Performance effectiveness of vital parameter combinations for early warning of sepsis—an exhaustive study using machine learning

Abstract

Objective To carry out exhaustive data-driven computations for the performance of noninvasive vital signs heart rate (HR), respiratory rate (RR), peripheral oxygen saturation (SpO2), and temperature (Temp), considered both independently and in all possible combinations, for early detection of sepsis. Materials and methods By extracting features interpretable by clinicians, we applied Gradient Boosted Decision Tree machine learning on a dataset of 2630 patients to build 240 models. Validation was performed on a geographically distinct dataset. Relative to onset, predictions were clocked as per 16 pairs of monitoring intervals and prediction times, and the outcomes were ranked. Results The combination of HR and Temp was found to be a minimal feature set yielding maximal predictability with area under receiver operating curve 0.94, sensitivity of 0.85, and specificity of 0.90. Whereas HR and RR each directly enhance prediction, the effects of SpO2 and Temp are significant only when combined with HR or RR. In benchmarking relative to standard methods Systemic Inflammatory Response Syndrome (SIRS), National Early Warning Score (NEWS), and quick-Sequential Organ Failure Assessment (qSOFA), Vital-SEP outperformed all 3 of them. Conclusion It can be concluded that using intensive care unit data even 2 vital signs are adequate to predict sepsis upto 6 h in advance with promising accuracy comparable to standard scoring methods and other sepsis predictive tools reported in literature. Vital-SEP can be used for fast-track prediction especially in limited resource hospital settings where laboratory based hematologic or biochemical assays may be unavailable, inaccurate, or entail clinically inordinate delays. A prospective study is essential to determine the clinical impact of the proposed sepsis prediction model and evaluate other outcomes such as mortality and duration of hospital stay.

A scoping review of machine learning for sepsis prediction- feature engineering strategies and model performance: a step towards explainability

“…Further innovation in feature engineering was demonstrated through the development of second-order derived features and aggregate features [ 17 ], capturing complex relationships and condensing data into insightful metrics. This approach yielded a comprehensive set of 672 features, with 192 identified as unique, revealing the synergistic effect of body temperature and heart rate, among others, on sepsis prediction accuracy and lead time.…”

“…of Features No. of Final Features Top 10 Features Intensive care unit Meicheng Yang et al [ 3 ] Wrappers 168 20 ICULOS, Hospital Admission, Time, Temp, Fio2, Fio2_interval, Lactate, WBC, Creatinine, Unit 1, BUN Maximiliano Mollura et al [ 12 ] Embedded + Wrapper 75 30 SDPAT, SD_Ratio, PAT_HF, AVPAT, Vent_Flag, NN50, pNN50, AVSAP, Avg _ssr_hfn, DAP_VLF Xin Zhao et al [ 13 ] Embedded 40 25 Temp, O2Stat, Resp, BUN, Magnesium, HR, Potassium, Bilirubin_total, DBP, PTT, PH Ekanath Srihari Rangan et al [ 17 ] Feature Extraction (2nd order derived aggregate features) 672 240 HR, Temp (baseline),Respiration, Temp Variance,SP02,HR(baseline), SP02(Delta between 2 and baseline), Temp(between 4 and 3) Yu Bai et al [ 18 ] Unsupervised 27 27 APACHE_4,HC03_max,Lactate_Max,Lactate_Min,HC03_Min,Creatinine_Min, Albumin_Min, Creatinine_Max, Albumin_Max and Glucose_Min Zhengling He et al [ 19 ] Feature Extraction (LSTM) 82 82 Bilirubin_total, Creatinine, Fi02, HR, MAP, PaCo2, Platelets, RR, SBP, SIRS_Resp Everton Osnei Cesario et al [ 20 ] Embedded 16 16 Age, DBP, HR, SBP, RR, Blood Glucose, Admission Days, Temp, Gender, Surgical Procedure (for RF) Kim Huat Goh et al [ 21 ] Filter Method 100 Topics...…”

Improving diagnosis of early complications (<1 week) through continuous vital sign monitoring following oncological gastrointestinal surgical procedures

“…These parameters are most valuable in identifying clinical deterioration. 23,24 The patch was specifically employed for patients who had undergone major oncological abdominal surgery, given their elevated risk of complications and the potential cost-effectiveness associated with early detection of clinical deterioration. Early identification of complications may benefit the patient in terms of diagnosis and treatment.…”

“…In November 2021, the Sensium® wireless patch was introduced in a large teaching hospital due to its capability to measure heart rate, respiratory rate, and temperature. These parameters are most valuable in identifying clinical deterioration 23,24 . The patch was specifically employed for patients who had undergone major oncological abdominal surgery, given their elevated risk of complications and the potential cost‐effectiveness associated with early detection of clinical deterioration.…”