Dataset Information

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

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.

SUBMITTER: Veiga RV

PROVIDER: S-EPMC7990918 | biostudies-literature | 2021 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Veiga Rafael V RV Schuler-Faccini Lavinia L França Giovanny V A GVA Andrade Roberto F S RFS Teixeira Maria Glória MG Costa Larissa C LC Paixão Enny S ES Costa Maria da Conceição N MDCN Barreto Maurício L ML Oliveira Juliane F JF Oliveira Wanderson K WK Cardim Luciana L LL Rodrigues Moreno S MS

Scientific reports 20210324 1

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 unstructur ...[more]

PMID: 33762667

Similar Datasets

Project description:BackgroundChildren with in utero Zika virus (ZIKV) exposure without congenital Zika syndrome (CZS) are at risk for abnormal neurodevelopment. Preschool-age outcomes for children with antenatal ZIKV exposure have not yet been established.MethodsChildren with in utero ZIKV exposure and non-exposed controls had neurodevelopmental evaluations at age 3-5 years in Sabanalarga, Colombia. Cases did not have CZS and were previously evaluated prenatally through age 18 months. Controls were born before ZIKV arrival to Colombia. Neurodevelopmental assessments included Pediatric Evaluation of Disability Inventory (PEDI-CAT), Behavior Rating Inventory of Executive Function (BRIEF-P), Bracken School Readiness Assessment (BSRA), and Movement Assessment Battery for Children (MABC). Family demographics and child medical history were recorded.ResultsFifty-five ZIKV-exposed children were evaluated at mean age 3.6 years and 70 controls were evaluated at 5.2 years. Family demographics were similar between groups. BRIEF-P t-scores were higher for cases than controls in shift and flexibility domains. Cases had lower PEDI-CAT mobility t-scores compared to controls. There was no difference in MABC between groups. In 11% of cases and 1% of controls, parents reported child mood problems.ConclusionsChildren with in utero ZIKV exposure without CZS may demonstrate emerging differences in executive function, mood, and adaptive mobility that require continued evaluation.ImpactPreschool neurodevelopmental outcome in children with in utero Zika virus exposure is not yet known, since the Zika virus epidemic occurred in 2015-2017 and these children are only now entering school age. This study finds that Colombian children with in utero Zika virus exposure without congenital Zika syndrome are overall developing well but may have emerging differences in executive function, behavior and mood, and adaptive mobility compared to children without in utero Zika virus exposure. Children with in utero Zika virus exposure require continued multi-domain longitudinal neurodevelopmental evaluation through school age.

Project description:ObjectiveTo examine the relationship between prenatal diagnostics (ultrasound examination and amniotic fluid Zika virus testing) and postnatal congenital Zika syndrome abnormalities.Data sourcesSystematic searches were performed in 27 databases, including ClinicalTrials.gov, from inception to July 1, 2019, for articles with the keywords "Zika," "prenatal," "ultrasound," and "amniocentesis."Methods of study selectionA total of 3,049 unique records were identified. Two reviewers independently assessed titles, abstracts, and full texts for relevance; 84 articles met the inclusion criteria. These articles describe 402 mother-fetus or mother-neonate dyads; 385 were included in the review of prenatal ultrasound examination, and 56 in the review of amniocentesis (39 in both).Tabulation, integration, and resultsAmong 195 fetuses with congenital Zika syndrome findings on prenatal ultrasound examination, postnatal congenital Zika syndrome abnormalities were reported for 153 (78%; 95% CI 7-84%). High proportions of microcephaly (76%; 95% CI 69-82%) and brain abnormalities (78%; 95% CI 69-86%) were confirmed postnatally. Among 190 fetuses without congenital Zika syndrome findings on prenatal ultrasound examination, 17% (95% CI 12-24%) had congenital Zika syndrome abnormalities identified postnatally. Structural congenital Zika syndrome abnormalities were identified postnatally in approximately equal proportions among dyads with and without Zika virus RNA detected in an amniotic fluid specimen (68% and 67%; 95% CI 52-82% and 95% CI 38-88%). In six pregnancies, Zika virus RNA was detected in amniotic fluid but not in a subsequent amniocentesis specimen.ConclusionPrenatal ultrasound examination frequently detects structural findings associated with Zika virus infection; however, not all abnormalities are detected, and some may represent transient findings. As with other congenital infections, prenatal detection may vary with timing of infection, timing of ultrasound examination, technical expertise, and severity of abnormalities. The detection of Zika virus RNA in amniotic fluid in the included studies did not predict the risk for congenital Zika syndrome abnormalities in these cases, and clearance of Zika virus RNA from amniotic fluid appears possible after maternal infection. Diagnostic testing for Zika virus infection remains a shared decision between patients and clinicians, and more data are needed to define clinical predictors that will inform these decisions.Systematic review registrationPROSPERO, CRD42018080959.

Project description:ImportanceHydrocephalus is a treatable but potentially fatal complication that has not been previously described in congenital Zika syndrome (CZS).ObjectiveTo describe the clinical features and imaging findings in 24 patients with congenital Zika syndrome (CZS) who developed hydrocephalus.Design, setting, and participantsThis case series included patients with hydrocephalus who were born in October and November 2015 and followed up until mid-2017 in the 2 largest national referral centers for CZS in Brazil. The participants included consecutively enrolled children with a clinical and laboratorial diagnosis of CZS who developed clinical and/or image findings suggestive of hydrocephalus and who were confirmed to experience increased intracranial hypertension during ventriculoperitoneal shunt procedures.Main outcomes and measuresTo retrospectively describe clinical and image findings in these 24 patients.ResultsThis multicenter cohort included 308 patients with CZS; 24 consecutive children were enrolled in this study. These children were aged between 3 to 18 months, and 13 of 24 (54%) were female. All patients presented with at least 1 positive test result for anti-Zika antibodies in cerebrospinal fluid or serum and had classic signs of CZS. At the time of hydrocephalus diagnosis, only 14 of 24 patients (58%) had symptoms and signs suggestive of hydrocephalus (mainly worsening seizures, vomiting, irritability, and/or sudden increase of head circumference percentile). Two of 24 patients (8%) had no symptoms suggestive of hydrocephalus but were found to have reduced brain volume on repeated imaging. Cerebellar or brainstem hypoplasia on baseline imaging were found in 18 of 23 patients (78%). At the second computed tomographic scan, all patients showed a marked increase of ventricular volume, compatible with communicating hydrocephalus, and reduction of brain tissue that was visibly worse than on baseline imaging for the 23 patients with repeated scans.Conclusions and relevanceWe present evidence that hydrocephalus is a complication of CZS in at least a proportion of patients. The clinical spectrum of this condition continues to evolve, but given that presenting signs and symptoms of hydrocephalus can be challenging to recognize in CZS, we provisionally recommend that high suspicion and appropriate monitoring for hydrocephalus should be part of the standard care of patients with CZS.

Dataset Information

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

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets