Real-time computer-aided diagnosis of focal pancreatic masses from endoscopic ultrasound imaging based on a hybrid convolutional and long short-term memory neural network model.

Differential diagnosis of focal pancreatic masses is based on endoscopic ultrasound (EUS) guided fine needle aspiration biopsy (EUS-FNA/FNB). Several imaging techniques (i.e. gray-scale, color Doppler, contrast-enhancement and elastography) are used for differential diagnosis. However, diagnosis remains highly operator dependent. To address this problem, machine learning algorithms (MLA) can generate an automatic computer-aided diagnosis (CAD) by analyzing a large number of clinical images in real-time. We aimed to develop a MLA to characterize focal pancreatic masses during the EUS procedure. The study included 65 patients with focal pancreatic masses, with 20 EUS images selected from each patient (grayscale, color Doppler, arterial and venous phase contrast-enhancement and elastography). Images were classified based on cytopathology exam as: chronic pseudotumoral pancreatitis (CPP), neuroendocrine tumor (PNET) and ductal adenocarcinoma (PDAC). The MLA is based on a deep learning method which combines convolutional (CNN) and long short-term memory (LSTM) neural networks. 2688 images were used for training and 672 images for testing the deep learning models. The CNN was developed to identify the discriminative features of images, while a LSTM neural network was used to extract the dependencies between images. The model predicted the clinical diagnosis with an area under curve index of 0.98 and an overall accuracy of 98.26%. The negative (NPV) and positive (PPV) predictive values and the corresponding 95% confidential intervals (CI) are 96.7%, [94.5, 98.9] and 98.1%, [96.81, 99.4] for PDAC, 96.5%, [94.1, 98.8], and 99.7%, [99.3, 100] for CPP, and 98.9%, [97.5, 100] and 98.3%, [97.1, 99.4] for PNET. Following further validation on a independent test cohort, this method could become an efficient CAD tool to differentiate focal pancreatic masses in real-time.

Endoscopic ultrasound-guided fine needle aspiration: from the past to the future.

Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) is a technique which allows the study of cells obtained through aspiration in different locations near the gastrointestinal tract. EUS-FNA is used to acquire tissue from mucosal/submucosal tumors, as well as peri-intestinal structures including lymph nodes, pancreas, adrenal gland, gallbladder, bile duct, liver, kidney, lung, etc. The pancreas and lymph nodes are still the most common organs targeted in EUS-FNA. The overall accuracy of EUS is superior to computed tomography scan and magnetic resonance imaging for detecting pancreatic lesions. In most cases it is possible to avoid unnecessary surgical interventions in advanced pancreatic cancer, and EUS is considered the preferred method for loco-regional staging of pancreatic cancer. FNA improved the sensitivity and specificity compared to EUS imaging alone in detection of malignant lymph nodes. The negative predictive value of EUS-FNA is relatively low. The presence of a cytopathologist during EUS-FNA improves the diagnostic yield, decreasing unsatisfactory samples or need for additional passes, and consequently the procedural time. The size of the needle is another factor that could modify the diagnostic accuracy of EUS-FNA. Even though the EUS-FNA technique started in early nineteen's, there are many remarkable progresses culminating nowadays with the discovery and performance of needle-based confocal laser endomicroscopy. Last, but not least, identification and quantification of potential molecular markers for pancreatic cancer on cellular samples obtained by EUS-FNA could be a promising approach for the diagnosis of solid pancreatic masses.