Clinical Workflows

We developed an automatic stereotactic radiosurgery (SRS) treatment planning workflow for brain metastases. This could reduce the workload of radiologists during time-consuming manual labeling and treatment planning simulation.

Lung cancer has the highest death rate among all cancers. We developed an automatically workflow for detecting pulmonary nodules and evaluating its malignancy. This workflow are abled to early detect suspicious nodules with high probability to become lung cancer.

Coronary artery disease (CAD) is one of the leading cause of death in the world. CAD was cause by atherosclerotic plaque which may lead to stenosis. We developed algorithms based on CCTA images, the combination of CCTA and CAG, and based on CFD to detect plaque and stenosis.

We differentiate optical coherence tomography (OCT) image among muscle, ligamentum flavum, and epidural space to assist anesthetist to know where the anesthesia needle goes during the injection.

Pancreatic cancer is one of the most difficult cancers to diagnose and the most lethal cancer. This computer-aided diagnosis tool can detect and classify pancreatic lesions from CT scans, which reduce the workload of radiologists and facilitate early detection of pancreatic cancer.

Prostate cancer only happens in men. It is hard to identify from MRI and it has no apparent symptoms. We proposed a segmentation method on prostate pathology images which are the whole slide images of patient tissue. Doctors can spend less time to look for what they need to focus on.