scholarly journals Using Machine Learning to Predict Mortality for COVID-19 Patients on Day Zero in the ICU

2021 ◽  
Author(s):  
Elham Jamshidi ◽  
Amirhossein Asgary ◽  
Nader Tavakoli ◽  
Alireza Zali ◽  
Hadi Esmaily ◽  
...  
Keyword(s):  
Machine Learning ◽  
Complete Blood Count ◽  
Treatment Strategies ◽  
International Normalized Ratio ◽  
Early Prediction ◽  
Icu Patients ◽  
Distribution Width ◽  
Predictive Values ◽  
Prediction Of Mortality ◽  
The Impact

AbstractRationaleGiven the expanding number of COVID-19 cases and the potential for upcoming waves of infection, there is an urgent need for early prediction of the severity of the disease in intensive care unit (ICU) patients to optimize treatment strategies.ObjectivesEarly prediction of mortality using machine learning based on typical laboratory results and clinical data registered on the day of ICU admission.MethodsWe studied retrospectively 263 COVID-19 ICU patients. To find parameters with the highest predictive values, Kolmogorov-Smirnov and Pearson chi-squared tests were used. Logistic regression and random forest (RF) algorithms were utilized to build classification models. The impact of each marker on the RF model predictions was studied by implementing the local interpretable model-agnostic explanation technique (LIME-SP).ResultsAmong 66 documented parameters, 15 factors with the highest predictive values were identified as follows: gender, age, blood urea nitrogen (BUN), creatinine, international normalized ratio (INR), albumin, mean corpuscular volume, white blood cell count, segmented neutrophil count, lymphocyte count, red cell distribution width (RDW), and mean cell hemoglobin along with a history of neurological, cardiovascular, and respiratory disorders. Our RF model can predict patients outcomes with a sensitivity of 70% and a specificity of 75%.ConclusionsThe most decisive variables in our model were increased levels of BUN, lowered albumin levels, increased creatinine, INR, and RDW along with gender and age. Complete blood count parameters were also crucial for some patients. Considering the importance of early triage decisions, this model can be a useful tool in COVID-19 ICU decision-making.

scholarly journals Using Machine Learning to Predict Mortality for COVID-19 Patients on Day 0 in the ICU

2022 ◽  
Vol 3 ◽  
Author(s):  
Elham Jamshidi ◽  
Amirhossein Asgary ◽  
Nader Tavakoli ◽  
Alireza Zali ◽  
Soroush Setareh ◽  
...  
Keyword(s):  
Machine Learning ◽  
Treatment Strategies ◽  
Machine Learning Algorithms ◽  
Gradient Boosting ◽  
Support Vector ◽  
Early Prediction ◽  
Distribution Width ◽  
Predictive Values ◽  
Prediction Of Mortality ◽  
The Impact

Rationale: Given the expanding number of COVID-19 cases and the potential for new waves of infection, there is an urgent need for early prediction of the severity of the disease in intensive care unit (ICU) patients to optimize treatment strategies.Objectives: Early prediction of mortality using machine learning based on typical laboratory results and clinical data registered on the day of ICU admission.Methods: We retrospectively studied 797 patients diagnosed with COVID-19 in Iran and the United Kingdom (U.K.). To find parameters with the highest predictive values, Kolmogorov-Smirnov and Pearson chi-squared tests were used. Several machine learning algorithms, including Random Forest (RF), logistic regression, gradient boosting classifier, support vector machine classifier, and artificial neural network algorithms were utilized to build classification models. The impact of each marker on the RF model predictions was studied by implementing the local interpretable model-agnostic explanation technique (LIME-SP).Results: Among 66 documented parameters, 15 factors with the highest predictive values were identified as follows: gender, age, blood urea nitrogen (BUN), creatinine, international normalized ratio (INR), albumin, mean corpuscular volume (MCV), white blood cell count, segmented neutrophil count, lymphocyte count, red cell distribution width (RDW), and mean cell hemoglobin (MCH) along with a history of neurological, cardiovascular, and respiratory disorders. Our RF model can predict patient outcomes with a sensitivity of 70% and a specificity of 75%. The performance of the models was confirmed by blindly testing the models in an external dataset.Conclusions: Using two independent patient datasets, we designed a machine-learning-based model that could predict the risk of mortality from severe COVID-19 with high accuracy. The most decisive variables in our model were increased levels of BUN, lowered albumin levels, increased creatinine, INR, and RDW, along with gender and age. Considering the importance of early triage decisions, this model can be a useful tool in COVID-19 ICU decision-making.

scholarly journals The Importance of Haemogram Parameters in the Diagnosis and Prognosis of Septic Patients

