Objective and Automated Detection of Diffuse White Matter Abnormality in Preterm Infants Using Deep Convolutional Neural Networks

Structural connectivity at term equivalent age and language in preterm children at 2 years corrected

We previously reported interhemispheric structural hyperconnectivity bypassing the corpus callosum in children born extremely preterm (<28 weeks) versus term children. This increased connectivity was positively associated with language performance at 4-6 years of age in our prior work. In the present study, we aim to investigate whether this extracallosal connectivity develops in extremely preterm infants at term equivalent age by leveraging a prospective cohort study of 350 very and extremely preterm infants followed longitudinally in the Cincinnati Infant Neurodevelopment Early Prediction Study. For this secondary analysis, we included only children born extremely preterm and without significant brain injury (n = 95). We use higher-order diffusion modelling to assess the degree to which extracallosal pathways are present in extremely preterm infants and predictive of later language scores at 22-26 months corrected age. We compare results obtained from two higher-order diffusion models: generalized q-sampling imaging and constrained spherical deconvolution. Advanced MRI was obtained at term equivalent age (39-44 weeks post-menstrual age). For structural connectometry analysis, we assessed the level of correlation between white matter connectivity at the whole-brain level at term equivalent age and language scores at 2 years corrected age, controlling for post-menstrual age, sex, brain abnormality score and social risk. For our constrained spherical deconvolution analyses, we performed connectivity-based fixel enhancement, using probabilistic tractography to inform statistical testing of the hypothesis that fibre metrics at term equivalent age relate to language scores at 2 years corrected age after adjusting for covariates. Ninety-five infants were extremely preterm with no significant brain injury. Of these, 53 had complete neurodevelopmental and imaging data sets that passed quality control. In the connectometry analyses adjusted for covariates and multiple comparisons (P < 0.05), the following tracks were inversely correlated with language: bilateral cerebellar white matter and middle cerebellar peduncles, bilateral corticospinal tracks, posterior commissure and the posterior inferior fronto-occipital fasciculus. No tracks from the constrained spherical deconvolution/connectivity-based fixel enhancement analyses remained significant after correction for multiple comparisons. Our findings provide critical information about the ontogeny of structural brain networks supporting language in extremely preterm children. Greater connectivity in more posterior tracks that include the cerebellum and connections to the regions of the temporal lobes at term equivalent age appears to be disadvantageous for language development.

Diffuse excessive high signal intensity in the preterm brain on advanced MRI represents widespread neuropathology

Preterm brains commonly exhibit elevated signal intensity in the white matter on T2-weighted MRI at term-equivalent age. This signal, known as diffuse excessive high signal intensity (DEHSI) or diffuse white matter abnormality (DWMA) when quantitatively assessed, is associated with abnormal microstructure on diffusion tensor imaging. However, postmortem data are largely lacking and difficult to obtain, and the pathological significance of DEHSI remains in question. In a cohort of 202 infants born preterm at ≤32 weeks gestational age, we leveraged two newer diffusion MRI models - Constrained Spherical Deconvolution (CSD) and neurite orientation dispersion and density index (NODDI) - to better characterize the macro and microstructural properties of DWMA and inform the ongoing debate around the clinical significance of DWMA. With increasing DWMA volume, fiber density broadly decreased throughout the white matter and fiber cross-section decreased in the major sensorimotor tracts. Neurite orientation dispersion decreased in the centrum semiovale, corona radiata, and temporal lobe. These findings provide insight into DWMA's biological underpinnings and demonstrate that it is a serious pathology.

Assessment of germinal matrix hemorrhage on head ultrasound with deep learning algorithms

BACKGROUND

Germinal matrix hemorrhage-intraventricular hemorrhage is among the most common intracranial complications in premature infants. Early detection is important to guide clinical management for improved patient prognosis.

OBJECTIVE

The purpose of this study was to assess whether a convolutional neural network (CNN) can be trained via transfer learning to accurately diagnose germinal matrix hemorrhage on head ultrasound.

MATERIALS AND METHODS

Over a 10-year period, 400 head ultrasounds performed in patients ages 6 months or younger were reviewed. Key sagittal images at the level of the caudothalamic groove were obtained from 200 patients with germinal matrix hemorrhage and 200 patients without hemorrhage; all images were reviewed by a board-certified pediatric radiologist. One hundred cases were randomly allocated from the total for validation and an additional 100 for testing of a CNN binary classifier. Transfer learning and data augmentation were used to train the model.

RESULTS

The median age of patients was 0 weeks old with a median gestational age of 30 weeks. The final trained CNN model had a receiver operating characteristic area under the curve of 0.92 on the validation set and accuracy of 0.875 on the test set, with 95% confidence intervals of [0.86, 0.98] and [0.81, 0.94], respectively.

CONCLUSION

A CNN trained on a small set of images with data augmentation can detect germinal matrix hemorrhage on head ultrasounds with strong accuracy.

