Electronic "Sniffer" Systems to Identify the Acute Respiratory Distress Syndrome

Machine Learning Tools for Acute Respiratory Distress Syndrome Detection and Prediction

Machine learning (ML) tools for acute respiratory distress syndrome (ARDS) detection and prediction are increasingly used. Therefore, understanding risks and benefits of such algorithms is relevant at the bedside. ARDS is a complex and severe lung condition that can be challenging to define precisely due to its multifactorial nature. It often arises as a response to various underlying medical conditions, such as pneumonia, sepsis, or trauma, leading to widespread inflammation in the lungs. ML has shown promising potential in supporting the recognition of ARDS in ICU patients. By analyzing a variety of clinical data, including vital signs, laboratory results, and imaging findings, ML models can identify patterns and risk factors associated with the development of ARDS. This detection and prediction could be crucial for timely interventions, diagnosis and treatment. In summary, leveraging ML for the early prediction and detection of ARDS in ICU patients holds great potential to enhance patient care, improve outcomes, and contribute to the evolving landscape of precision medicine in critical care settings. This article is a concise definitive review on artificial intelligence and ML tools for the prediction and detection of ARDS in critically ill patients.

Automatic ARDS surveillance with chest X-ray recognition using convolutional neural networks

OBJECTIVE

This study aims to design, validate and assess the accuracy a deep learning model capable of differentiation Chest X-Rays between pneumonia, acute respiratory distress syndrome (ARDS) and normal lungs.

MATERIALS AND METHODS

A diagnostic performance study was conducted using Chest X-Ray images from adult patients admitted to a medical intensive care unit between January 2003 and November 2014. X-ray images from 15,899 patients were assigned one of three prespecified categories: "ARDS", "Pneumonia", or "Normal".

RESULTS

A two-step convolutional neural network (CNN) pipeline was developed and tested to distinguish between the three patterns with sensitivity ranging from 91.8% to 97.8% and specificity ranging from 96.6% to 98.8%. The CNN model was validated with a sensitivity of 96.3% and specificity of 96.6% using a previous dataset of patients with Acute Lung Injury (ALI)/ARDS.

DISCUSSION

The results suggest that a deep learning model based on chest x-ray pattern recognition can be a useful tool in distinguishing patients with ARDS from patients with normal lungs, providing faster results than digital surveillance tools based on text reports.

CONCLUSION

A CNN-based deep learning model showed clinically significant performance, providing potential for faster ARDS identification. Future research should prospectively evaluate these tools in a clinical setting.

How Electronic Medical Record Integration Can Support More Efficient Critical Care Clinical Trials

Large volumes of data are collected on critically ill patients, and using data science to extract information from the electronic medical record (EMR) and to inform the design of clinical trials represents a new opportunity in critical care research. Using improved methods of phenotyping critical illnesses, subject identification and enrollment, and targeted treatment group assignment alongside newer trial designs such as adaptive platform trials can increase efficiency while lowering costs. Some tools such as the EMR to automate data collection are already in use. Refinement of data science approaches in critical illness research will allow for better clinical trials and, ultimately, improved patient outcomes.

Long short-term memory model identifies ARDS and in-hospital mortality in both non-COVID-19 and COVID-19 cohort

OBJECTIVE

To identify the risk of acute respiratory distress syndrome (ARDS) and in-hospital mortality using long short-term memory (LSTM) framework in a mechanically ventilated (MV) non-COVID-19 cohort and a COVID-19 cohort.

METHODS

We included MV ICU patients between 2017 and 2018 and reviewed patient records for ARDS and death. Using active learning, we enriched this cohort with MV patients from 2016 to 2019 (MV non-COVID-19, n=3905). We collected a second validation cohort of hospitalised patients with COVID-19 in 2020 (COVID+, n=5672). We trained an LSTM model using 132 structured features on the MV non-COVID-19 training cohort and validated on the MV non-COVID-19 validation and COVID-19 cohorts.

RESULTS

Applying LSTM (model score 0.9) on the MV non-COVID-19 validation cohort had a sensitivity of 86% and specificity of 57%. The model identified the risk of ARDS 10 hours before ARDS and 9.4 days before death. The sensitivity (70%) and specificity (84%) of the model on the COVID-19 cohort are lower than MV non-COVID-19 cohort. For the COVID-19 + cohort and MV COVID-19 + patients, the model identified the risk of in-hospital mortality 2.4 days and 1.54 days before death, respectively.