2017 ◽  
Vol 3 (3) ◽  
pp. 105-110 ◽  
Author(s):  
Alina Elena Orfanu ◽  
Cristina Popescu ◽  
Anca Leuștean ◽  
Anca Ruxandra Negru ◽  
Cătălin Tilişcan ◽  
...  
Keyword(s):  
Immune Status ◽  
Complete Blood Count ◽  
Therapy Monitoring ◽  
Distribution Width ◽  
Predictive Values ◽  
Prolonged Hospitalisation ◽  
Diagnosis And Prognosis ◽  
Infectious Focus ◽  
Severe Pathology ◽  
Sensitivity Specificity

AbstractSepsis represents a severe pathology that requires both rapid and precise positive and differential diagnosis to identify patients who need immediate antimicrobial therapy. Monitoring septic patients′ outcome leads to prolonged hospitalisation and antibacterial therapy, often accompanied by substantial side effects, complications and a high mortality risk. Septic patients present with complex pathophysiological and immunological disorders and with a predominance of pro-inflammatory or anti-inflammatory mediators which are heterogeneous with respect to the infectious focus, the aetiology of sepsis or patients′ immune status or comorbidities. Previous studies performed have analysed inflammatory biomarkers, but a test or combinations of tests that can quickly and precisely establish a diagnosis or prognosis of septic patients has yet to be discovered. Recent research has focused on re-analysing older accessible parameters found in the complete blood count to determine the sensitivity, specificity, positive and negative predictive values for the diagnosis and prognosis of sepsis. The neutrophil/lymphocyte count ratio (NLCR), mean platelet volume (MPV) and red blood cells distribution width (RDW) are haemogram indicators which have been evaluated and which are of proven use in septic patients′ management.

scholarly journals A Machine Learning Approach for the Prediction of Traumatic Brain Injury Induced Coagulopathy

2021 ◽  
Vol 8 ◽  
Author(s):  
Fan Yang ◽  
Chi Peng ◽  
Liwei Peng ◽  
Jian Wang ◽  
Yuejun Li ◽  
...  
Keyword(s):  
Machine Learning ◽  
Traumatic Brain Injury ◽  
Intensive Care ◽  
Brain Injury ◽  
Blood Cell ◽  
Red Blood Cell ◽  
Research Database ◽  
Distribution Width ◽  
Predictive Values ◽  
Extreme Gradient Boosting

Background: Traumatic brain injury-induced coagulopathy (TBI-IC), is a disease with poor prognosis and increased mortality rate.Objectives: Our study aimed to identify predictors as well as develop machine learning (ML) models to predict the risk of coagulopathy in this population.Methods: ML models were developed and validated based on two public databases named Medical Information Mart for Intensive Care (MIMIC)-IV and the eICU Collaborative Research Database (eICU-CRD). Candidate predictors, including demographics, family history, comorbidities, vital signs, laboratory findings, injury type, therapy strategy and scoring system were included. Models were compared on area under the curve (AUC), accuracy, sensitivity, specificity, positive and negative predictive values, and decision curve analysis (DCA) curve.Results: Of 999 patients in MIMIC-IV included in the final cohort, a total of 493 (49.35%) patients developed coagulopathy following TBI. Recursive feature elimination (RFE) selected 15 variables, including international normalized ratio (INR), prothrombin time (PT), sepsis related organ failure assessment (SOFA), activated partial thromboplastin time (APTT), platelet (PLT), hematocrit (HCT), red blood cell (RBC), hemoglobin (HGB), blood urea nitrogen (BUN), red blood cell volume distribution width (RDW), creatinine (CRE), congestive heart failure, myocardial infarction, sodium, and blood transfusion. The external validation in eICU-CRD demonstrated that adapting boosting (Ada) model had the highest AUC of 0.924 (95% CI: 0.902–0.943). Furthermore, in the DCA curve, the Ada model and the extreme Gradient Boosting (XGB) model had relatively higher net benefits (ie, the correct classification of coagulopathy considering a trade-off between false- negatives and false-positives)—over other models across a range of threshold probability values.Conclusions: The ML models, as indicated by our study, can be used to predict the incidence of TBI-IC in the intensive care unit (ICU).

scholarly journals Early Prediction of Mortality, Severity, and Length of Stay in the Intensive Care Unit of Sepsis Patients Based on Sepsis 3.0 by Machine Learning Models

