Idiopathic Normal Pressure Hydrocephalus (INPH) is a syndrome affecting older adults, characterized by impaired gait, cognitive decline, and incontinence. The underlying cause of INPH is not fully understood, but it involves the abnormal distribution of cerebrospinal fluid (CSF) despite normal intracranial pressure. Interestingly, draining intracranial CSF often alleviates all three major symptoms, with gait improvement being the most notable. Since its initial description in 1950, the primary treatment for INPH has been the insertion of a shunt to facilitate CSF drainage. However, not all patients undergoing shunt surgery experience improvement. Several methods exist to predict which patients will benefit from shunt surgery, ranging from radiological assessments to infusion tests, with varying degrees of correlation to outcome. Many of these tests are time-consuming and carry a risk of infection. This study aims to take the first step toward developing a deep learning-based MRI clinical decision tool for shunt surgery in INPH patients. Specifically, it investigates whether brain morphology differences between shunt responders and non-responders can be identified using a Convolutional Neural Network (CNN) classifier applied to preoperative brain MR images. The CNN was developed in a previous project at Alvis. Methods and Materials The study utilizes MR brain images (volumes) from approximately 350 INPH patients who have undergone pre- and post-surgery gait assessments. Improvement in gait speed serves as the outcome variable. Pre-processing of the images includes co-registration to the T1 ICBM152 template brain, segmentation of CSF, white matter, and grey matter using SPM, and application of a brain mask. An existing CNN ensemble model will be trained to classify patients as shunt responders or non-responders based on their preoperative brain MR images. Significance The results of this study will determine the feasibility of predicting shunt surgery outcomes using solely radiological data. If successful, this approach will represent a significant step toward developing a reliable clinical decision tool for shunt surgery in INPH patients, potentially improving patient outcomes and optimizing surgical decisions.