The role of MRI-based texture analysis to predict the severity of brain injury in neonates with perinatal asphyxia

Temporal dynamics of neonatal hypoxic-ischemic encephalopathy injuries on magnetic resonance imaging

Moderate to severe perinatal hypoxic-ischemic encephalopathy occurs in ~ 1 to 3/1000 live births in high-income countries and is associated with a significant risk of death or neurodevelopmental disability. Detailed assessment is important to help identify high-risk infants, to help families, and to support appropriate interventions. A wide range of monitoring tools is available to assess changes over time, including urine and blood biomarkers, neurological examination, and electroencephalography. At present, magnetic resonance imaging is unique as although it is expensive and not suited to monitoring the early evolution of hypoxic-ischemic encephalopathy by a week of life it can provide direct insight into the anatomical changes in the brain after hypoxic-ischemic encephalopathy and so offers strong prognostic information on the long-term outcome after hypoxic-ischemic encephalopathy. This review investigated the temporal dynamics of neonatal hypoxic-ischemic encephalopathy injuries, with a particular emphasis on exploring the correlation between the prognostic implications of magnetic resonance imaging scans in the first week of life and their relationship to long-term outcome prediction, particularly for infants treated with therapeutic hypothermia. A comprehensive literature search, from 2016 to 2024, identified 20 pertinent articles. This review highlights that while the optimal timing of magnetic resonance imaging scans is not clear, overall, it suggests that magnetic resonance imaging within the first week of life provides strong prognostic accuracy. Many challenges limit the timing consistency, particularly the need for intensive care and clinical monitoring. Conversely, although most reports examined the prognostic value of scans taken between 4 and 10 days after birth, there is evidence from small numbers of cases that, at times, brain injury may continue to evolve for weeks after birth. This suggests that in the future it will be important to explore a wider range of times after hypoxic-ischemic encephalopathy to fully understand the optimal timing for predicting long-term outcomes.

Texture-Based Classification of Fetal Growth Restriction From Intrauterine Neurosonographic Image

OBJECTIVE

Fetal growth restriction (FGR) is a condition where fetuses fail to reach their genetic potential for growth, posing a significant health challenge for newborns. The aim of this research was to explore the efficacy of texture-based analysis of neurosonographic images in identifying FGR in fetuses, which may provide a promising tool for early assessment of FGR.

METHODS

A retrospective analysis collected 100 intrauterine neurosonographic images from 50 FGR and 50 gestational age-appropriate fetuses. Using MaZda software, approximately 300 texture features were extracted from occipital white matter (OWM) and cerebellum of intrauterine neurosonographic images, respectively. Then 10 optimal features were separately selected by 3 algorithms, including the Fisher coefficient method, the method of minimizing classification error probability and average correlation coefficients, and the mutual information coefficient method. Further, the 10 statistically most significant features were selected from these sets to form the mixed feature set. After nonlinear discriminant analysis was performed to reduce feature dimensionality, the artificial neural network (ANN) classifier was conducted, respectively.

RESULTS

For OWM and cerebellum, a total of 11 and 14 statistically significant features were selected. When the mixed feature sets of OWM and cerebellum were applied to ANN classifier, classification accuracy were 90.00% (κ = 0.800; P < .001) and 93.00% (κ = 0.860; P < .001), and the receiver operating characteristic curve for identifying FGR showed an area under the curve of 0.82 and 0.87.

CONCLUSIONS

Texture analysis of fetal intrauterine neurosonographic images is a feasible and noninvasive strategy for evaluating FGR fetuses.

Evaluation of Basal Ganglia in Paediatric Patients With Primary Nephrotic Syndrome by Brain Magnetic Resonance Histogram Analysis

BACKGROUND

Primary nephrotic syndrome is an important cause of chronic renal failure in childhood. Important neuronal complications may develop during the disease.

AIMS

This study aims to demonstrate basal ganglia involvement in children with nephrotic syndrome by texture analysis.

METHODS

Brain MRI images of 22 paediatric patients with primary nephrotic syndrome and 40 healthy children of similar age groups were analysed. Brain MRI T2-weighted images were extracted from the thalamus, lentiform nucleus and nucleus caudatus and texture analysis was performed.

