Software System to Predict Outcome of Hemorrhage in Neonates using Cranial Ultrasound
Researchers at Children's National Hospital have developed an automatic software system and model to predict brain injury in neonates using 2D cranial ultrasound. Using ventricle segmentation, this system enables the prediction of the outcome of intraventricular hemorrhage and post hemorrhagic hydrocephalus in premature neonates which are the most recognized reasons for brain injury in newborns.
Newborns born before 37 weeks of gestation are considered to be premature. Such premature neonates born with intraventricular hemorrhage (IVH) followed by hydrocephalus (post-hemorrhagic hydrocephalus, PHH) are at the highest risk to develop severe cerebral palsy and other adverse neurodevelopmental disorders. Cranial ultrasound is a non-invasive way to image the neonates for brain injury. The new software system segments cranial ultrasound images using computational analyses (shape analysis, machine learning) to offer the following advantages:
- Qualitative and quantitative evaluation of cranial ultrasound.
- Greater sensitivity in predicting the course of IVH and therefore brain injury.
- Early prediction allows for early intervention and hence improved outcomes in treating brain injury, cerebral palsy and other neurodevelopmental disorders.
- Useful to the entire care team with specialists from neurology, neonatology, radiology and neurosurgery involved in treating neonates with IVH and PHH.
- Applicable to diagnosing brain injury in neonates with IVH and PHH
- Potentially applicable to imaging other brain ventricular hemorrhages
Stage of Development
- Validated pre-clinical and human clinical data obtained.
Intellectual Property Status
- Patent Pending
This technology is available for exclusive or non-exclusive licensing.
Haiyin Chang, Ph.D.
- Senior Licensing Associate