireviewer disagreements and generating gold standard data sets. APL's unique combination of methods and open source framework may accelerate the creation of CDS algorithms for ventilator management, and may serve as a model for future research into physiologic waveform annotation systems.