RESULTS

The images of 22 children with primary nephrotic syndrome and 40 children in the control group were evaluated. There were no notable distinctions identified in terms of age and gender between the patient and control groups (P value 0,410; 0,516, respectively). Accordingly, a significant difference was found between mean, 1.P, 10.P, 50.P, 90.P, 99.P values of histogram parameters obtained from thalamus (P values were 0.001; 0.000; 0.001; 0.002; 0.004; 0.009, respectively). A significant difference was found between mean, 1.P, 10.P, 50.P, 90.P, 99.P values of histogram parameters obtained from lentiform nuclei (P values were 0.031; 0.019; 0.006; 0.006; 0.003; 0.003; 0.001; 0.002, respectively). A significant difference was found between the mean, 1.P, 10.P, 50.P, 90.P, 99.P values of the histogram parameters obtained from the nucleus caudatus (P values 0,002; 0,005; 0,002; 0,002; 0,002; 0,003; 0,003, respectively).

CONCLUSION

Texture analysis may be helpful in demonstrating brain parenchymal involvement in paediatric patients with primary nephrotic syndrome by showing changes that are not recognised on conventional images.

Volumetric apparent diffusion coefficient histogram analysis in term neonatal asphyxia treated with hypothermia

OBJECTIVES

Our aim is to estimate the long-term neurological sequelae and prognosis in term neonatal asphyxia treated with hypothermia via volumetric apparent diffusion coefficient (ADC) map histogram analysis (HA).

METHODS

Brain MRI studies of 83 term neonates with asphyxia who received whole-body hypothermia treatment and examined between postnatal (PN) fourth and sixth days were retrospectively re-evaluated by 2 radiologists. Volumetric HA was performed for the areas frequently affected in deep and superficial asphyxia (thalamus, lentiform nucleus, posterior limb of internal capsule, corpus callosum forceps major, and perirolandic cortex-subcortical white matter) on ADC map. The quantitative ADC values were obtained separately for each region. Qualitative-visual (conventional) MRI findings were also re-evaluated. Neonates were examined neurodevelopmentally according to the Revised Brunet-Lezine scale. The distinguishability of long-term neurodevelopmental outcomes was statistically investigated.

RESULTS

With HA, the adverse neurodevelopmental outcomes could only be distinguished from mild-moderated impairment and normal development at the thalamus with 10th percentile ADC (P = .02 and P = .03, respectively) and ADCmin (P = .03 and P = .04, respectively). Also with the conventional MRI findings, adverse outcome could be distinguished from mild-moderated impairment (P = .04) and normal development (P = .04) via cytotoxic oedema of the thalamus, corpus striatum, and diffuse cerebral cortical.

CONCLUSION

The long-term adverse neurodevelopmental outcomes in newborns with asphyxia who received whole-body hypothermia treatment can be estimated similarly with volumetric ADC-HA and the conventional assessment of the ADC map.

ADVANCES IN KNOWLEDGE

This study compares early MRI ADC-HA with neurological sequelae in term newborns with asphyxia who received whole-body hypothermia treatment. We could not find any significant difference in predicting adverse neurological sequelae between the visual-qualitative evaluation of the ADC map and HA.

Clinical-functional correlation with brain volumetry in severe perinatal asphyxia: a case report

BACKGROUND

Hypoxic-ischemic encephalopathy (HIE) appears in neurological conditions where some brain areas are likely to be injured, such as deep grey matter, basal ganglia area, and white matter subcortical periventricular áreas. Moreover, modeling these brain areas in a newborn is challenging due to significant variability in the intensities associated with HIE conditions. This paper aims to evaluate functional measurements and 3D machine learning models of a given HIE case by correlating the affected brain areas with the pathophysiology and clinical neurodevelopmental.

CASE PRESENTATION

A comprehensive analysis of a term infant with perinatal asphyxia using longitudinal 3D brain information from Machine Learning Models is presented. The clinical analysis revealed the perinatal asphyxia diagnosis with APGAR <5 at 5 and 10 minutes, umbilical arterial pH of 7.0 BE of -21.2 mmol / L), neonatal seizures, and invasive ventilation mechanics. Therapeutic interventions: physical, occupational, and language neurodevelopmental therapies. Epilepsy treatment: vagus nerve stimulation, levetiracetam, and phenobarbital. Furthermore, the 3D analysis showed how the volume decreases due to age, exhibiting an increasing asymmetry between hemispheres. The results of the basal ganglia area showed that thalamus asymmetry, caudate, and putamen increase over time while globus pallidus decreases.

CLINICAL OUTCOMES

spastic cerebral palsy, microcephaly, treatment-refractory epilepsy.