Extracallosal Structural Connectivity Is Positively Associated With Language Performance in Well-Performing Children Born Extremely Preterm

Children born extremely preterm (<28 weeks gestation) are at risk for language delay or disorders. Decreased structural connectivity in preterm children has been associated with poor language outcome. Previously, we used multimodal imaging techniques to demonstrate that increased functional connectivity during a stories listening task was positively associated with language scores for preterm children. This functional connectivity was supported by extracallosal structural hyperconnectivity when compared to term-born children. Here, we attempt to validate this finding in a distinct cohort of well-performing extremely preterm children (EPT, n = 16) vs. term comparisons (TC, n = 28) and also compare this to structural connectivity in a group of extremely preterm children with a history of language delay or disorder (EPT-HLD, n = 8). All participants are 4-6 years of age. We perform q-space diffeomorphic reconstruction and functionally-constrained structural connectometry (based on fMRI activation), including a novel extension enabling between-groups comparisons with non-parametric ANOVA. There were no significant differences between groups in age, sex, race, ethnicity, parental education, family income, or language scores. For EPT, tracks positively associated with language scores included the bilateral posterior inferior fronto-occipital fasciculi and bilateral cerebellar peduncles and additional cerebellar white matter. Quantitative anisotropy in these pathways accounted for 55% of the variance in standardized language scores for the EPT group specifically. Future work will expand this cohort and follow longitudinally to investigate the impact of environmental factors on developing language networks and resiliency in the preterm brain.

Deep Multimodal Learning From MRI and Clinical Data for Early Prediction of Neurodevelopmental Deficits in Very Preterm Infants

The prevalence of disabled survivors of prematurity has increased dramatically in the past 3 decades. These survivors, especially, very preterm infants (VPIs), born ≤ 32 weeks gestational age, are at high risk for neurodevelopmental impairments. Early and clinically effective personalized prediction of outcomes, which forms the basis for early treatment decisions, is urgently needed during the peak neuroplasticity window—the first couple of years after birth—for at-risk infants, when intervention is likely to be most effective. Advances in MRI enable the noninvasive visualization of infants' brains through acquired multimodal images, which are more informative than unimodal MRI data by providing complementary/supplementary depicting of brain tissue characteristics and pathology. Thus, analyzing quantitative multimodal MRI features affords unique opportunities to study early postnatal brain development and neurodevelopmental outcome prediction in VPIs. In this study, we investigated the predictive power of multimodal MRI data, including T2-weighted anatomical MRI, diffusion tensor imaging, resting-state functional MRI, and clinical data for the prediction of neurodevelopmental deficits. We hypothesize that integrating multimodal MRI and clinical data improves the prediction over using each individual data modality. Employing the aforementioned multimodal data, we proposed novel end-to-end deep multimodal models to predict neurodevelopmental (i.e., cognitive, language, and motor) deficits independently at 2 years corrected age. We found that the proposed models can predict cognitive, language, and motor deficits at 2 years corrected age with an accuracy of 88.4, 87.2, and 86.7%, respectively, significantly better than using individual data modalities. This current study can be considered as proof-of-concept. A larger study with external validation is important to validate our approach to further assess its clinical utility and overall generalizability.

Automatic Segmentation of Diffuse White Matter Abnormality on T2-weighted Brain MR Images Using Deep Learning in Very Preterm Infants

About 50%-80% of very preterm infants (VPIs) (≤ 32 weeks gestational age) exhibit diffuse white matter abnormality (DWMA) on their MR images at term-equivalent age. It remains unknown if DWMA is associated with developmental impairments, and further study is warranted. To aid in the assessment of DWMA, a deep learning model for DWMA quantification on T2-weighted MR images was developed. This secondary analysis of prospective data was performed with an internal cohort of 98 VPIs (data collected from December 2014 to April 2016) and an external cohort of 28 VPIs (data collected from January 2012 to August 2014) who had already undergone MRI at term-equivalent age. Ground truth DWMA regions were manually annotated by two human experts with the guidance of a prior published semiautomated algorithm. In a twofold cross-validation experiment using the internal cohort of 98 infants, the three-dimensional (3D) ResU-Net model accurately segmented DWMA with a Dice similarity coefficient of 0.907 ± 0.041 (standard deviation) and balanced accuracy of 96.0% ± 2.1, outperforming multiple peer deep learning models. The 3D ResU-Net model that was trained with the whole internal cohort (n = 98) was further tested on an independent external test cohort (n = 28) and achieved a Dice similarity coefficient of 0.877 ± 0.059 and balanced accuracy of 92.3% ± 3.9. The externally validated 3D ResU-Net deep learning model for accurately segmenting DWMA may facilitate the clinical diagnosis of DWMA in VPIs. Supplemental material is available for this article. Keywords: Brain/Brain Stem, Convolutional Neural Network (CNN), MR-Imaging, Pediatrics, Segmentation, Supervised learning © RSNA, 2021.

