Machine Learning to Predict ICU Admission, ICU Mortality and Survivors’ Length of Stay Among COVID-19 Patients: Toward Optimal Allocation of ICU Resources

2020 ◽  
Author(s):  
Tingting Dan ◽  
Yang Li ◽  
Ziwei Zhu ◽  
Xijie Chen ◽  
Wuxiu Quan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Length Of Stay ◽  
Optimal Allocation ◽  
Icu Mortality ◽  
Icu Admission

Machine learning combining CT findings and clinical parameters improves prediction of length of stay and ICU admission in torso trauma

2021 ◽  
Author(s):  
Pedro Vinícius Staziaki ◽  
Di Wu ◽  
Jesse C. Rayan ◽  
Irene Dixe de Oliveira Santo ◽  
Feng Nan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Length Of Stay ◽  
Clinical Parameters ◽  
Icu Admission ◽  
Ct Findings ◽  
Torso Trauma

scholarly journals Early Prediction of Seven-Day Mortality in Intensive Care Unit Using a Machine Learning Model: Results from the SPIN-UTI Project

2021 ◽  
Vol 10 (5) ◽  
pp. 992
Author(s):  
Martina Barchitta ◽  
Andrea Maugeri ◽  
Giuliana Favara ◽  
Paolo Marco Riela ◽  
Giovanni Gallo ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care ◽  
Intensive Care Units ◽  
Learning Algorithm ◽  
Area Under The Curve ◽  
Support Vector ◽  
Icu Admission ◽  
Risk Of Death ◽  
Saps Ii ◽  
Svm Algorithm

Patients in intensive care units (ICUs) were at higher risk of worsen prognosis and mortality. Here, we aimed to evaluate the ability of the Simplified Acute Physiology Score (SAPS II) to predict the risk of 7-day mortality, and to test a machine learning algorithm which combines the SAPS II with additional patients’ characteristics at ICU admission. We used data from the “Italian Nosocomial Infections Surveillance in Intensive Care Units” network. Support Vector Machines (SVM) algorithm was used to classify 3782 patients according to sex, patient’s origin, type of ICU admission, non-surgical treatment for acute coronary disease, surgical intervention, SAPS II, presence of invasive devices, trauma, impaired immunity, antibiotic therapy and onset of HAI. The accuracy of SAPS II for predicting patients who died from those who did not was 69.3%, with an Area Under the Curve (AUC) of 0.678. Using the SVM algorithm, instead, we achieved an accuracy of 83.5% and AUC of 0.896. Notably, SAPS II was the variable that weighted more on the model and its removal resulted in an AUC of 0.653 and an accuracy of 68.4%. Overall, these findings suggest the present SVM model as a useful tool to early predict patients at higher risk of death at ICU admission.

scholarly journals Healthcare trajectories before and after critical illness: population-based insight on diverse patients clusters

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Youenn Jouan ◽  
Leslie Grammatico-Guillon ◽  
Noémie Teixera ◽  
Claire Hassen-Khodja ◽  
Christophe Gaborit ◽  
...  
Keyword(s):  
Septic Shock ◽  
Mechanical Ventilation ◽  
Intensive Care ◽  
Critical Illness ◽  
Length Of Stay ◽  
Population Based ◽  
Hospital Length Of Stay ◽  
Healthcare Use ◽  
Icu Admission ◽  
Icu Discharge

Abstract Background The post intensive care syndrome (PICS) gathers various disabilities, associated with a substantial healthcare use. However, patients’ comorbidities and active medical conditions prior to intensive care unit (ICU) admission may partly drive healthcare use after ICU discharge. To better understand retative contribution of critical illness and PICS—compared to pre-existing comorbidities—as potential determinant of post-critical illness healthcare use, we conducted a population-based evaluation of patients’ healthcare use trajectories. Results Using discharge databases in a 2.5-million-people region in France, we retrieved, over 3 years, all adult patients admitted in ICU for septic shock or acute respiratory distress syndrome (ARDS), intubated at least 5 days and discharged alive from hospital: 882 patients were included. Median duration of mechanical ventilation was 11 days (interquartile ranges [IQR] 8;20), mean SAPS2 was 49, and median hospital length of stay was 42 days (IQR 29;64). Healthcare use (days spent in healthcare facilities) was analyzed 2 years before and 2 years after ICU admission. Prior to ICU admission, we observed, at the scale of the whole study population, a progressive increase in healthcare use. Healthcare trajectories were then explored at individual level, and patients were assembled according to their individual pre-ICU healthcare use trajectory by clusterization with the K-Means method. Interestingly, this revealed diverse trajectories, identifying patients with elevated and increasing healthcare use (n = 126), and two main groups with low (n = 476) or no (n = 251) pre-ICU healthcare use. In ICU, however, SAPS2, duration of mechanical ventilation and length of stay were not different across the groups. Analysis of post-ICU healthcare trajectories for each group revealed that patients with low or no pre-ICU healthcare (which represented 83% of the population) switched to a persistent and elevated healthcare use during the 2 years post-ICU. Conclusion For 83% of ARDS/septic shock survivors, critical illness appears to have a pivotal role in healthcare trajectories, with a switch from a low and stable healthcare use prior to ICU to a sustained higher healthcare recourse 2 years after ICU discharge. This underpins the hypothesis of long-term critical illness and PICS-related quantifiable consequences in healthcare use, measurable at a population level.

scholarly journals Validation of an Internationally Derived Patient Severity Phenotype to Support COVID-19 Analytics from Electronic Health Record Data

2021 ◽  
Author(s):  
Jeffrey G Klann ◽  
Griffin M Weber ◽  
Hossein Estiri ◽  
Bertrand Moal ◽  
Paul Avillach ◽  
...  
Keyword(s):  
Machine Learning ◽  
Electronic Health Record ◽  
Chart Review ◽  
Learning Approach ◽  
Health Record ◽  
Learning Approaches ◽  
Electronic Health Record Data ◽  
Icu Admission ◽  
Machine Learning Approach ◽  
Electronic Health

Abstract Introduction The Consortium for Clinical Characterization of COVID-19 by EHR (4CE) is an international collaboration addressing COVID-19 with federated analyses of electronic health record (EHR) data. Objective We sought to develop and validate a computable phenotype for COVID-19 severity. Methods Twelve 4CE sites participated. First we developed an EHR-based severity phenotype consisting of six code classes, and we validated it on patient hospitalization data from the 12 4CE clinical sites against the outcomes of ICU admission and/or death. We also piloted an alternative machine-learning approach and compared selected predictors of severity to the 4CE phenotype at one site. Results The full 4CE severity phenotype had pooled sensitivity of 0.73 and specificity 0.83 for the combined outcome of ICU admission and/or death. The sensitivity of individual code categories for acuity had high variability - up to 0.65 across sites. At one pilot site, the expert-derived phenotype had mean AUC 0.903 (95% CI: 0.886, 0.921), compared to AUC 0.956 (95% CI: 0.952, 0.959) for the machine-learning approach. Billing codes were poor proxies of ICU admission, with as low as 49% precision and recall compared to chart review. Discussion We developed a severity phenotype using 6 code classes that proved resilient to coding variability across international institutions. In contrast, machine-learning approaches may overfit hospital-specific orders. Manual chart review revealed discrepancies even in the gold-standard outcomes, possibly due to heterogeneous pandemic conditions. Conclusion We developed an EHR-based severity phenotype for COVID-19 in hospitalized patients and validated it at 12 international sites.

scholarly journals Machine learning prediction of breast cancer survival using age, sex, length of stay, mode of diagnosis and location of cancer

2021 ◽  
Author(s):  
Hilary I. Okagbue ◽  
Patience I. Adamu ◽  
Pelumi E. Oguntunde ◽  
Emmanuela C. M. Obasi ◽  
Oluwole A. Odetunmibi
Keyword(s):  
Breast Cancer ◽  
Machine Learning ◽  
Length Of Stay ◽  
Cancer Survival ◽  
Breast Cancer Survival

scholarly journals Development of a machine learning model for predicting pediatric mortality in the early stages of intensive care unit admission

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Bongjin Lee ◽  
Kyunghoon Kim ◽  
Hyejin Hwang ◽  
You Sun Kim ◽  
Eun Hee Chung ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Validation Cohort ◽  
External Validation ◽  
Learning Model ◽  
Derivation Cohort ◽  
Icu Admission ◽  
Early Stages ◽  
Machine Learning Model

