Dataset Information

Use of Fetal Magnetic Resonance Image Analysis and Machine Learning to Predict the Need for Postnatal Cerebrospinal Fluid Diversion in Fetal Ventriculomegaly.

ABSTRACT: Importance:Which children with fetal ventriculomegaly, or enlargement of the cerebral ventricles in utero, will develop hydrocephalus requiring treatment after birth is unclear. Objective:To determine whether extraction of multiple imaging features from fetal magnetic resonance imaging (MRI) and integration using machine learning techniques can predict which patients require postnatal cerebrospinal fluid (CSF) diversion after birth. Design, Setting, and Patients:This retrospective case-control study used an institutional database of 253 patients with fetal ventriculomegaly from January 1, 2008, through December 31, 2014, to generate a predictive model. Data were analyzed from January 1, 2008, through December 31, 2015. All 25 patients who required postnatal CSF diversion were selected and matched by gestational age with 25 patients with fetal ventriculomegaly who did not require CSF diversion (discovery cohort). The model was applied to a sample of 24 consecutive patients with fetal ventriculomegaly who underwent evaluation at a separate institution (replication cohort) from January 1, 1998, through December 31, 2007. Data were analyzed from January 1, 1998, through December 31, 2009. Exposures:To generate the model, linear measurements, area, volume, and morphologic features were extracted from the fetal MRI, and a machine learning algorithm analyzed multiple features simultaneously to find the combination that was most predictive of the need for postnatal CSF diversion. Main Outcomes and Measures:Accuracy, sensitivity, and specificity of the model in correctly classifying patients requiring postnatal CSF diversion. Results:A total of 74 patients (41 girls [55%] and 33 boys [45%]; mean [SD] gestational age, 27.0 [5.6] months) were included from both cohorts. In the discovery cohort, median time to CSF diversion was 6 days (interquartile range [IQR], 2-51 days), and patients with fetal ventriculomegaly who did not develop symptoms were followed up for a median of 29 months (IQR, 9-46 months). The model correctly classified patients who required CSF diversion with 82% accuracy, 80% sensitivity, and 84% specificity. In the replication cohort, the model achieved 91% accuracy, 75% sensitivity, and 95% specificity. Conclusion and Relevance:Image analysis and machine learning can be applied to fetal MRI findings to predict the need for postnatal CSF diversion. The model provides prognostic information that may guide clinical management and select candidates for potential fetal surgical intervention.

SUBMITTER: Pisapia JM

PROVIDER: S-EPMC5796744 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ImportanceDiversion of cerebrospinal fluid (CSF) has been used for decades as a treatment for children with severe traumatic brain injury (TBI) and is recommended by evidenced-based guidelines. However, these recommendations are based on limited studies.ObjectiveTo determine whether CSF diversion is associated with improved Glasgow Outcome Score-Extended for Pediatrics (GOS-EP) and decreased intracranial pressure (ICP) in children with severe TBI.Design, setting, and participantsThis observational comparative effectiveness study was performed at 51 clinical centers that routinely care for children with severe TBI in 8 countries (US, United Kingdom, Spain, the Netherlands, Australia, New Zealand, South Africa, and India) from February 2014 to September 2017, with follow-up at 6 months after injury (final follow-up, October 22, 2021). Children with severe TBI were included if they had Glasgow Coma Scale (GCS) scores of 8 or lower, had intracranial pressure (ICP) monitor placed on-site, and were aged younger than 18 years. Children were excluded if they were pregnant or an ICP monitor was not placed at the study site. Consecutive children were screened and enrolled, data regarding treatments were collected, and at discharge, consent was obtained for outcomes testing. Propensity matching for pretreatment characteristics was performed to develop matched pairs for primary analysis. Data analyses were completed on April 18, 2022.ExposuresClinical care followed local standards, including the use of CSF diversion (or not), with patients stratified at the time of ICP monitor placement (CSF group vs no CSF group).Main outcomes and measuresThe primary outcome was GOS-EP at 6 months, while ICP was considered as a secondary outcome. CSF vs no CSF was treated as an intention-to-treat analysis, and a sensitivity analysis was performed for children who received delayed CSF diversion.ResultsA total of 1000 children with TBI were enrolled, including 314 who received CSF diversion (mean [SD] age, 7.18 [5.45] years; 208 [66.2%] boys) and 686 who did not (mean [SD] age, 7.79 [5.33] years; 437 [63.7%] boys). The propensity-matched analysis included 98 pairs. In propensity score-matched analyses, there was no difference between groups in GOS-EP (median [IQR] difference, 0 [-3 to 1]; P = .08), but there was a decrease in overall ICP in the CSF group (mean [SD] difference, 3.97 [0.12] mm Hg; P < .001).Conclusions and relevanceIn this comparative effectiveness study, CSF diversion was not associated with improved outcome at 6 months after TBI, but a decrease in ICP was observed. Given the higher quality of evidence generated by this study, current evidence-based guidelines related to CSF diversion should be reconsidered.

Project description:BackgroundChromosomal abnormalities are important causes of ventriculomegaly (VM). In mild and isolated cases of fetal VM, obstetricians rarely give clear indications for pregnancy termination. We aimed to calculate the incidence of chromosomal abnormalities and incremental yield of chromosomal microarray analysis (CMA) in VM, providing more information on genetic counseling and prognostic evaluation for fetuses with VM.MethodsThe Chinese language databases Wanfang Data, China National Knowledge Infrastructure, and China Biomedical Literature Database (from January 1, 1991 to April 29, 2020) and English language databases PubMed, Embase, and Cochrane Library (from January 1, 1945 to April 29, 2020) were systematically searched for articles on fetal VM. Diagnostic criteria were based on ultrasonographic or magnetic resonance imaging (MRI) assessment of lateral ventricular atrium width: ≥10 to <15 mm for mild VM, and ≥15 mm for severe VM. Isolated VM was defined by the absence of structural abnormalities other than VM detected by ultrasonography or MRI. R software was used for the meta-analysis to determine the incidence of chromosomal abnormalities and incremental yield of CMA in VM, and the combined rate and 95% confidence interval (CI) were calculated.ResultsTwenty-three articles involving 1635 patients were included. The incidence of chromosomal abnormalities in VM was 9% (95% CI: 5%-12%) and incremental yield of CMA in VM was 11% (95% CI: 7%-16%). The incidences of chromosomal abnormalities in mild, severe, isolated, and non-isolated VM were 9% (95% CI: 4%-16%), 5% (95% CI: 1%-11%), 3% (95% CI: 1%-6%), and 13% (95% CI: 4%-25%), respectively.ConclusionsApplying CMA in VM improved the detection rate of abnormalities. When VM is confirmed by ultrasound or MRI, obstetricians should recommend fetal karyotype analysis to exclude chromosomal abnormalities. Moreover, CMA should be recommended preferentially in pregnant women with fetal VM who are undergoing invasive prenatal diagnosis. CMA cannot completely replace chromosome karyotype analysis.

Dataset Information

Use of Fetal Magnetic Resonance Image Analysis and Machine Learning to Predict the Need for Postnatal Cerebrospinal Fluid Diversion in Fetal Ventriculomegaly.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets