Machine learning of cell population data, complete blood count, and differential count parameters for early prediction of bacteremia among adult patients with suspected bacterial infections and blood culture sampling in emergency departments

被引：5

作者：

Chang, Yu-Hsin ^{[1
,2
]}

Hsiao, Chiung-Tzu ^{[3
]}

Chang, Yu-Chang ^{[3
]}

Lai, Hsin-Yu ^{[3
]}

Lin, Hsiu-Hsien ^{[3
]}

Chen, Chien-Chih ^{[4
]}

Hsu, Lin-Chen ^{[5
]}

Wu, Shih-Yun ^{[6
]}

Shih, Hong-Mo ^{[1
,2
,7
,12
]}

Hsueh, Po-Ren ^{[2
,3
,8
,10
,11
]}

Cho, Der-Yang ^{[9
,13
]}

机构：

[1] China Med Univ Hosp, Dept Emergency Med, Taichung, Taiwan

[2] China Med Univ, Coll Med, Sch Med, Taichung, Taiwan

[3] China Med Univ Hosp, Dept Lab Med, Taichung, Taiwan

[4] Wei Gong Mem Hosp, Dept Lab, Miaoli, Taiwan

[5] China Med Univ, Annan Hosp, Dept Lab, Tainan, Taiwan

[6] Chang Gung Univ, Sch Med, Taoyuan, Taiwan

[7] China Med Univ, Dept Publ Hlth, Taichung, Taiwan

[8] China Med Univ, China Med Univ Hosp, Dept Internal Med, Div Infect Dis, Taichung, Taiwan

[9] China Med Univ Hosp, Dept Neurosurg, Taichung, Taiwan

[10] China Med Univ Hosp, Dept Lab Med, 2 Yude Rd Dist, Taichung 40447, Taiwan

[11] China Med Univ Hosp, Dept Internal Med, 2 Yude Rd Dist, Taichung 40447, Taiwan

[12] China Med Univ Hosp, Dept Emergency Med, 2 Yude Rd, Taichung 40447, Taiwan

[13] China Med Univ Hosp, Dept Neurosurg, 2 Yude Rd, Taichung 404332, Taiwan

来源：

JOURNAL OF MICROBIOLOGY IMMUNOLOGY AND INFECTION | 2023年 / 56卷 / 04期

关键词：

Bacteremia early prediction; Machine learning; Cell population data; Complete blood count; Cell differential count; LEUKOCYTES; DIAGNOSIS; ANALYZER; SUBSETS; SEPSIS; AREAS;

D O I：

10.1016/j.jmii.2023.05.001

中图分类号：

R392 [医学免疫学]; Q939.91 [免疫学];

学科分类号：

100102 ;

摘要：

Background: Bacteremia is a life-threatening complication of infectious diseases. Bacteremia can be predicted using machine learning (ML) models, but these models have not utilized cell population data (CPD).Methods: The derivation cohort from emergency department (ED) of China Medical University Hospital (CMUH) was used to develop the model and was prospectively validated in the same hospital. External validation was performed using cohorts from ED of Wei-Gong Memorial Hos-pital (WMH) and Tainan Municipal An-Nan Hospital (ANH). Adult patients who underwent com-plete blood count (CBC), differential count (DC), and blood culture tests were enrolled in the present study. The ML model was developed using CBC, DC, and CPD to predict bacteremia from positive blood cultures obtained within 4 h before or after the acquisition of CBC/DC blood samples.Results: This study included 20,636 patients from CMUH, 664 from WMH, and 1622 patients from ANH. Another 3143 patients were included in the prospective validation cohort of CMUH. The CatBoost model achieved an area under the receiver operating characteristic curve of 0.844 in the derivation cross-validation, 0.812 in the prospective validation, 0.844 in the WMH external validation, and 0.847 in the ANH external validation. The most valuable predic-tors of bacteremia in the CatBoost model were the mean conductivity of lymphocytes, nucle-ated red blood cell count, mean conductivity of monocytes, and neutrophil-to-lymphocyte ratio. Conclusions: ML model that incorporated CBC, DC, and CPD showed excellent performance in predicting bacteremia among adult patients with suspected bacterial infections and blood cul-ture sampling in emergency departments. Copyright 2023, Taiwan Society of Microbiology. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

引用

页码：782 / 792

页数：11

共 47 条

[1] Randomized Trial of Rapid Multiplex Polymerase Chain Reaction-Based Blood Culture Identification and Susceptibility Testing [J].

Banerjee, Ritu ;

Teng, Christine B. ;

Cunningham, Scott A. ;

Ihde, Sherry M. ;

Steckelberg, James M. ;

Moriarty, James P. ;

Shah, Nilay D. ;

Mandrekar, Jayawant N. ;

Patel, Robin .

CLINICAL INFECTIOUS DISEASES, 2015, 61 (07) :1071-1080

[2] CONTAMINANT BLOOD CULTURES AND RESOURCE UTILIZATION - THE TRUE CONSEQUENCES OF FALSE-POSITIVE RESULTS [J].

BATES, DW ;

GOLDMAN, L ;

LEE, TH .

JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION, 1991, 265 (03) :365-369

[3] Bacteremias: A leading cause of death [J].

Bearman, GML ;

Wenzel, RP .

ARCHIVES OF MEDICAL RESEARCH, 2005, 36 (06) :646-659

[4]

Bisong E., 2019, BUILDING MACHINE LEA, P59, DOI [10.1007/978-1-4842-4470-87, DOI 10.1007/978-1-4842-4470-8_7, 10.1007/978-1-4842-4470-8_19, DOI 10.1007/978-1-4842-4470-8_19]

[5] Random forests [J].