CONCLUSIONS

Slight changes in the basal ganglia and cerebellum require 3D volumetry for detection, as standard MRI examinations cannot fully reveal their complex shape variations. Quantifying these subtle neurodevelopmental changes helps in understanding their clinical implications. Besides, neurophysiological evaluations can boost neuroplasticity in children with neurological sequelae by stimulating new neuronal connections.

Abnormal neonatal brain microstructure in gestational diabetes mellitus revealed by MRI texture analysis

To investigate the value of MRI texture analysis in evaluating the effect of gestational diabetes mellitus (GDM) on neonatal brain microstructure development, we retrospectively collected images of neonates undergoing head MRI scans, including a GDM group (N1 = 37) and a healthy control group (N2 = 34). MaZda texture analysis software was used to extract the texture features from different sequence images and perform dimensionality reduction, and then the texture features selected by the lowest misjudgement rate method were imported into SPSS software for statistical analysis. In our study, we found that GDM affects the development of the microstructure of the neonatal brain, and different combinations of texture features have different recognition performances, such as different sequences and different brain regions. As a consequence, texture analysis combining multiple conventional MRI sequences has a high recognition performance in revealing the abnormal development of the brain microstructure of neonates born of mothers with GDM.

Volumetric apparent diffusion coefficient (ADC) histogram analysis of the brain in paediatric patients with hypoxic ischaemic encephalopathy

Purpose

To evaluate the whole brain, hippocampus, thalamus, and lentiform nucleus by volumetric apparent diffusion coefficient (ADC) histogram analysis in paediatric patients with hypoxic-ischaemic encephalopathy (HIE).

Material and methods

This retrospective study included 25 patients with HIE and 50 patients as the control group. Diffusion-weighted imaging was obtained at b-values of 1000 mm2/s. The histogram parameters of ADC values, including the mean, minimum, maximum, 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles, as well as skewness, kurtosis, and variance were determined. The interclass correlation coefficient (ICC) was used to assess the inter-observer agreement.

Results

ADCmin, ADCmean, and ADCmax, as well as the 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles of ADC values for the HIE group were all lower than those of the control group (p < 0.001) in the volumetric histogram analysis of the hippocampus, thalamus, and lentiform nucleus. In the whole-brain histogram analysis, ADC min, and the 50th and 75th percentiles of ADC values did not differ significantly, while other parameters were lower in the HIE group. The ROC curve revealed that the ADC histogram parameters of the hippocampus provided the most accurate results for the diagnosis of HIE. The area under the curve (AUC) of the 95th percentile of ADC values was the highest (AUC = 0.915; cut-off 1.262 × 10-3 mm2/s; sensitivity 88% and specificity 84%).

Conclusions

Volumetric ADC histogram analysis of the whole brain, hippocampus, thalamus, and lentiform nucleus with b-values of 1000 mm2/s can serve as an imaging marker for determining HIE.

Can radiomics be used to detect hypoxic-ischemic encephalopathy in neonates without magnetic resonance imaging abnormalities?

BACKGROUND

No study has assessed normal magnetic resonance imaging (MRI) findings to predict potential brain injury in neonates with hypoxic-ischemic encephalopathy (HIE).

OBJECTIVE

We aimed to evaluate the efficacy of MRI-based radiomics models of the basal ganglia, thalami and deep medullary veins to differentiate between HIE and the absence of MRI abnormalities in neonates.

MATERIALS AND METHODS

In this study, we included 38  full-term neonates with HIE and normal MRI findings and 89 normal neonates. Radiomics features were extracted from T1-weighted images, T2-weighted images, diffusion-weighted imaging and susceptibility-weighted imaging (SWI). The different models were evaluated using receiver operating characteristic curve analysis. Clinical utility was evaluated using decision curve analysis.

RESULTS

The SWI model exhibited the best performance among the seven single-sequence models. For the training and validation cohorts, the area under the curves (AUCs) of the SWI model were 1.00 and 0.98, respectively. The combined nomogram model incorporating SWI Rad-scores and independent predictors of clinical characteristics was not able to distinguish HIE in patients without MRI abnormalities from the control group (AUC, 1.00). A high degree of fitting and favorable clinical utility was detected using the calibration curve with the Hosmer-Lemeshow test. Decision curve analysis was used for the SWI, clinical and combined nomogram models. The decision curve showed that the SWI and combined nomogram models had better predictive performance than the clinical model.

CONCLUSIONS

HIE can be detected in patients without MRI abnormalities using an MRI-based radiomics model. The SWI model performed better than the other models.