2021 ◽  
Vol 8 ◽  
Author(s):  
Longxiang Su ◽  
Zheng Xu ◽  
Fengxiang Chang ◽  
Yingying Ma ◽  
Shengjun Liu ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Random Forest ◽  
Random Forest Classifier ◽  
Early Prediction ◽  
Learning Models ◽  
Prediction Of Mortality ◽  
Severity Prediction ◽  
Machine Learning Models

Background: Early prediction of the clinical outcome of patients with sepsis is of great significance and can guide treatment and reduce the mortality of patients. However, it is clinically difficult for clinicians.Methods: A total of 2,224 patients with sepsis were involved over a 3-year period (2016–2018) in the intensive care unit (ICU) of Peking Union Medical College Hospital. With all the key medical data from the first 6 h in the ICU, three machine learning models, logistic regression, random forest, and XGBoost, were used to predict mortality, severity (sepsis/septic shock), and length of ICU stay (LOS) (>6 days, ≤ 6 days). Missing data imputation and oversampling were completed on the dataset before introduction into the models.Results: Compared to the mortality and LOS predictions, the severity prediction achieved the best classification results, based on the area under the operating receiver characteristics (AUC), with the random forest classifier (sensitivity = 0.65, specificity = 0.73, F1 score = 0.72, AUC = 0.79). The random forest model also showed the best overall performance (mortality prediction: sensitivity = 0.50, specificity = 0.84, F1 score = 0.66, AUC = 0.74; LOS prediction: sensitivity = 0.79, specificity = 0.66, F1 score = 0.69, AUC = 0.76) among the three models. The predictive ability of the SOFA score itself was inferior to that of the above three models.Conclusions: Using the random forest classifier in the first 6 h of ICU admission can provide a comprehensive early warning of sepsis, which will contribute to the formulation and management of clinical decisions and the allocation and management of resources.

Inflammatory indexes neutrophils/lymphocytes, platelets/lymphocytes and red-cell distribution width (RDW) as prognostic biomarkers in advanced solitary fibrous tumors (SFT) treated with pazopanib: Correlative study of GEIS-32 trial.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 11511-11511
Author(s):  
Javier Martin Broto ◽  
Samuel Hidalgo ◽  
David Silva Moura ◽  
Silvia Stacchiotti ◽  
Antonio Lopez-Pousa ◽  
...  
Keyword(s):  
Overall Survival ◽  
Multivariate Analysis ◽  
Complete Blood Count ◽  
Progression Free Survival ◽  
Statistical Correlation ◽  
Phase Iii ◽  
Pivotal Phase ◽  
Distribution Width ◽  
Best Response ◽  
The Impact

11511 Background: Pazopanib (P) was assessed prospectively in a phase 2 study in SFT resulting in a longer progression free survival (PFS) and overall survival (OS) compared to historical controls treated with chemotherapy. No statistical correlation was found between angiogenic factors and P in its pivotal phase III sarcoma trial. In the last two years, a soaring interest on the prognostic and predictive value of inflammatory indexes such as neutrophil/lymphocyte (NLR) and platelet/lymphocyte (PLR) is emerging in sarcomas. A retrospective analysis of inflammatory indexes of patients who entered the GEIS-32 (NCT02066285) trial was performed. In that trial advanced SFT patients were treated with P from front-line. Methods: All eligible patients who entered in the typical- and malignant-SFT cohort of the GEIS-32 trial were included in this analysis. To determine NLR and PLR, baseline values of platelets (10e9/L), neutrophils (10e9/L) and lymphocytes (10e9/L) were obtained from complete blood count tests. Additionally, RDW (standardized as 1 = upper value of normal range) values at baseline were also determined. The impact of NLR, PLR and RDW on OS, PFS and Choi response were analyzed by univariate and multivariate analysis. MAXSTAT was used to determine optimal cut-off points for overall survival. Metastasis free interval (MFI), mitotic count and ECOG were also analyzed, among others. Results: Sixty-seven out of 70 enrolled SFT patients, median age 63-y and 57% female, were considered for this analysis. The median follow-up from treatment initiation was 20.0 months. High standardized RDW value at baseline (cut-off 1.03) was significantly associated with worse OS [10.7 months (95% CI 3.8-17.5) vs 49.8 months (95% CI 9.4-90.2), p < 0.001] and worse PFS [8.8 months (95% CI 0.9-7.0) vs 9.8 months (7.4-12.3), p = 0.001]. High PLR (cut-off 242) significantly correlated with worse OS [10.7 months (95% CI 5.2-16.2) vs 49.8 months (95% CI 14.6-85.0), p < 0.001] and worse PFS [4.5 months (95% CI 2.0-7.0) vs 10.1 months (95% CI 6.3-13.9), p = 0.005], and high NLR (cut-off 3.78) was significantly associated with worse OS [11.7 months (95% CI 3.5-19.8) vs NA, p < 0.001] and worse PFS [4.5 months (95% CI 1.9-7.0) vs 10.8 months (95% CI 8.7-12.9), p = 0.010]. Independent variables in multivariate analysis were NLR, RDW, MFI and mitosis for PFS; while RDW and ECOG for OS (see table). Further, NLR and mitosis were independent factors for Choi progressive disease (as best response). Conclusions: High NLR and RDW values were independent biomarkers of worse outcome in advanced SFT patients treated with pazopanib.[Table: see text]

