Fronto-temporal horn ratio: yet another marker of ventriculomegaly?

Comprehensive Morphometric Analysis to Identify Key Neuroimaging Biomarkers for the Diagnosis of Adult Hydrocephalus Using Artificial Intelligence

BACKGROUND AND OBJECTIVES

Hydrocephalus involves abnormal cerebrospinal fluid accumulation in brain ventricles. Early and accurate diagnosis is crucial for timely intervention and preventing progressive neurological deterioration. The aim of this study was to identify key neuroimaging biomarkers for the diagnosis of hydrocephalus using artificial intelligence to develop practical and accurate diagnostic tools for neurosurgeons.

METHODS

Fifteen 1-dimensional (1-D) neuroimaging parameters and ventricular volume of adult patients with non-normal pressure hydrocephalus and healthy subjects were measured using manual image processing, and 10 morphometric indices were also calculated. The data set was analyzed using 8 machine, ensemble, and deep learning classifiers to predict hydrocephalus. SHapley Additive exPlanations (SHAP) feature importance analysis identified key neuroimaging diagnostic biomarkers.

RESULTS

Gradient Boosting achieved the highest performance, with an accuracy of 0.94 and an area under the curve of 0.97. SHAP analysis identified ventricular volume as the most important parameter. Given the challenges of measuring volume for clinicians, we identified key 1-D morphometric biomarkers that are easily measurable yet provide similar classifier performance. The results showed that the frontal-temporal horn ratio, modified Evan index, modified cella media index, sagittal maximum lateral ventricle height, and coronal posterior callosal angle are key 1-D diagnostic biomarkers. Notably, higher modified Evan index, modified cella media index, and sagittal maximum lateral ventricle height, and lower frontal-temporal horn ratio and coronal posterior callosal angle values were associated with hydrocephalus prediction. The results also elucidated the relationships between these key 1-D morphometric parameters and ventricular volume, providing potential diagnostic insights.

CONCLUSION

This study highlights the importance of a multifaceted diagnostic approach incorporating 5 easily measurable 1-D neuroimaging biomarkers for neurosurgeons to differentiate non-normal pressure hydrocephalus from healthy subjects. Incorporating our artificial intelligence model, interpreted through SHAP analysis, into routine clinical workflows may transform the diagnostic landscape for hydrocephalus by standardizing diagnosis and overcoming the limitations of visual evaluations, particularly in early stages and challenging cases.

Neuroimaging to guide neuroprognostication in the neonatal intensive care unit

PURPOSE OF REVIEW

Neurological problems are common in infants admitted to the neonatal intensive care unit (NICU). Various neuroimaging modalities are available for neonatal brain imaging and are selected based on presenting problem, timing and patient stability.

RECENT FINDINGS

Neuroimaging findings, taken together with clinical factors and serial neurological examination can be used to predict future neurodevelopmental outcomes. In this narrative review, we discuss neonatal neuroimaging modalities, and how these can be optimally utilized to assess infants in the NICU. We will review common patterns of brain injury and neurodevelopmental outcomes in hypoxic-ischemic encephalopathy, perinatal arterial ischemic stroke and preterm brain injury.

SUMMARY

Timely and accurate neuroprognostication can identify infants at risk for neurodevelopmental impairment and allow for early intervention and targeted therapies to improve outcomes.

Ventriculomegaly thresholds for prediction of symptomatic post-hemorrhagic ventricular dilatation in preterm infants

BACKGROUND

Benefits from early surgical intervention in preterm infants with intraventricular hemorrhage (IVH) prior to symptomatic ventriculomegaly must be weighed against risks of surgery. We calculated thresholds of common ventriculomegaly indices at a late-intervention institution to predict subsequent symptomatic ventriculomegaly requiring neurosurgery.

