Yuli Wang

2024 Engineering Prize

Artificial Intelligence for Monitoring and Predicting Disease in Patients with Idiopathic Normal Pressure Hydrocephalus Using Imaging and Clinical Data

Yuli Wang, MS

Under the mentorship of Dr. Harrison Bai

PhD candidate, Yuli Wang developed an advanced artificial intelligence-based deep learning model to improve the accuracy of idiopathic normal pressure hydrocephalus (iNPH) diagnosis using medical imaging. This is done by creating segments of the brain’s ventricles, where CSF accumulates, which the model then analyzes the images for certain markers of iNPH. This method is more reliable than traditional approaches, especially in cases where the ventricles are enlarged.

Their project also sought to automate the calculation of the Evans Ratio, which is the ratio of the maximum width of the frontal horns of the lateral ventricles and the maximal internal diameter of the skull at the same level employed in axial CT and MRI images. The Evans Ratio is crucial in diagnosing iNPH, which in the project’s case, has achieved results that are on par with manual measurements by experts.

The goal for his research is to more accurately identify iNPH from other conditions with similar traits. The researchers are developing a broader deep learning model that combines imaging features and clinical data to distinguish iNPH from other similar conditions. These early and precise diagnoses are crucial for treatment, and to ensure better outcomes for intervention and reduce the chance of unnecessary treatments.