Forecasting US movies box office performances in Turkey using machine learning algorithms

2020 ◽  
Vol 39 (5) ◽  
pp. 6579-6590
Author(s):  
Sandy Çağlıyor ◽  
Başar Öztayşi ◽  
Selime Sezgin
Keyword(s):  
Machine Learning ◽  
Global Economy ◽  
Learning Algorithms ◽  
Forecast Model ◽  
Machine Learning Algorithms ◽  
Gradient Boosting ◽  
High Stakes ◽  
Box Office ◽  
Industry Forecast ◽  
The Impact

The motion picture industry is one of the largest industries worldwide and has significant importance in the global economy. Considering the high stakes and high risks in the industry, forecast models and decision support systems are gaining importance. Several attempts have been made to estimate the theatrical performance of a movie before or at the early stages of its release. Nevertheless, these models are mostly used for predicting domestic performances and the industry still struggles to predict box office performances in overseas markets. In this study, the aim is to design a forecast model using different machine learning algorithms to estimate the theatrical success of US movies in Turkey. From various sources, a dataset of 1559 movies is constructed. Firstly, independent variables are grouped as pre-release, distributor type, and international distribution based on their characteristic. The number of attendances is discretized into three classes. Four popular machine learning algorithms, artificial neural networks, decision tree regression and gradient boosting tree and random forest are employed, and the impact of each group is observed by compared by the performance models. Then the number of target classes is increased into five and eight and results are compared with the previously developed models in the literature.

Presepsin - New Marker of Sepsis Romanian Neonatal Intensive Care Unit Experience

2019 ◽  
Vol 70 (8) ◽  
pp. 3008-3013
Author(s):  
Silvia Maria Stoicescu ◽  
Ramona Mohora ◽  
Monica Luminos ◽  
Madalina Maria Merisescu ◽  
Gheorghita Jugulete ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Neonatal Intensive Care Unit ◽  
Neonatal Sepsis ◽  
Predictive Value ◽  
Neonatal Intensive Care ◽  
Complete Blood Count ◽  
Clinical Manifestations ◽  
Receiver Operating Curve ◽  
Predictive Values

Difficulties in establishing the onset of neonatal sepsis has directed the medical research in recent years to the possibility of identifying early biological markers of diagnosis. Overdiagnosing neonatal sepsis leads to a higher rate and duration in the usage of antibiotics in the Neonatal Intensive Care Unit (NICU), which in term leads to a rise in bacterial resistance, antibiotherapy complications, duration of hospitalization and costs.Concomitant analysis of CRP (C Reactive Protein), procalcitonin, complete blood count, presepsin in newborn babies with suspicion of early or late neonatal sepsis. Presepsin sensibility and specificity in diagnosing neonatal sepsis. The study group consists of newborns admitted to Polizu Neonatology Clinic between 15th February- 15th July 2017, with suspected neonatal sepsis. We analyzed: clinical manifestations and biochemical markers values used for diagnosis of sepsis, namely the value of CRP, presepsin and procalcitonin on the onset day of the disease and later, according to evolution. CRP values may be influenced by clinical pathology. Procalcitonin values were mainly influenced by the presence of jaundice. Presepsin is the biochemical marker with the fastest predictive values of positive infection. Presepsin can be a useful tool for early diagnosis of neonatal sepsis and can guide the antibiotic treatment. Presepsin value is significantly higher in neonatal sepsis compared to healthy newborns (939 vs 368 ng/mL, p [ 0.0001); area under receiver operating curve (AUC) for presepsine was 0.931 (95% confidence interval 0.86-1.0). PSP has a greater sensibility and specificity compared to classical sepsis markers, CRP and PCT respectively (AUC 0.931 vs 0.857 vs 0.819, p [ 0.001). The cut off value for presepsin was established at 538 ng/mLwith a sensibility of 79.5% and a specificity of 87.2 %. The positive predictive value (PPV) is 83.8 % and negative predictive value (NPV) is 83.3%.

Sign in / Sign up

Export Citation Format

Share Document