713 A New EUS Imaging Diagnosis for Pancreatic Cancer: A Pilot Study of Differential Diagnosis Between Chronic Pancreatitis and Pancreatic Cancer Based on the Quantitative Analysis With New Technology Using Raw Image

Abstract

A New EUS Imaging Diagnosis for Pancreatic Cancer: A Pilot Study of Differential Diagnosis Between Chronic Pancreatitis and Pancreatic Cancer Based on the Quantitative Analysis With New Technology Using Raw Image Atsushi Irisawa, Tadayuki Takagi, Tsunehiko Ikeda, Rei Suzuki, AI Sato, Goro Shibukawa, Takuto Hikichi, Masaki Sato, Katsutoshi Obara, Hiromasa Ohira Gastroenterology, Fukushima Medical University Aizu Medical Center, Fukushima, Japan; Gadteroenterology and Rheumatology, Fukushima Medical University, Fukushima, Japan; Endoscopy, Fukushima Medical University Hospital, Fukushima, Japan (Background and Aim) Reportedly, patients with chronic pancreatitis (CP) are 16 times more likely to develop pancreatic cancer (PC) than normal individuals. Strong evidence exists for the association of hereditary CP and PC. Although EUS enables detection of subtle pancreatic abnormalities and EUS-FNA is highly useful for diagnosing PC, it is frequently hard to detect the small cancer that invades slowly into the rough pancreatic parenchyma due to CP. The aim of this study is to develop a new EUS imaging diagnosis for PC using quantitative analysis with new technology. This method will be strongly help to excavate the small cancer in patients with CP. (Patients and Methods) Twenty-eight patients were enrolled in this study (CP in 10, PC in 28). The final diagnosis was established for all patients by various methods including EUS, EUS-FNA, surgery and long-term follow up more than 6 months. The ultrasound images were recorded in each case with using the EUS center (EU-ME1, Olympus Medical Systems Corp., Tokyo, Japan) and the curved linear-arrayed echoendoscope (GFUC240P-AL5 or GF-UCT240-AL5, Olympus Medical Systems Corp., Tokyo, Japan) as a raw image format (RAW image). The RAW image contains minimally processed data from the image sensor. It is not yet processed and therefore is not ready to be printed or edited with a graphics editor. The captured RAW image were quantitatively analyzed with a new technology (developed by Olympus Medical Systems Corp.), and attempted to create an image for the analysis of CP and PC. The image was described using two parameters on each RAW image: 1) mid-band fit and 2) intercept, which are both obtained from the spectrum intensity on the regression line of frequency spectrum available from the RAW imaging data. The quantitative analyses were performed by measuring the average value of intercepts and the average of mid-band fits. The results of the quantitative analyses were compared with the EUS/EUS-FNA diagnosis. (Results) There were significant differences between CP and PC in either 2 parameters (the average value of intercepts; 56 /-2.8 dB CP vs. 62 /-4.6 dB in PC, p 0.001. the average of mid-band fits; 47 / 2.3 dB in CP vs. 54 / 3.4 dB in PC, p 0.001). PC was clearly distinguished using 2 parameters from CP (figure) (Conclusion) It was considered that EUS examination with this new quantitative analysis using RAW imaging data would be useful for excavation and diagnosis the small cancer in patients with CP.