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.

Development of an automated, general-purpose prediction tool for postoperative respiratory failure using machine learning: A retrospective cohort study.

STUDY OBJECTIVE: Postoperative respiratory failure is a major surgical complication and key quality metric. Existing prediction tools underperform, are limited to specific populations, and necessitate manual calculation. This limits their implementation. We aimed to create an improved, machine learning powered prediction tool with ideal characteristics for automated calculation. DESIGN, SETTING, AND PATIENTS: We retrospectively reviewed 101,455 anesthetic procedures from 1/2018 to 6/2021. The primary outcome was the Standardized Endpoints in Perioperative Medicine consensus definition for postoperative respiratory failure. Secondary outcomes were respiratory quality metrics from the National Surgery Quality Improvement Sample, Society of Thoracic Surgeons, and CMS. We abstracted from the electronic health record 26 procedural and physiologic variables previously identified as respiratory failure risk factors. We randomly split the cohort and used the Random Forest method to predict the composite outcome in the training cohort. We coined this the RESPIRE model and measured its accuracy in the validation cohort using area under the receiver operating curve (AUROC) analysis, among other measures, and compared this with ARISCAT and SPORC-1, two leading prediction tools. We compared performance in a validation cohort using score cut-offs determined in a separate test cohort. MAIN RESULTS: The RESPIRE model exhibited superior accuracy with an AUROC of 0.93 (95% CI, 0.92-0.95) compared to 0.82 for both ARISCAT and SPORC-1 (P-for-difference < 0.0001 for both). At comparable 80-90% sensitivities, RESPIRE had higher positive predictive value (11%, 95% CI: 10-12%) and lower false positive rate (12%, 95% CI: 12-13%) compared to 4% and 37% for both ARISCAT and SPORC-1. The RESPIRE model also better predicted the established quality metrics for postoperative respiratory failure. CONCLUSIONS: We developed a general-purpose, machine learning powered prediction tool with superior performance for research and quality-based definitions of postoperative respiratory failure.

Development and validation of a machine learning ASA-score to identify candidates for comprehensive preoperative screening and risk stratification.

OBJECTIVE: The ASA physical status (ASA-PS) is determined by an anesthesia provider or surgeon to communicate co-morbidities relevant to perioperative risk. Assigning an ASA-PS is a clinical decision and there is substantial provider-dependent variability. We developed and externally validated a machine learning-derived algorithm to determine ASA-PS (ML-PS) based on data available in the medical record. DESIGN: Retrospective multicenter hospital registry study. SETTING: University-affiliated hospital networks. PATIENTS: Patients who received anesthesia at Beth Israel Deaconess Medical Center (Boston, MA, training [n = 361,602] and internal validation cohorts [n = 90,400]) and Montefiore Medical Center (Bronx, NY, external validation cohort [n = 254,412]). MEASUREMENTS: The ML-PS was created using a supervised random forest model with 35 preoperatively available variables. Its predictive ability for 30-day mortality, postoperative ICU admission, and adverse discharge were determined by logistic regression. MAIN RESULTS: The anesthesiologist ASA-PS and ML-PS were in agreement in 57.2% of the cases (moderate inter-rater agreement). Compared with anesthesiologist rating, ML-PS assigned more patients into extreme ASA-PS (I and IV), (p < 0.01), and less patients in ASA II and III (p < 0.01). ML-PS and anesthesiologist ASA-PS had excellent predictive values for 30-day mortality, and good predictive values for postoperative ICU admission and adverse discharge. Among the 3594 patients who died within 30 days after surgery, net reclassification improvement analysis revealed that using the ML-PS, 1281 (35.6%) patients were reclassified into the higher clinical risk category compared with anesthesiologist rating. However, in a subgroup of multiple co-morbidity patients, anesthesiologist ASA-PS had a better predictive accuracy than ML-PS. CONCLUSIONS: We created and validated a machine learning physical status based on preoperatively available data. The ability to identify patients at high risk early in the preoperative process independent of the provider's decision is a part of the process we use to standardize the stratified preoperative evaluation of patients scheduled for ambulatory surgery.

Improving Adjuvant Liver-Directed Treatment Recommendations for Unresectable Hepatocellular Carcinoma: An Artificial Intelligence-Based Decision-Making Tool.

PURPOSE: Liver-directed therapy after transarterial chemoembolization (TACE) can lead to improvement in survival for selected patients with unresectable hepatocellular carcinoma (HCC). However, there is uncertainty in the appropriate application and modality of therapy in current clinical practice guidelines. The aim of this study was to develop a proof-of-concept, machine learning (ML) model for treatment recommendation in patients previously treated with TACE and select patients who might benefit from additional treatment with combination stereotactic body radiotherapy (SBRT) or radiofrequency ablation (RFA). METHODS: This retrospective observational study was based on data from an urban, academic hospital system selecting for patients diagnosed with stage I-III HCC from January 1, 2008, to December 31, 2018, treated with TACE, followed by adjuvant RFA, SBRT, or no additional liver-directed modality. A feedforward, ML ensemble model provided a treatment recommendation on the basis of pairwise assessments evaluating each potential treatment option and estimated benefit in survival. RESULTS: Two hundred thirty-seven patients met inclusion criteria, of whom 54 (23%) and 49 (21%) received combination of TACE and SBRT or TACE and RFA, respectively. The ML model suggested a different consolidative modality in 32.7% of cases among patients who had previously received combination treatment. Patients treated in concordance with model recommendations had significant improvement in progression-free survival (hazard ratio 0.5; P = .007). The most important features for model prediction were cause of cirrhosis, stage of disease, and albumin-bilirubin grade (a measure of liver function). CONCLUSION: In this proof-of-concept study, an ensemble ML model was able to provide treatment recommendations for HCC who had undergone prior TACE. Additional treatment in line with model recommendations was associated with significant improvement in progression-free survival, suggesting a potential benefit for ML-guided medical decision making.

Epidemiology of Organ Failure Before and During COVID-19 Pandemic Surge Conditions.

BACKGROUND: Understanding the distribution of organ failure before and during the COVID-19 pandemic surge can provide a deeper understanding of how the pandemic strained health care systems and affected outcomes. OBJECTIVE: To assess the distribution of organ failure in 3 New York City hospitals during the COVID-19 pandemic. METHODS: A retrospective cohort study of adult admissions across hospitals from February 1, 2020, through May 31, 2020, was conducted. The cohort was stratified into those admitted before March 17, 2020 (prepandemic) and those admitted on or after that date (SARS-CoV-2-positive and non-SARS-CoV-2). Sequential Organ Failure Assessment scores were computed every 2 hours for each admission. RESULTS: A total of 1 794 975 scores were computed for 20 704 admissions. Before and during the pandemic, renal failure was the most common type of organ failure at admission and respiratory failure was the most common type of hospital-onset organ failure. The SARS-CoV-2-positive group showed a 231% increase in respiratory failure compared with the prepandemic group. More than 65% of hospital-onset organ failure in the prepandemic group and 83% of hospital-onset respiratory failure in the SARS-CoV-2-positive group occurred outside intensive care units. The SARS-CoV-2-positive group showed a 341% increase in multiorgan failure compared with the prepandemic group. Compared with the prepandemic and non-SARS-CoV-2 patients, SARS-CoV-2-positive patients had significantly higher mortality for the same admission and maximum organ failure score. CONCLUSION: Most hospital-onset organ failure began outside intensive care units, with a marked increase in multiorgan failure during pandemic surge conditions and greater hospital mortality for the severity of organ failure.