Computer-assisted Endoscopic Ultrasound for Improved Diagnosis and Treatment of Pancreatic Cancer
Over 100,000 new cases of pancreatic cancer are identified in Europe each year, but the disease has a very poor prognosis with less than 1 in 5 patients surviving beyond one year following diagnosis. Earlier diagnosis and safe and effective treatments are critical to reducing mortality in pancreatic cancer. Recently, an increased use of MRI and CT imaging has contributed to earlier diagnosis through an increased detection of lesions in the pancreas, such as pancreatic cysts, some of which (about 20%) will be cancerous or progress into cancer. Therefore, further investigation to determine the type of lesion is extremely important. Endoscopic ultrasound – (EUS-) guided tissue sampling is the standard clinical method for evaluating pancreatic lesions, but the technique is highly operator dependent and technically difficult to master. Similar problems arise when EUS is used to guide minimally-invasive therapies, such as radiofrequency ablation and photodynamic therapy. Such treatments potentially offer a low-risk alternative to conventional surgery – pancreatic resection carries a significant risk with a mortality rate of 4.5% and complication rate of 31% – but precise knowledge of lesion location in relation to nearby anatomy is required for successful therapy planning and delivery, and this is challenging using standard technology. The aim of this project is to develop and test a computer-assisted navigation system, which combines realtime 3D endoscopic instrument tracking with multimodal (ultrasound-to-MRI/CT) image registration/fusion software, to improve instrument placement accuracy and make performing EUS-guided procedures much easier, quicker, and more reliable.