Machine Learning-Based Feature Selection and Classification for the Experimental Diagnosis of Trypanosoma cruzi

Abstract

Chagas disease, caused by the Trypanosoma cruzi (T. cruzi) parasite, is the third most common parasitosis worldwide. Most of the infected subjects can remain asymptomatic without an opportune and early detection or an objective diagnostic is not conducted. Frequently, the disease manifests itself after a long time, accompanied by severe heart disease or by sudden death. Thus, the diagnosis is a complex and challenging process where several factors must be considered. In this paper, a novel pipeline is presented integrating temporal data from four modalities (electrocardiography signals, echocardiography images, Doppler spectrum, and ELISA antibody titers), multiple features selection analyses by a univariate analysis and a machine learning-based selection. The method includes an automatic dichotomous classification of animal status (control vs. infected) based on Random Forest, Extremely Randomized Trees, Decision Trees, and Support Vector Machine. The most relevant multimodal attributes found were ELISA (IgGT, IgG1, IgG2a), electrocardiography (SR mean, QT and ST intervals), ascending aorta Doppler signals, and echocardiography (left ventricle diameter during diastole). Concerning automatic classification from selected features, the best accuracy of control vs. acute infection groups was 93.3 ± 13.3% for cross-validation and 100% in the final test; for control vs. chronic infection groups, it was 100% and 100%, respectively. We conclude that the proposed machine learning-based approach can be of help to obtain a robust and objective diagnosis in early T. cruzi infection stages.

Cardiac Abnormalities in a Predictive Mouse Model of Chagas Disease

“…As a general trend, these abnormalities became less pronounced when the infection transitioned to the chronic stage (21-25 wpi) and the parasite burden had been reduced by the immune response. ST and QT show abnormalities in ventricular repolarisation, and HRV denotes changes in cardiac autonomic function [45]. The main ECG change reported in T. cruzi-infected C57BL/6 mice was an increase in the duration of the PR interval, reflecting a delay in electrical conduction through atrioventricular (AV) nodal conduction in chronically infected mice [46].…”

ECG Marker Evaluation for the Machine-Learning-Based Classification of Acute and Chronic Phases of Trypanosoma cruzi Infection in a Murine Model

“…The results obtained from the automatic classification ( Table 2 and Table 3 ) of Control 1 vs. the acute group (ACC up to 93.8% for CV and up to 87.5 during FT) and of Control 2 vs. mice with chronic infection (ACC up to 87.5% for CV and up to 66.6% during FT) appear to be competitive with that reported in the state of the art. In [ 21 ], using only ECG variables, performances of up to an ACC of 66.7% were obtained during CV and FT for the classification of controls vs. the acute groups. For the classification of control vs. the chronic groups, an ACC of up to 70% was obtained for CV and 100% during FT.…”

“…This algorithm has been selected because of its ability to assign the importance of biological features [ 11 , 23 , 24 ]. Furthermore, in a previous work in which various feature selection algorithms were compared, the MDI-based approach was one of the best performing [ 21 ]. To calculate the MDI, the Extremely Randomized Trees Classifier algorithm was used with a total of 500 decision trees.…”

“…They have also proposed the principal component analysis of average circadian profiles together with HRV indices [ 20 ]. In a previous work [ 21 ], the intervals and durations of some ECG complexes were analyzed in order to integrate them into a multimodal automatic classification strategy. However, in that work, the voltages and the characteristics of the P, J, and T waves were not analyzed.…”

ECG signal processing by feature selection with classification based on machine learning architectures in cardiovascular disease detection

“…They combined the common MIT -BIH arrhythmia database with a MATLAB-based FFNN. author who created the collection of temporal and amplitude features obtained from SIEMENS ECG equipment is found in [9].…”