METHODS

We retrospectively reviewed neuroimaging and neurosurgical outcomes in preterm infants with grade III/IV IVH between 2007 and 2020. Frontal-occipital horn ratio (FOHR), frontal-temporal horn ratio (FTHR), anterior horn width (AHW), and ventricular index (VI) were measured. Area under the receiver operating curve (AUC) for predicting intervention (initiated after progressive symptomatic ventriculomegaly) was calculated for diagnostic scan, scans during weeks 1-4, and maximum measurement prior to intervention. Threshold values that optimized sensitivity and specificity were derived.

RESULTS

A total of 1254 scans in 132 patients were measured. In all, 37 patients had a neurosurgical intervention. All indices differed between those with and without intervention from the first diagnostic scan (p < 0.001). AUC of maximum measurement was 97.1% (95% CI 94.6-99.7) for FOHR, 97.7% (95% CI 95.6-99.8) for FTHR, 96.6% (95% CI 93.9-99.4) for AHW, and 96.8% (95% CI 94.0-99.5) for VI. Calculated thresholds were FOHR 0.66, FTHR 0.62, AHW 15.5 mm, and VI 8.4 mm > p97 (sensitivities >86.8%, specificities >90.1%).

CONCLUSION

Ventriculomegaly indices were greater for patients who developed progressive persistent ventriculomegaly from the first diagnostic scan and predicted neurosurgical intervention.

IMPACT

We derived thresholds of common ventriculomegaly indices (ventricular index, anterior frontal horn width, fronto-occipital horn and fronto-temporal horn index) to best predict the development of progressive symptomatic post-hemorrhage hydrocephalus in preterm infants with intraventricular hemorrhage. While current thresholds were established by a priori expert consensus, we report the first data-driven derivation of ventriculomegaly thresholds across all indices for the prediction of symptomatic hydrocephalus. Data-derived thresholds will more precisely weigh the risks and benefits of early intervention.

Degree of ventriculomegaly predicts school-aged functional outcomes in preterm infants with intraventricular hemorrhage

BACKGROUND

Greater ventriculomegaly in preterm infants with intraventricular hemorrhage (IVH) has been associated with worse neurodevelopmental outcomes in infancy. We aim to explore the relationship between ventriculomegaly and school-age functional outcome.

METHODS

Retrospective review of preterm infants with Grade III/IV IVH from 2006 to 2020. Frontal-occipital horn ratio (FOHR) was measured on imaging throughout hospitalization and last available follow-up scan. Pediatric Cerebral Performance Category (PCPC) scale was used to assess functional outcome at ≥4 years. Ordinal logistic regression was used to determine the relationship between functional outcome and FOHR at the time of Neurosurgery consult, neurosurgical intervention, and last follow-up scan while adjusting for confounders.

RESULTS

One hundred and thirty-four infants had Grade III/IV IVH. FOHR at consult was 0.62 ± 0.12 and 0.75 ± 0.13 at first intervention (p < 0.001). On univariable analysis, maximum FOHR, FOHR at the last follow-up scan, and at Neurosurgery consult predicted worse functional outcome (p < 0.01). PVL, longer hospital admission, and gastrotomy/tracheostomy tube also predicted worse outcome (p < 0.05). PVL, maximum FOHR, and FOHR at consult remained significant on multivariable analysis (p < 0.05). Maximum FOHR of 0.61 is a fair predictor for moderate-severe impairment (AUC 75%, 95% CI: 62-87%).

CONCLUSIONS

Greater ventricular dilatation and PVL were independently associated with worse functional outcome in Grade III/IV IVH regardless of neurosurgical intervention.

IMPACT

Ventriculomegaly measured by frontal-occipital horn ratio (FOHR) and periventricular leukomalacia are independent correlates of school-age functional outcomes in preterm infants with intraventricular hemorrhage regardless of need for neurosurgical intervention. These findings extend the known association between ventriculomegaly and neurodevelopmental outcomes in infancy to functional outcomes at school age. FOHR is a fair predictor of school-age functional outcome, but there are likely other factors that influence functional status, which highlights the need for prospective studies to incorporate other clinical and demographic variables in predictive models.