Predictive Value of MRI in Hypoxic-Ischemic Encephalopathy Treated with Therapeutic Hypothermia

Background: Hypoxic-ischemic encephalopathy (HIE) is a severe pathology, and no unique predictive biomarker has been identified. Our aims are to identify associations of perinatal and outcome parameters with morphological anomalies and ADC values from MRI. The secondary aims are to define a predictive ADC threshold value and detect ADC value fluctuations between MRIs acquired within 7 days (MR0) and at 1 year (MR1) of birth in relation to perinatal and outcome parameters. Methods: Fifty-one term children affected by moderate HIE treated with hypothermia and undergoing MRI0 and MRI1 were recruited. Brain MRIs were evaluated through the van Rooij score, while ADC maps were co-registered on a standardized cerebral surface, on which 29 ROIs were drawn. Statistical analysis was performed in Matlab, with the statistical significance value at 0.05. Results: ADC0 < ADC1 in the left and right thalami, left and right frontal white matter, right visual cortex, and the left dentate nucleus of children showing abnormal perinatal and neurodevelopmental parameters. At ROC analysis, the best prognostic ADC cut-off value was 1.535 mm2/s × 10−6 (sensitivity 80%, specificity 86%) in the right frontal white matter. ADC1 > ADC0 in the right visual cortex and left dentate nucleus, positively correlated with multiple abnormal perinatal and neurodevelopmental parameters. The van Rooij score was significantly higher in children presenting with sleep disorders. Conclusions: ADC values could be used as prognostic biomarkers to predict children’s neurodevelopmental outcomes. Further studies are needed to address these crucial topics and validate our results. Early and multidisciplinary perinatal evaluation and the subsequent re-assessment of children are pivotal to identify physical and neuropsychological disorders to guarantee early and tailored therapy.

A radiomics-based study of deep medullary veins in infants: Evaluation of neonatal brain injury with hypoxic-ischemic encephalopathy via susceptibility-weighted imaging

Objective

The deep medullary veins (DMVs) can be evaluated using susceptibility-weighted imaging (SWI). This study aimed to apply radiomic analysis of the DMVs to evaluate brain injury in neonatal patients with hypoxic-ischemic encephalopathy (HIE) using SWI.

Methods

This study included brain magnetic resonance imaging of 190 infants with HIE and 89 controls. All neonates were born at full-term (37+ weeks gestation). To include the DMVs in the regions of interest, manual drawings were performed. A Rad-score was constructed using least absolute shrinkage and selection operator (LASSO) regression to identify the optimal radiomic features. Nomograms were constructed by combining the Rad-score with a clinically independent factor. Receiver operating characteristic curve analysis was applied to evaluate the performance of the different models. Clinical utility was evaluated using a decision curve analysis.

Results

The combined nomogram model incorporating the Rad-score and clinical independent predictors, was better in predicting HIE (in the training cohort, the area under the curve was 0.97, and in the validation cohort, it was 0.95) and the neurologic outcomes after hypoxic-ischemic (in the training cohort, the area under the curve was 0.91, and in the validation cohort, it was 0.88).

Conclusion

Based on radiomic signatures and clinical indicators, we developed a combined nomogram model for evaluating neonatal brain injury associated with perinatal asphyxia.

Radiomics based of deep medullary veins on susceptibility-weighted imaging in infants: predicting the severity of brain injury of neonates with perinatal asphyxia

OBJECTIVE

This study aimed to apply radiomics analysis of the change of deep medullary veins (DMV) on susceptibility-weighted imaging (SWI), and to distinguish mild hypoxic-ischemic encephalopathy (HIE) from moderate-to-severe HIE in neonates.

METHODS

A total of 190 neonates with HIE (24 mild HIE and 166 moderate-to-severe HIE) were included in this study. All of them were born at 37 gestational weeks or later. The DMVs were manually included in the regions of interest (ROI). For the purpose of identifying optimal radiomics features and to construct Rad-scores, 1316 features were extracted. LASSO regression was used to identify the optimal radiomics features. Using the Red-score and the clinical independent factor, a nomogram was constructed. In order to evaluate the performance of the different models, receiver operating characteristic (ROC) curve analysis was applied. Decision curve analysis (DCA) was implemented to evaluate the clinical utility.