DISCUSSION

Our LSTM algorithm accurately and timely identified the risk of ARDS or death in MV non-COVID-19 and COVID+ patients. By alerting the risk of ARDS or death, we can improve the implementation of evidence-based ARDS management and facilitate goals-of-care discussions in high-risk patients.

CONCLUSION

Using the LSTM algorithm in hospitalised patients identifies the risk of ARDS or death.

Evaluating construct validity of computable acute respiratory distress syndrome definitions in adults hospitalized with COVID-19: an electronic health records based approach

BACKGROUND

Evolving ARDS epidemiology and management during COVID-19 have prompted calls to reexamine the construct validity of Berlin criteria, which have been rarely evaluated in real-world data. We developed a Berlin ARDS definition (EHR-Berlin) computable in electronic health records (EHR) to (1) assess its construct validity, and (2) assess how expanding its criteria affected validity.

METHODS

We performed a retrospective cohort study at two tertiary care hospitals with one EHR, among adults hospitalized with COVID-19 February 2020-March 2021. We assessed five candidate definitions for ARDS: the EHR-Berlin definition modeled on Berlin criteria, and four alternatives informed by recent proposals to expand criteria and include patients on high-flow oxygen (EHR-Alternative 1), relax imaging criteria (EHR-Alternatives 2-3), and extend timing windows (EHR-Alternative 4). We evaluated two aspects of construct validity for the EHR-Berlin definition: (1) criterion validity: agreement with manual ARDS classification by experts, available in 175 patients; (2) predictive validity: relationships with hospital mortality, assessed by Pearson r and by area under the receiver operating curve (AUROC). We assessed predictive validity and timing of identification of EHR-Berlin definition compared to alternative definitions.

RESULTS

Among 765 patients, mean (SD) age was 57 (18) years and 471 (62%) were male. The EHR-Berlin definition classified 171 (22%) patients as ARDS, which had high agreement with manual classification (kappa 0.85), and was associated with mortality (Pearson r = 0.39; AUROC 0.72, 95% CI 0.68, 0.77). In comparison, EHR-Alternative 1 classified 219 (29%) patients as ARDS, maintained similar relationships to mortality (r = 0.40; AUROC 0.74, 95% CI 0.70, 0.79, Delong test P = 0.14), and identified patients earlier in their hospitalization (median 13 vs. 15 h from admission, Wilcoxon signed-rank test P < 0.001). EHR-Alternative 3, which removed imaging criteria, had similar correlation (r = 0.41) but better discrimination for mortality (AUROC 0.76, 95% CI 0.72, 0.80; P = 0.036), and identified patients median 2 h (P < 0.001) from admission.

CONCLUSIONS

The EHR-Berlin definition can enable ARDS identification with high criterion validity, supporting large-scale study and surveillance. There are opportunities to expand the Berlin criteria that preserve predictive validity and facilitate earlier identification.

An alert tool to promote lung protective ventilation for possible acute respiratory distress syndrome

Objective

Computer-aided decision tools may speed recognition of acute respiratory distress syndrome (ARDS) and promote consistent, timely treatment using lung-protective ventilation (LPV). This study evaluated implementation and service (process) outcomes with deployment and use of a clinical decision support (CDS) synchronous alert tool associated with existing computerized ventilator protocols and targeted patients with possible ARDS not receiving LPV.

Materials and Methods

We performed an explanatory mixed methods study from December 2019 to November 2020 to evaluate CDS alert implementation outcomes across 13 intensive care units (ICU) in an integrated healthcare system with >4000 mechanically ventilated patients annually. We utilized quantitative methods to measure service outcomes including CDS alert tool utilization, accuracy, and implementation effectiveness. Attitudes regarding the appropriateness and acceptability of the CDS tool were assessed via an electronic field survey of physicians and advanced practice providers.

