An AI-based auxiliary empirical antibiotic therapy model for children with bacterial pneumonia using low-dose chest CT images

Mudan Zhang,Siwei Yu,Rongpin Wang,Xiaoyong Zhang,Zhiyan Shen,Xianchun Zeng,Chencui Huang,Xuntao Yin,Xinfeng Liu

doi:10.1007/s11604-021-01136-2

Abstract

PurposeTo construct an auxiliary empirical antibiotic therapy (EAT) multi-class classification model for children with bacterial pneumonia using radiomics features based on artificial intelligence and low-dose chest CT images.Materials and methodsData were retrospectively collected from children with pathogen-confirmed bacterial pneumonia including Gram-positive bacterial pneumonia (122/389, 31%), Gram-negative bacterial pneumonia (159/389, 41%) and atypical bacterial pneumonia (108/389, 28%) from January 1 to June 30, 2019. Nine machine-learning models were separately evaluated based on radiomics features extracted from CT images; three optimal submodels were constructed and integrated to form a multi-class classification model.ResultsWe selected five features to develop three radiomics submodels: a Gram-positive model, a Gram-negative model and an atypical model. The comprehensive radiomics model using support vector machine method yielded an average area under the curve (AUC) of 0.75 [95% confidence interval (CI), 0.65–0.83] and accuracy (ACC) of 0.58 [sensitivity (SEN), 0.57; specificity (SPE), 0.78] in the training set, and an average AUC of 0.73 (95% CI 0.61–0.79) and ACC of 0.54 (SEN, 0.52; SPE, 0.75) in the test set.ConclusionThis auxiliary EAT radiomics multi-class classification model was deserved to be researched in differential diagnosing bacterial pneumonias in children.

Highlights

Community-acquired pneumonia (CAP) is a leading cause of childhood morbidity and mortality worldwide
We separately found five different radiomics features related to the multi-class classification of Gram-positive, Gram-negative and atypical bacterial pneumonias
We constructed and validated a comprehensive multi-class classification model for differential diagnosis of these three common types of bacterial pneumonia by synthesizing three radiomics submodels based on radiomics features extracted from low-dose chest CT images

Summary

Introduction

Community-acquired pneumonia (CAP) is a leading cause of childhood morbidity and mortality worldwide. According to 2016 Global Burden of Disease (GBD) data [2], approximately 64% of pneumonia deaths in children under 5 years old were bacterial in etiology. Life-threatening bacterial infections require immediate and precise antibiotic therapy. Because of the spread of antibiotic resistance, the altered microbial-community structure caused by antibiotic use and the fewer new antibacterials released in recent years due to the high costs involved, rational prescription of antibiotics for CAP is more important [3]. It is very difficult to precisely prescribe antibiotics to individual children with CAP based on bacterial pathogens that are proven by testing to be the cause of infection. Most early-stage antibiotic therapy for bacterial pneumonia is empirical; that is, it covers a range of possible target bacteria while culture results are pending

Methods

Results

Conclusion