RESULTS

A total of 15 potential predictors were selected and contributed to Red-score construction. Compared with the radiomics model, the nomogram combined model incorporating Red-score and urea nitrogen did not better distinguish between the mild HIE and moderate-to-severe HIE group. For the training cohort, the AUC of the radiomics model and the combined nomogram model was 0.84 and 0.84. For the validation cohort, the AUC of the radiomics model and the combined nomogram model was 0.80 and 0.79, respectively. The addition of clinical characteristics to the nomogram failed to distinguish mild HIE from moderate-to-severe HIE group.

CONCLUSION

We developed a radiomics model and combined nomogram model as an indicator to distinguish mild HIE from moderate-to-severe HIE group.

Distribution of IntraThalamic Injury According to Nuclei and Vascular Territories in Children With Term Hypoxic-Ischemic Injury

BACKGROUND

Term hypoxic-ischemic injury (HII) on magnetic resonance imaging (MRI) is described as the basal ganglia thalamus [BGT], watershed [WS], or combined [BGT/WS] groups. We aimed to determine differences between HII groups in intrathalamic distribution.

METHODS

Delayed MRIs of children with HII and thalamic injury were reviewed. Custom tools were placed over T2-weighted and/or fluid-attenuated inversion recovery axial images to determine distribution of intrathalamic injury: (1) six subjective (whole/near-whole, central, anterior, posterior, lateral, medial); (2) four nuclear (anterior [AN], ventrolateral [VLN], medial [MN], and pulvinar [PN]); and (3) three arterial (thalamoperforating arteries [TPA], thalamogeniculate arteries [TGA], and posterior choroidal arteries [PCA]) locations. We compared the frequency of injury of the aforementioned intrathalamic locations between HII groups.

RESULTS

The 128 children (mean age at MRI 7.35 ± 3.6 years) comprised 41% (n = 53) BGT, 26% (n = 33) WS, and 33% (n = 42) BGT/WS. The VLN was the most frequent injured nuclear region (66%, n = 85), and the TGA (93%, n = 128) was the most frequent arterial region involved. VLN injury occurred more frequently in the BGT group (P < 0.001), PN in the WS group (P < 0.001), and AN (P < 0.001), MN (P < 0.001), PN (P = 0.001), and all nuclei together (P < 0.001) in the BGT/WS group. The combination of all vascular territories was significantly associated with BGT/WS (P < 0.001).

CONCLUSIONS

There are significant differences in intrathalamic nuclear and arterial injuries between the different types of HII.

Whole-volume ADC histogram of the brain as an image biomarker in evaluating disease severity of neonatal hypoxic-ischemic encephalopathy

Purpose

To examine the diagnostic significance of the apparent diffusion coefficient (ADC) histogram in quantifying neonatal hypoxic ischemic encephalopathy (HIE).

Methods

An analysis was conducted on the MRI data of 90 HIE patients, 49 in the moderate-to-severe group, and the other in the mild group. The 3D Slicer software was adopted to delineate the whole brain region as the region of interest, and 22 ADC histogram parameters were obtained. The interobserver consistency of the two radiologists was assessed by the interclass correlation coefficient (ICC). The difference in parameters (ICC > 0.80) between the two groups was compared by performing the independent sample t-test or the Mann–Whitney U test. In addition, an investigation was conducted on the correlation between parameters and the neonatal behavioral neurological assessment (NBNA) score. The ROC curve was adopted to assess the efficacy of the respective significant parameters. Furthermore, the binary logistic regression was employed to screen out the independent risk factors for determining the severity of HIE.

Results

The ADCmean, ADCmin, ADCmax,10th−70th, 90th percentile of ADC values of the moderate-to-severe group were smaller than those of the mild group, while the group's variance, skewness, kurtosis, heterogeneity, and mode-value were higher than those of the mild group (P < 0.05). All the mentioned parameters, the ADCmean, ADCmin, and 10th−70th and 90th percentile of ADC displayed positive correlations with the NBNA score, mode-value and ADCmax displayed no correlations with the NBNA score, the rest showed negative correlations with the NBNA score (P < 0.05). The area under the curve (AUC) of variance was the largest (AUC = 0.977; cut-off 972.5, sensitivity 95.1%; specificity 87.8%). According to the logistic regression analysis, skewness, kurtosis, variance, and heterogeneity were independent risk factors for determining the severity of HIE (OR > 1, P < 0.05).

Conclusions

The ADC histogram contributes to the HIE diagnosis and is capable of indicating the diffusion information of the brain objectively and quantitatively. It refers to a vital method for assessing the severity of HIE.