Radiomics and machine learning for the diagnosis of pediatric cervical non-tuberculous mycobacterial lymphadenitis

Yarab Al Bulushi,Nikesh Muthukrishnan,Caroline Reinhold,Christine Saint-Martin,Farhad Maleki,Reza Forghani

doi:10.1038/s41598-022-06884-3

Abstract

Non-tuberculous mycobacterial (NTM) infection is an emerging infectious entity that often presents as lymphadenitis in the pediatric age group. Current practice involves invasive testing and excisional biopsy to diagnose NTM lymphadenitis. In this study, we performed a retrospective analysis of 249 lymph nodes selected from 143 CT scans of pediatric patients presenting with lymphadenopathy at the Montreal Children’s Hospital between 2005 and 2018. A Random Forest classifier was trained on the ten most discriminative features from a set of 1231 radiomic features. The model classifying nodes as pyogenic, NTM, reactive, or proliferative lymphadenopathy achieved an accuracy of 72%, a precision of 68%, and a recall of 70%. Between NTM and all other causes of lymphadenopathy, the model achieved an area under the curve (AUC) of 89%. Between NTM and pyogenic lymphadenitis, the model achieved an AUC of 90%. Between NTM and the reactive and proliferative lymphadenopathy groups, the model achieved an AUC of 93%. These results indicate that radiomics can achieve a high accuracy for classification of NTM lymphadenitis. Such a non-invasive highly accurate diagnostic approach has the potential to reduce the need for invasive procedures in the pediatric population.

Highlights

Non-tuberculous mycobacterial (NTM) infection is an emerging infectious entity that often presents as lymphadenitis in the pediatric age group
The classifiers designed to distinguish NTM, reactive, and proliferative lymphadenopathy achieved an average accuracy of 72%, precision of 68%, recall of 70%, F1-score of 67%, and area under the curve of 90%, when applied to samples in the test set
For distinction of NTM lymphadenitis from all other causes of lymphadenopathy, on the test set, the model achieved a precision of 65%, recall of 80%, accuracy of 82%, negative predictive value (NPV) of 91%, and area under the curve of 89%

Summary

Introduction

Non-tuberculous mycobacterial (NTM) infection is an emerging infectious entity that often presents as lymphadenitis in the pediatric age group. It is imperative to exclude other possible organisms, such as TB, to guarantee proper m anagement[6] Diagnostic approaches such as the Tuberculin Skin Test (TST), Fine Needle Aspiration, or excisional biopsy have limitations in terms of sensitivity and specificity as well as the inadvertent complications of invasive testing[5,6,18,21,22]. NTM lymphadenitis has been associated with hypoechogenicity and intranodal liquefaction on u ltrasound[19] These imaging findings are not considered specific; diagnostic models rely on the clinical and invasive test results[5,16,17,18,19,21,22,23,24,25].

Objectives

Methods

Results

Conclusion