The goal of this observational study is to learn about deep learning radiogenomics for individualized therapy in unresectable gallbladder cancer. The main questions it aims to answer are: (i) whether a deep learning radiomics (DLR) model can be used for identification of HER2status and prediction of response to anti-HER2 directed therapy in unresectable GBC. (ii) validation of the deep learning radiomics (DLR) model for identification of HER2 status and prediction of response to anti-HER2 directed therapy in unresectable GBC. Participants will be asked to 1. Undergo biopsy of the gallbladder mass after a baseline CT scan 2. Based on the results of the biopsy, patients will be given chemotherapy either targeted (if Her2 positive) or non-targeted 3. Response to treatment will be assessed with a CT scan at 12 weeks of chemotherapy