Diffuse excessive high signal intensity on term equivalent MRI does not predict disability: a systematic review and meta-analysis

OBJECTIVE

To evaluate whether diffuse excessive high signal intensity (DEHSI) on term equivalent age MRI (TEA-MRI) predicts disability in preterm infants.

DESIGN

This is a systematic review and meta-analysis. Medline, EMBASE, Cochrane Library, EMCARE, Google Scholar and MedNar databases were searched in July 2019. Studies comparing developmental outcomes of isolated DEHSI on TEA-MRI versus normal TEA-MRI were included. Two reviewers independently extracted data and assessed the risk of bias. Meta-analysis was undertaken where data were available in a format suitable for pooling.

MAIN OUTCOME MEASURES

Neurodevelopmental outcomes ≥1 year of corrected age based on validated tools.

RESULTS

A total of 15 studies (n=1832) were included, of which data from 9 studies were available for meta-analysis. The pooled estimate (n=7) for sensitivity of DEHSI in predicting cognitive/mental disability was 0.58 (95% CI 0.34 to 0.79) and for specificity was 0.46 (95% CI 0.20 to 0.74). The summary area under the receiver operating characteristics (ROC) curve was low at 0.54 (CI 0.50 to 0.58). A pooled diagnostic OR (DOR) of 1 indicated that DEHSI does not discriminate preterm infants with and without mental disability. The pooled estimate (n=8) for sensitivity of DEHSI in predicting cerebral palsy (CP) was 0.57 (95% CI 0.37 to 0.75) and for specificity was 0.41 (95% CI 0.24 to 0.62). The summary area under the ROC curve was low at 0.51 (CI 0.46 to 0.55). A pooled DOR of 1 indicated that DEHSI does not discriminate between preterm infants with and without CP.

CONCLUSIONS

DEHSI on TEA-MRI did not predict future development of cognitive/mental disabilities or CP.

PROSPERO REGISTRATION NUMBER

CRD42019130576.

Novel diffuse white matter abnormality biomarker at term-equivalent age enhances prediction of long-term motor development in very preterm children

Our objective was to evaluate the independent prognostic value of a novel MRI biomarker-objectively diagnosed diffuse white matter abnormality volume (DWMA; diffuse excessive high signal intensity)-for prediction of motor outcomes in very preterm infants. We prospectively enrolled a geographically-based cohort of very preterm infants without severe brain injury and born before 32 weeks gestational age. Structural brain MRI was obtained at term-equivalent age and DWMA volume was objectively quantified using a published validated algorithm. These results were compared with visually classified DWMA. We used multivariable linear regression to assess the value of DWMA volume, independent of known predictors, to predict motor development as assessed using the Bayley Scales of Infant & Toddler Development, Third Edition at 3 years of age. The mean (SD) gestational age of the cohort was 28.3 (2.4) weeks. In multivariable analyses, controlling for gestational age, sex, and abnormality on structural MRI, DWMA volume was an independent prognostic biomarker of Bayley Motor scores ([Formula: see text]= -12.59 [95% CI -18.70, -6.48] R² = 0.41). Conversely, visually classified DWMA was not predictive of motor development. In conclusion, objectively quantified DWMA is an independent prognostic biomarker of long-term motor development in very preterm infants and warrants further study.

Automated Detection of Juvenile Myoclonic Epilepsy using CNN based Transfer Learning in Diffusion MRI

Epilepsy is one of the largest neurological diseases in the world, and juvenile myoclonic epilepsy (JME) usually occurs in adolescents, giving patients tremendous burdens during growth, which really needs the early diagnosis. Advanced diffusion magnetic resonance imaging (MRI) could detect the subtle changes of the white matter, which could be a non-invasive early diagnosis biomarker for JME. Transfer learning can solve the problem of insufficient clinical samples, which could avoid overfitting and achieve a better detection effect. However, there is almost no research to detect JME combined with diffusion MRI and transfer learning. In this study, two advanced diffusion MRI methods, high angle resolved diffusion imaging (HARDI) and neurite orientation dispersion and density imaging (NODDI), were used to generate the connectivity matrix which can describe tiny changes in white matter. And three advanced convolutional neural networks (CNN) based transfer learning were applied to detect JME. A total of 30 participants (15 JME patients and 15 normal controls) were analyzed. Among the three CNN models, Inception_resnet_v2 based transfer learning is better at detecting JME than Inception_v3 and Inception_v4, indicating that the "short cut" connection can improve the ability to detect JME. Inception_resnet_v2 achieved to detect JME with the accuracy of 75.2% and the AUC of 0.839. The results support that diffusion MRI and CNN based transfer learning have the potential to improve the automated detection of JME.