Results

Thirty-eight percent of study encounters had at least one episode of LPV nonadherence. Addition of LPV treatment detection logic prevented an estimated 1812 alert messages (41%) over use of disease detection logic alone. Forty-eight percent of alert recommendations were implemented within 2 h. Alert accuracy was estimated at 63% when compared to gold standard ARDS adjudication, with sensitivity of 85% and positive predictive value of 62%. Fifty-seven percent of survey respondents observed one or more benefits associated with the alert.

Conclusion

Introduction of a CDS alert tool based upon ARDS risk factors and integrated with computerized ventilator protocol instructions increased visibility to gaps in LPV use and promoted increased adherence to LPV.

Rule-Based Cohort Definitions for Acute Respiratory Distress Syndrome: A Computable Phenotyping Strategy Based on the Berlin Definition

Supplemental Digital Content is available in the text.

OBJECTIVES:

Accurate identification of acute respiratory distress syndrome is essential for understanding its epidemiology, patterns of care, and outcomes. We aimed to design a computable phenotyping strategy to detect acute respiratory distress syndrome in electronic health records of critically ill patients.

DESIGN:

This is a retrospective cohort study. Using a near real-time copy of the electronic health record, we developed a computable phenotyping strategy to detect acute respiratory distress syndrome based on the Berlin definition.

SETTING:

Twenty multidisciplinary ICUs in Mayo Clinic Health System.

SUBJECTS:

The phenotyping strategy was applied to 196,487 consecutive admissions from year 2009 to 2019.

INTERVENTIONS:

The acute respiratory distress syndrome cohort generated by this novel strategy was compared with the acute respiratory distress syndrome cohort documented by clinicians during the same period. The sensitivity and specificity of the phenotyping strategy were calculated in randomly selected patient cohort (50 patients) using the results from manual medical record review as gold standard.

MEASUREMENTS AND MAIN RESULTS:

Among the patients who did not have acute respiratory distress syndrome documented, the computable phenotyping strategy identified 3,169 adult patients who met the Berlin definition, 676 patients (21.3%) were classified to have severe acute respiratory distress syndrome (Pao₂/Fio₂ ratio ≤ 100), 1,535 patients (48.4%) had moderate acute respiratory distress syndrome (100 < Pao₂/Fio₂ ratio ≤ 200), and 958 patients (30.2%) had mild acute respiratory distress syndrome (200 < Pao₂/Fio₂ ratio ≤ 300). The phenotyping strategy achieved a sensitivity of 94.4%, specificity of 96.9%, positive predictive value of 94.4%, and negative predictive value of 96.9% in a randomly selected patient cohort. The clinicians documented acute respiratory distress syndrome in 1,257 adult patients during the study period. The clinician documentation rate of acute respiratory distress syndrome was 28.4%. Compared with the clinicians’ documentation, the phenotyping strategy identified a cohort that had higher acuity and complexity of illness suggested by higher Sequential Organ Failure Assessment score (9 vs 7; p < 0.0001), higher Acute Physiology and Chronic Health Evaluation score (76 vs 63; p < 0.0001), higher rate of requiring invasive mechanical ventilation (99.1% vs 71.8%; p < 0.0001), higher ICU mortality (20.6% vs 16.8%; p < 0.0001), and longer ICU length of stay (5.1 vs 4.2 d; p < 0.0001).

CONCLUSIONS:

Our rule-based computable phenotyping strategy can accurately detect acute respiratory distress syndrome in critically ill patients in the setting of high clinical complexity. This strategy can be applied to enhance early recognition of acute respiratory distress syndrome and to facilitate best-care delivery and clinical research in acute respiratory distress syndrome.

Deep learning to detect acute respiratory distress syndrome on chest radiographs: a retrospective study with external validation

Background

Acute respiratory distress syndrome (ARDS) is a common, but under-recognised, critical illness syndrome associated with high mortality. An important factor in its under-recognition is the variability in chest radiograph interpretation for ARDS. We sought to train a deep convolutional neural network (CNN) to detect ARDS findings on chest radiographs.

Methods

CNNs were pretrained on 595 506 radiographs from two centres to identify common chest findings (eg, opacity and effusion), and then trained on 8072 radiographs annotated for ARDS by multiple physicians using various transfer learning approaches. The best performing CNN was tested on chest radiographs in an internal and external cohort, including a subset reviewed by six physicians, including a chest radiologist and physicians trained in intensive care medicine. Chest radiograph data were acquired from four US hospitals.

Findings

In an internal test set of 1560 chest radiographs from 455 patients with acute hypoxaemic respiratory failure, a CNN could detect ARDS with an area under the receiver operator characteristics curve (AUROC) of 0·92 (95% CI 0·89–0·94). In the subgroup of 413 images reviewed by at least six physicians, its AUROC was 0·93 (95% CI 0·88–0·96), sensitivity 83·0% (95% CI 74·0–91·1), and specificity 88·3% (95% CI 83·1–92·8). Among images with zero of six ARDS annotations (n=155), the median CNN probability was 11%, with six (4%) assigned a probability above 50%. Among images with six of six ARDS annotations (n=27), the median CNN probability was 91%, with two (7%) assigned a probability below 50%. In an external cohort of 958 chest radiographs from 431 patients with sepsis, the AUROC was 0·88 (95% CI 0·85–0·91). When radiographs annotated as equivocal were excluded, the AUROC was 0·93 (0·92–0·95).

Interpretation

A CNN can be trained to achieve expert physician-level performance in ARDS detection on chest radiographs. Further research is needed to evaluate the use of these algorithms to support real-time identification of ARDS patients to ensure fidelity with evidence-based care or to support ongoing ARDS research.

Funding

National Institutes of Health, Department of Defense, and Department of Veterans Affairs.

Use of Machine Learning to Screen for Acute Respiratory Distress Syndrome Using Raw Ventilator Waveform Data

To develop and characterize a machine learning algorithm to discriminate acute respiratory distress syndrome from other causes of respiratory failure using only ventilator waveform data.

Design

Retrospective, observational cohort study.

Setting

Academic medical center ICU.

Patients

Adults admitted to the ICU requiring invasive mechanical ventilation, including 50 patients with acute respiratory distress syndrome and 50 patients with primary indications for mechanical ventilation other than hypoxemic respiratory failure.

Interventions

None.

Measurements and Main Results

Pressure and flow time series data from mechanical ventilation during the first 24-hours after meeting acute respiratory distress syndrome criteria (or first 24-hr of mechanical ventilation for non-acute respiratory distress syndrome patients) were processed to extract nine physiologic features. A random forest machine learning algorithm was trained to discriminate between the patients with and without acute respiratory distress syndrome. Model performance was assessed using the area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and negative predictive value. Analyses examined performance when the model was trained using data from the first 24 hours and tested using withheld data from either the first 24 hours (24/24 model) or 6 hours (24/6 model). Area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and negative predictive value were 0.88, 0.90, 0.71, 0.77, and 0.90 (24/24); and 0.89, 0.90, 0.75, 0.83, and 0.83 (24/6).

Conclusions

Use of machine learning and physiologic information derived from raw ventilator waveform data may enable acute respiratory distress syndrome screening at early time points after intubation. This approach, combined with traditional diagnostic criteria, could improve timely acute respiratory distress syndrome recognition and enable automated clinical decision support, especially in settings with limited availability of conventional diagnostic tests and electronic health records.

Performance Measure Development, Use, and Measurement of Effectiveness Using the Guideline on Mechanical Ventilation in Acute Respiratory Distress Syndrome. An Official American Thoracic Society Workshop Report

Guideline implementation tools are designed to improve uptake of guideline recommendations in clinical settings but do not uniformly accompany the clinical practice guideline documents. Performance measures are a type of guideline implementation tool with the potential to catalyze behavior change and greater adherence to clinical practice guidelines. However, many performance measures suffer from serious flaws in their design and application, prompting the American Thoracic Society (ATS) to define its own performance measure development standards in a previous workshop in 2012. This report summarizes the proceedings of a follow-up workshop convened to advance the ATS’s work in performance measure development and guideline implementation. To illustrate the application of the ATS’s performance measure development framework, we used the example of a low–tidal volume ventilation performance measure created de novo from the 2017 ATS/European Society of Intensive Care Medicine/Society of Critical Care Medicine mechanical ventilation in acute respiratory distress syndrome clinical practice guideline. We include a detailed explanation of the rationale for the specifications chosen, identification of areas in need of further validity testing, and a preliminary strategy for pilot testing of the performance measure. Pending additional resources and broader performance measure expertise, issuing “preliminary performance measures” and their specifications alongside an ATS clinical practice guideline offers a first step to further the ATS’s guideline implementation agenda. We recommend selectively proceeding with full performance measure development for those measures with positive early user feedback and the greatest potential impact in accordance with ATS leadership guidance.

