Computer-aided Diagnosis of Endoscopic Ultrasound Elastography Used in the Differentiation of Focal Pancreatic Masses

Ultrasound elastography is a recent ultrasound method used for the calculation of tissue elasticity distribution in real-time. The method allows the reconstruction of tissue elasticity (i.e. the elasticity modulus) and reveals directly the physical properties of the tissue, consequently showing different tissue hardness patterns that are determined by diseases. Tissue elastography can be easily performed real-time with conventional probes, including the linear EUS probes used for the examination of the pancreas. The calculation of tissue elasticity distribution is performed in real-time and the examination results are represented as transparent overlay colour images overimposed on the conventional gray-scale B-mode images.

Ultrasound elastography was previously used for the diagnosis of breast lesions, prostate cancer and thyroid nodules. However, the value of endoscopic ultrasound elastography for the diagnosis of pancreatic focal masses is not clear for the current moment, as some authors couldn't differentiate benign and malignant pancreatic tumors. Moreover, the intense fibrotic reaction and calcifications in chronic pancreatitis induce strain differences, and it is not clear if elastography is sensitive enough to detect them.

The study protocol is based on a semi-quantitative approach of EUS elastography data (movies) consisting of characterization of manually user-defined regions of interest, based on the hue histograms of the individual focal masses. Due to the inherent bias induced by selection of images from a dynamic sequence of EUS elastography, we have previously reported on the utility of using computer-aided diagnosis by averaging images from a dynamic sequence of EUS elastography. A special plugin (based on the ImageJ software, NIH, Bethesda, MD, USA) is used to compute hue histograms on average EUS elastography images, while the hue histograms values for each patient (0 to 255 values) are further used to classify the patients with benign and malignant lesions.

Ultrasound elastography will be performed during an usual EUS examination (7.5 MHz frequency), with two movies of 10 seconds recorded on the embedded HDD in order to minimize variability and to increase repeatability of acquisition. A two panel image with the usual conventional gray-scale B-mode EUS image on the right side and with the elastography image on the left side will be used. The region of interest will be preferably larger than the focal mass, in order to include the surrounding structures. In order to minimize the human bias, all the post-processing and computer analysis of digital movies will be performed within the IT Center in Craiova, with all programmers and statisticians being blinded to the clinical, pathological and imaging data, with the exception of the average hue histogram values calculated from a second region of interest manually traced around the focal mass.

The study design is prospective, blinded and multi-center, comparing endoscopic ultrasound elastography (EUS-EG) results for the detection and characterization of focal pancreatic masses by using artificial intelligence techniques, in comparison with the gold standard represented by pathology. The study will be performed with the approval of the institutional board review of each center.