AbstractThe aim of this study was to develop a predictive model of pediatric mortality in the early stages of intensive care unit (ICU) admission using machine learning. Patients less than 18 years old who were admitted to ICUs at four tertiary referral hospitals were enrolled. Three hospitals were designated as the derivation cohort for machine learning model development and internal validation, and the other hospital was designated as the validation cohort for external validation. We developed a random forest (RF) model that predicts pediatric mortality within 72 h of ICU admission, evaluated its performance, and compared it with the Pediatric Index of Mortality 3 (PIM 3). The area under the receiver operating characteristic curve (AUROC) of RF model was 0.942 (95% confidence interval [CI] = 0.912–0.972) in the derivation cohort and 0.906 (95% CI = 0.900–0.912) in the validation cohort. In contrast, the AUROC of PIM 3 was 0.892 (95% CI = 0.878–0.906) in the derivation cohort and 0.845 (95% CI = 0.817–0.873) in the validation cohort. The RF model in our study showed improved predictive performance in terms of both internal and external validation and was superior even when compared to PIM 3.

Hospitalization in patients with systemic lupus erythematosus at Tawam Hospital, United Arab Emirates (UAE): Rates, causes, and factors associated with length of stay

Lupus ◽  
2021 ◽  
pp. 096120332199008
Author(s):  
Reem Aldarmaki ◽  
Hiba I Al Khogali ◽  
Ali M Al Dhanhani
Keyword(s):  
Systemic Lupus Erythematosus ◽  
Length Of Stay ◽  
Lupus Nephritis ◽  
United Arab Emirates ◽  
Lupus Erythematosus ◽  
Hospitalization Rate ◽  
Systemic Lupus ◽  
Factors Associated ◽  
Icu Admission ◽  
The Mean

Introduction Systemic lupus erythematosus (SLE) is a relapsing and remitting multiorgan disease associated with significant morbidity and mortality. The survival rate of patients with SLE has recently improved, which was associated with increased morbidity and hospitalization rates. Therefore, this study aimed to examine the rate and causes of hospitalization in patients with SLE and explore factors associated with increased length of stay (LOS). Methods Patients who visited rheumatology clinics (Tawam hospital, United Arab Emirates (UAE)) and fulfilled the American College of Rheumatology (ACR) SLE criteria were identified. Retrospective charts were reviewed to determine previous admissions. Demographic data, reason for hospitalization, duration of hospitalization, intensive care unit (ICU) admission, number of specialist consultations, medications used, and SLE characteristics at time of admission were collected. The hospitalization rate was calculated as the number of hospitalized patients divided by the total number of patients with the disease. We performed multivariable regression analysis for factors associated with increased LOS. Results A total of 91 patients with SLE (88 women and 3 men) met the inclusion criteria with a mean disease duration of 10.2 years (SD 5.5). A total of 222 admissions were identified, and 66 of 91 patients were admitted at least once. The mean crude hospitalization rate calculated was 29.8%. The primary reason for admission was pregnancy (29%), SLE activity (24%), and infection (20%). When combining primary and secondary reasons, the proportion of admissions due to SLE activity increased to 32%. The mean LOS was 5.9 (SD 6.0) days. About 7% of admitted patients required ICU admission. In multivariable analysis, patients with lupus nephritis, complications during hospitalization, and increased number of specialists consultations and who were admitted to ICU and started new medication were all associated with increased LOS. Conclusion A significant proportion of patients with SLE were hospitalized during their disease course. The hospitalization rate in this study appears to be higher than those reported elsewhere. Disease flare is the leading cause of admission in patients with SLE in this relatively young cohort. Lupus nephritis has been found to be significantly related to longer LOS. Measurements taken to reduce the incidence and severity of flares would likely decrease hospitalization rate and LOS in patients with SLE.

Measurement of Dead Space Fraction Upon ICU Admission Predicts Length of Stay and Clinical Outcomes Following Bidirectional Cavopulmonary Anastomosis*

2018 ◽  
Vol 19 (1) ◽  
pp. 23-31 ◽  
Author(s):  
Claire L. Cigarroa ◽  
Sarah J. van den Bosch ◽  
Xiaoqi Tang ◽  
Kimberlee Gauvreau ◽  
Christopher W. Baird ◽  
...  
Keyword(s):  
Length Of Stay ◽  
Clinical Outcomes ◽  
Dead Space ◽  
Icu Admission ◽  
Cavopulmonary Anastomosis ◽  
Bidirectional Cavopulmonary Anastomosis
Sign in / Sign up

Export Citation Format

Share Document