Respiration rate (RR) is an important vital sign for clinical monitoring of hospitalized patients, with changes in RR being strongly tied to changes in clinical status leading to adverse events. Human labels for RR, based on counting breaths, are known to be inaccurate and time consuming for medical staff. Automated monitoring of RR is in place for some patients, typically those in intensive care units (ICUs), but is absent for the majority of inpatients on standard medical wards who are still at risk for clinical deterioration. This work trains a neural network (NN) to label RR from electrocardiogram (ECG) telemetry waveforms, which like many biosignals, carry multiple signs of respiratory variation. The NN shows high accuracy on multiple validation sets (internal and external, same and different sources of RR labels), with mean absolute errors less than 1.78 breaths per minute (bpm) in the worst case. The clinical utility of such a technology is exemplified by performing a retrospective analysis of two patient cohorts that suffered adverse events including respiratory failure, showing that continuous RR monitoring could reveal dynamics that strongly tracked with intubation events. This work exemplifies the method of combining pre-existing telemetry monitoring systems and artificial intelligence (AI) to provide accurate, automated and scalable patient monitoring, all of which builds towards an AI-based hospital-wide early warning system (EWS).

We present a foundation model-derived method to identify highly informative tokens and events in electronic health records. Our approach considers incoming data in the entire context of a patient's hospitalization and so can flag anomalous events that rule-based approaches would consider within a normal range. We demonstrate that the events our model flags are significant for predicting downstream patient outcomes and that a fraction of events identified as carrying little information can safely be dropped. Additionally, we show how informativeness can help interpret the predictions of prognostic models trained on foundation model-derived representations.

Sepsis-induced acute respiratory failure (ARF) is a serious complication with a poor prognosis. This paper presents a deep representation learningbased phenotyping method to identify distinct groups of clinical trajectories of septic patients with ARF. For this retrospective study, we created a dataset from electronic medical records (EMR) consisting of data from sepsis patients admitted to medical intensive care units who required at least 24 hours of invasive mechanical ventilation at a quarternary care academic hospital in southeast USA for the years 2016-2021. A total of N=3349 patient encounters were included in this study. Clustering Representation Learning on Incomplete Time Series Data (CRLI) algorithm was applied to a parsimonious set of EMR variables in this data set. To validate the optimal number of clusters, the K-means algorithm was used in conjunction with dynamic time warping. Our model yielded four distinct patient phenotypes that were characterized as liver dysfunction/heterogeneous, hypercapnia, hypoxemia, and multiple organ dysfunction syndrome by a critical care expert. A Kaplan-Meier analysis to compare the 28-day mortality trends exhibited significant differences (p < 0.005) between the four phenotypes. The study demonstrates the utility of our deep representation learning-based approach in unraveling phenotypes that reflect the heterogeneity in sepsis-induced ARF in terms of different mortality outcomes and severity. These phenotypes might reveal important clinical insights into an effective prognosis and tailored treatment strategies.

The development of respiratory failure is common among patients in intensive care units (ICU). Large data quantities from ICU patient monitoring systems make timely and comprehensive analysis by clinicians difficult but are ideal for automatic processing by machine learning algorithms. Early prediction of respiratory system failure could alert clinicians to patients at risk of respiratory failure and allow for early patient reassessment and treatment adjustment. We propose an early warning system that predicts moderate/severe respiratory failure up to 8 hours in advance. Our system was trained on HiRID-II, a data-set containing more than 60,000 admissions to a tertiary care ICU. An alarm is typically triggered several hours before the beginning of respiratory failure. Our system outperforms a clinical baseline mimicking traditional clinical decision-making based on pulse-oximetric oxygen saturation and the fraction of inspired oxygen. To provide model introspection and diagnostics, we developed an easy-to-use web browser-based system to explore model input data and predictions visually.

Although recent multi-task learning methods have shown to be effective in improving the generalization of deep neural networks, they should be used with caution for safety-critical applications, such as clinical risk prediction. This is because even if they achieve improved task-average performance, they may still yield degraded performance on individual tasks, which may be critical (e.g., prediction of mortality risk). Existing asymmetric multi-task learning methods tackle this negative transfer problem by performing knowledge transfer from tasks with low loss to tasks with high loss. However, using loss as a measure of reliability is risky since it could be a result of overfitting. In the case of time-series prediction tasks, knowledge learned for one task (e.g., predicting the sepsis onset) at a specific timestep may be useful for learning another task (e.g., prediction of mortality) at a later timestep, but lack of loss at each timestep makes it difficult to measure the reliability at each timestep. To capture such dynamically changing asymmetric relationships between tasks in time-series data, we propose a novel temporal asymmetric multi-task learning model that performs knowledge transfer from certain tasks/timesteps to relevant uncertain tasks, based on feature-level uncertainty. We validate our model on multiple clinical risk prediction tasks against various deep learning models for time-series prediction, which our model significantly outperforms, without any sign of negative transfer. Further qualitative analysis of learned knowledge graphs by clinicians shows that they are helpful in analyzing the predictions of the model.

Logistic regression (LR) is widely used in clinical prediction because it is simple to deploy and easy to interpret. Nevertheless, being a linear model, LR has limited expressive capability and often has unsatisfactory performance. Generalized additive models (GAMs) extend the linear model with transformations of input features, though feature interaction is not allowed for all GAM variants. In this paper, we propose a factored generalized additive model (F-GAM) to preserve the model interpretability for targeted features while allowing a rich model for interaction with features fixed within the individual. We evaluate F-GAM on prediction of two targets, postoperative acute kidney injury and acute respiratory failure, from a single-center database. We find superior model performance of F-GAM in terms of AUPRC and AUROC compared to several other GAM implementations, random forests, support vector machine, and a deep neural network. We find that the model interpretability is good with results with high face validity.

There is no shortage of corporate AI projects underway and most will fail because companies have not laid the proper foundation in the form of pilot projects or proof of concept approaches needed to scale AI efforts. That's the central conclusion of a new survey of about 80 companies jumping on the AI bandwagon but neglecting to look before they leap. "While enterprises have high expectations of the impact of [intelligent automation], they are not ready to implement it from the top down and at scale," concludes technology consultant KPMG in a report released this week. Among the reasons is a fundamental lack of understanding that the corporate implementation of AI referred to as intelligent automation is "about changing business processes, and then restructuring the organization around those new processes now driven by technologies that didn't exist before." Another part of the problem is unrealistic expectations by early adopters of AI who have done little planning to scale out deployment efforts.

It's been a big year in healthcare technology. Healthcare Analytics News reached out to experts across our 8 coverage areas to determine which companies, people, and projects made the biggest waves. The winners of 2017 ushered in advances that have turned heads, resulted in measurable improvements, and given reason to believe that this high-speed sector is not built on hype alone. Big Data: Montefiore Medical Center Numbers can save lives. The traditional relational database in place at Montefiore Medical Center, Albert Einstein College of Medicine, in Bronx, New York, has been dredged out and filled in with the center's innovative semantic data lake project, to plumb the depths of predictive analytics.