Triage and monitoring of COVID-19 patients in intensive care using unsupervised machine learning

Abstract

Background: We designed an algorithm to assess COVID-19 patients severity and dynamic intubation needs and predict their length of stay using the breathing frequency (BF) and oxygen saturation (SpO₂) signals.

Methods: We recorded the BF and SpO₂ signals for confirmed COVID-19 patients admitted to the ICU of a teaching hospital during both the first and subsequent outbreaks of the pandemic in France. An unsupervised machine-learning algorithm (the Gaussian mixture model) was applied to the patients' data for clustering. The algorithm's robustness was ensured by comparing its results against actual intubation rates. We predicted intubation rates using the algorithm every hour, thus conducting a severity evaluation. We designed a S₂₄ severity score that represented the patient's severity over the previous 24 h; the validity of MS₂₄, the maximum S₂₄ score, was checked against rates of intubation risk and prolonged ICU stay.

Results: Our sample included 279 patients. . The unsupervised clustering had an accuracy rate of 87.8% for intubation recognition (AUC = 0.94, True Positive Rate 86.5%, true Negative Rate 90.9%). The S₂₄ score of intubated patients was significantly higher than that of non-intubated patients at 48 h before intubation. The MS₂₄ score allowed for the distinguishing between three severity levels with an increased risk of intubation: green (3.4%), orange (37%), and red (77%). A MS₂₄ score over 40 was highly predictive of an ICU stay greater than 5 days at an accuracy rate of 81.0% (AUC = 0.87).

Conclusions: Our algorithm uses simple signals and seems to efficiently visualize the patients' respiratory situations, meaning that it has the potential to assist staffs' in decision-making. Additionally, real-time computation is easy to implement.

Keywords: Artificial intelligence; COVID-19; Intubation; Monitoring; Prediction.

Graphical abstract

Fig. 1

General Principles: A-The SpO₂ and BF signals are used to calculate a state vector representing the patient's respiratory status every hour. This state vector is averaged over the last 4 h before intubation. We set a fictitious date of intubation for patients who were not intubated, which is 91 h after admission (91 h the average date of intubation for patients). The state vector is also averaged over the last 4 h prior to this fictitious intubation date. B- All state vectors of the 279 patients are provided to an unsupervised clustering algorithm (GMM). We compared the performance of this algorithm to the actual classification. C- The algorithm calculates the probability of intubation in 4 h every hour. However, this prediction is often distorted by brief periods of instability. This probability is therefore averaged over 24 h.

Fig. 2

Study flow chart, including patients from ongoing pandemic, March 2020 to September 2021.

Fig. 3

Unsupervised clustering performances of Model 4 as listed in Table 2. A – The confusion matrix shows an accuracy of 87.8%. Class 0 is non-intubated and class 1 intubated. Numbers are actual number of patients in each predicted class according to the ground true class. Green percentages (resp. red) are true rate (positive or negative) or positive predictive values (resp.: false and negative predictive values). B - ROC curve had an AUC = 0.94.

Fig. 4

The cohort dynamic changes of the S₂₄ score during the 80 h prior to intubation. Patients from the intubated group are in red. The plain thick line is the mean and the fine lines correspond to the 25%–75% confidence interval. The green area represents the non-intubated group; the dashed thick line is the mean and the fine lines correspond to the 25–75% confidence interval. The S₂₄ score discriminates both groups at least 80 h before intubation. We noted a net increase in S₂₄ score at 24 h prior to intubation.

Fig. 5

A - Evolution of the cumulative incidence of intubation with the severity score expressed as the rate of intubation. B - Evolution of the probability of intubation according to MS₂₄ score. The probability increased continuously until it reached almost 1 for MS₂₄ = 100.

Fig. 6

A - Score evolution of a stable patient (green). The S₂₄ score had the potential to increase, but quickly returned to zero, staying under 33 overall. The orange line shows an unstable patient with large score increases; this patient's condition improved without tracheal intubation. B – Score evolution of an unstable patient with a prolonged stay in the warning orange zone. This patient had to be intubated. C – Score evolution of an unstable patient with a terminal increase to a final S₂₄ score of 96 (MS₂₄ = 88).

Fig. 7

A - Changes in length of stay alongside MS₂₄ score (lengths of stay greater than 50 days were not plotted). Green dots represent non-intubated patients while red dots represent intubated patients. The purple line represents the moving average of length of stay with MS₂₄. B - Changes in length of stay with MS₂₄ score for non-intubated patients (red line is the linear regression) C -Confusion matrix of a prediction of a length of stay of >5 days using a MS₂₄ score greater than 40 that corresponds to the optimum of the ROC curve plotted in D.