Classification algorithm for congenital Zika Syndrome: characterizations, diagnosis and validation

Rafael V Veiga,Maurício L Barreto,Larissa C Costa,Wanderson K Oliveira,Maria Glória Teixeira,Lavinia Schuler-Faccini,Maria Da Conceição N Costa,Luciana L Cardim,Enny S Paixão,Giovanny V A França,Moreno S Rodrigues,Juliane F Oliveira,Roberto F S Andrade

doi:10.1038/s41598-021-86361-5

Abstract

Zika virus was responsible for the microcephaly epidemic in Brazil which began in October 2015 and brought great challenges to the scientific community and health professionals in terms of diagnosis and classification. Due to the difficulties in correctly identifying Zika cases, it is necessary to develop an automatic procedure to classify the probability of a CZS case from the clinical data. This work presents a machine learning algorithm capable of achieving this from structured and unstructured available data. The proposed algorithm reached 83% accuracy with textual information in medical records and image reports and 76% accuracy in classifying data without textual information. Therefore, the proposed algorithm has the potential to classify CZS cases in order to clarify the real effects of this epidemic, as well as to contribute to health surveillance in monitoring possible future epidemics.

Highlights

Zika virus was responsible for the microcephaly epidemic in Brazil which began in October 2015 and brought great challenges to the scientific community and health professionals in terms of diagnosis and classification
The Zika virus (ZIKV) related microcephaly epidemic that occurred in several cities of Brazil starting in October 20151–3, posed a series of challenges for health managers and the scientific community
Considering that a team of specialists from different areas of health manually classified 1501 cases of Congenital Zika Syndrome (CZS) reported in the RESP by February 26, 2016, our objective was the development of ML algorithms to classify suspected cases of CZS from that database, using as a gold standard the classification developed by these specialists

Summary

Introduction

Zika virus was responsible for the microcephaly epidemic in Brazil which began in October 2015 and brought great challenges to the scientific community and health professionals in terms of diagnosis and classification. The Zika virus (ZIKV) related microcephaly epidemic that occurred in several cities of Brazil starting in October 20151–3, posed a series of challenges for health managers and the scientific community. This includes case classification and diagnosis certainty for the clinical spectrum of Congenital Zika Syndrome (CZS)[4,5]. The focus of expert systems based on ML is not to identify the rules made by human specialists, but to obtain a decision as close as possible to that of a human specialists For this purpose, the learning algorithms usually use simple and complex relations like conditional and nonlinear of the predictors to reach their decision. In spite of the fact that ML techniques often generate models that are difficult to interpret, several health fields have already benefited from this approach such as cancer identification[7,8], diagnosis by medical image9,10, epidemiology11, genetics[12], medical diagnostic[13], among others

Objectives

Methods

Results

Conclusion