Identifying Clinical Research Priorities in Adult Pulmonary and Critical Care. NHLBI Working Group Report

Preventing, treating, and promoting recovery from critical illness due to pulmonary disease are foundational goals of the critical care community and the NHLBI. Decades of clinical research in acute respiratory distress syndrome, acute respiratory failure, pneumonia, and sepsis have yielded improvements in supportive care, which have translated into improved patient outcomes. Novel therapeutics have largely failed to translate from promising preclinical findings into improved patient outcomes in late-phase clinical trials. Recent advances in personalized medicine, "big data," causal inference using observational data, novel clinical trial designs, preclinical disease modeling, and understanding of recovery from acute illness promise to transform the methods of pulmonary and critical care clinical research. To assess the current state of, research priorities for, and future directions in adult pulmonary and critical care research, the NHLBI assembled a multidisciplinary working group of investigators. This working group identified recommendations for future research, including 1) focusing on understanding the clinical, physiological, and biological underpinnings of heterogeneity in syndromes, diseases, and treatment response with the goal of developing targeted, personalized interventions; 2) optimizing preclinical models by incorporating comorbidities, cointerventions, and organ support; 3) developing and applying novel clinical trial designs; and 4) advancing mechanistic understanding of injury and recovery to develop and test interventions targeted at achieving long-term improvements in the lives of patients and families. Specific areas of research are highlighted as especially promising for making advances in pneumonia, acute hypoxemic respiratory failure, and acute respiratory distress syndrome.

Supervised machine learning for the early prediction of acute respiratory distress syndrome (ARDS)

PURPOSE

Acute respiratory distress syndrome (ARDS) is a serious respiratory condition with high mortality and associated morbidity. The objective of this study is to develop and evaluate a novel application of gradient boosted tree models trained on patient health record data for the early prediction of ARDS.

MATERIALS AND METHODS

9919 patient encounters were retrospectively analyzed from the Medical Information Mart for Intensive Care III (MIMIC-III) data base. XGBoost gradient boosted tree models for early ARDS prediction were created using routinely collected clinical variables and numerical representations of radiology reports as inputs. XGBoost models were iteratively trained and validated using 10-fold cross validation.

RESULTS

On a hold-out test set, algorithm classifiers attained area under the receiver operating characteristic curve (AUROC) values of 0.905 when tested for the detection of ARDS at onset and 0.827, 0.810, and 0.790 for the prediction of ARDS at 12-, 24-, and 48-h windows prior to onset, respectively.

CONCLUSION

Supervised machine learning predictions may help predict patients with ARDS up to 48 h prior to onset.

Personalized pharmacological therapy for ARDS: a light at the end of the tunnel

Introduction: Pharmacotherapy for the acute respiratory distress syndrome (ARDS) has been tested in preclinical and clinical studies. However, to date, no pharmacological interventions have proven effective. This may be attributed to lack of proper identification of different ARDS phenotypes.Areas covered: We designed inclusive search strings and searched four bibliographic databases (Cochrane Database of Systematic Reviews, PubMed, Web of Science, and clinicaltrials.gov) to identify relevant research. Search results were mainly restricted to papers published from 2009 through 2019. ARDS is a heterogeneous syndrome, and its different phenotypes - defined according to clinical, radiological, and biological parameters - may affect response to therapy. The most promising pharmacological approaches to date have been based on ARDS pathophysiology. They focus on reducing inflammation and pulmonary edema, promoting selective vasodilation, and repairing alveolar epithelial and endothelial cells.Expert opinion: Pharmacotherapeutic approaches targeting ARDS pathophysiology have failed to exert beneficial effects. Personalized medicine targeting the different ARDS phenotypes has emerged as an option to improve survival. Identification of specific ARDS patient phenotypes that respond to specific therapies seems to be the most important challenge for the next decade. Additional research is warranted before personalized medicine approaches can be applied at bedside for ARDS patients.