Search

Novel artificial intelligence techniques are emerging in all fields of healthcare, including gastroenterology. The aim of this review is to give an overview of artificial intelligence applications in the management of pancreatic diseases. We performed a systematic literature search in PubMed and Medline up to May 2020 to identify relevant articles. Our results showed that the development of machine-learning based applications is rapidly evolving in the management of pancreatic diseases, guiding precision medicine in clinical, endoscopic and radiologic settings. Before implementation into clinical practice, further research should focus on the external validation of novel techniques, clarifying the accuracy and robustness of these models.

BACKGROUND: Current recommendations on training in endoscopic retrograde cholangiopancreatography are predicated on a performance-centred approach designed to ensure that trainees achieve appropriate skills. We aimed to analyse how competence in endoscopic retrograde cholangiopancreatography is defined in the literature and what proportion of trainees actually reach this threshold. METHODS: We conducted a systematic MEDLINE search for studies reporting on endoscopic retrograde cholangiopancreatography training programmes in a clinical setting. The main outcome measure was threshold for achieving competence in endoscopic retrograde cholangiopancreatography; the secondary outcome measure was assessment of trainee performance. Quality was assessed using the Cochrane Risk of Bias tool and the Methodological Index for Non-Randomized Studies criteria. RESULTS: Of 522 initially identified articles, 20 were included in the analysis; most studies showed a high risk of bias. Cannulation rate of the desired duct was the main marker of competence in all studies; however, only 8/20 studies reported on the performance of individual trainees, who achieved their respective standard of competence in only 25.6% of reported cases. CONCLUSIONS: Current literature identifies cannulation rate of a native papilla to be the most appropriate measure of endoscopic retrograde cholangiopancreatography competence; however, most trainees do not reach predefined competence thresholds. Furthermore, due to the limitations of available studies, the most appropriate competence measure remains subject for debate.

Background and Aim: Data on non-surgical treatment of groove pancreatitis (GP) and the risk of cancer are lacking. We aimed to determine the prevalence and predictors of cancer in patients in whom the diagnosis GP was considered, and to evaluate symptom improvement after treatment. Study: Patients referred with possible GP (2001-2014) were retrospectively included. An experienced radiologist reassessed imaging. GP patients received questionnaires evaluating their symptoms. Results: From the 38 possible GP patients, 10 had cancer (26%) and 28 GP (74%). Compared with cancer patients, GP patients more frequently had cysts (2/10 vs. 18/28, P = 0.03), less often jaundice (6/10 vs 3/27, P

Purpose Biliary stents may cause susceptibility artifacts, gradient-induced artifacts, and radio frequency (RF) induced artifacts on magnetic resonance images, which can hinder accurate target volume delineation in radiotherapy. In this study, the authors investigated and quantified the magnitude of these artifacts for stents of different materials. Methods Eight biliary stents made of nitinol, platinum-cored nitinol, stainless steel, or polyethylene from seven vendors, with different lengths (57–98 mm) and diameters (3.0–11.7 mm), were placed in a phantom. To quantify the susceptibility artifacts sequence-independently, ΔB0-maps and T2∗-maps were acquired at 1.5 and 3 T. To study the effect of the gradient-induced artifacts at 3 T, signal decay in images obtained with maximum readout gradient-induced artifacts was compared to signal decay in reference scans. To quantify the RF induced artifacts at 3 T, B1-maps were acquired. Finally, ΔB0-maps and T2∗-maps were acquired at 3 T of two pancreatic cancer patients who had received platinum-cored nitinol biliary stents. Results Outside the stent, susceptibility artifacts dominated the other artifacts. The stainless steel stent produced the largest susceptibility artifacts. The other stents caused decreased T2∗ up to 5.1 mm (1.5 T) and 8.5 mm (3 T) from the edge of the stent. For sequences with a higher bandwidth per voxel (1.5 T: BWvox > 275 Hz/voxel; 3 T: BWvox > 500 Hz/voxel), the B0-related susceptibility artifacts were negligible (

Background When assessing the feasibility of surveillance for pancreatic cancer (PC), it is important to address its psychological burden. The aim of this ongoing study is to evaluate the psychological burden of annual pancreatic surveillance for individuals at high risk to develop PC. Methods This is a multicenter prospective study. High-risk individuals who undergo annual pancreatic surveillance with magnetic resonance imaging (MRI) and endoscopic ultrasound (EUS) were invited to complete questionnaires to assess motivations for participating in surveillance, experiences with participation, perceived PC risk, topics of concern, and psychological distress. Questionnaires were sent after intake for participation (T1), after the first MRI and EUS (T2), and after the MRI and EUS 1 (T3), 2 (T4), and 3 years (T5) after first surveillance. Results In total, 140 out of 152 individuals returned one or more of the questionnaires (response 92%); 477 questionnaires were analyzed. The most frequently reported motivation for participating in surveillance was the possible early detection of (a precursor stage of) cancer (95–100%). Only a minority of respondents experienced MRI and EUS as uncomfortable (10% and 11%, respectively), and respondents dreaded their next EUS investigation less as surveillance progressed. Respondents' cancer worries decreased significantly over time, and both their anxiety and depression scores remained stable and low over the 3-year period of follow-up. Conclusions The psychological burden of pancreatic surveillance is low at all assessments. Therefore, from a psychological point of view, participation of high-risk individuals in an annual pancreatic surveillance program is feasible.Copyright © 2015 John Wiley & Sons, Ltd.

Purpose: In radiation therapy of pancreatic cancer, tumor alignment prior to each treatment fraction is improved when intratumoral gold fiducial markers (from here onwards: markers), which are visible on computed tomography (CT) and cone beam CT, are used. Visibility of these markers on magnetic resonance imaging (MRI) might improve image registration between CT and magnetic resonance (MR) images for tumor delineation purposes. However, concomitant image artifacts induced by markers are undesirable. The extent of visibility and artifact size depend on MRI-sequence parameters. The authors’ goal was to determine for various markers their potential to be visible and to generate artifacts, using measures that are independent of the MRI-sequence parameters. Methods: The authors selected ten different markers suitable for endoscopic placement in the pancreas and placed them into a phantom. The markers varied in diameter (0.28–0.6 mm), shape, and iron content (0%–0.5%). For each marker, the authors calculated T 2 ∗-maps and ΔB 0-maps using MRI measurements. A decrease in relaxation time T 2 ∗ can cause signal voids, associated with visibility, while a change in the magnetic field B 0 can cause signal shifts, which are associated with artifacts. These shifts inhibit accurate tumor delineation. As a measure for potential visibility, the authors used the volume of low T 2 ∗, i.e., the volume for which T 2 ∗ differed from the background by >15 ms. As a measure for potential artifacts, the authors used the volume for which |ΔB 0| > 9.4 × 10−8 T (4 Hz). To test whether there is a correlation between visibility and artifact size, the authors calculated the Spearman’s correlation coefficient (Rs ) between the volume of low T 2 ∗ and the volume of high |ΔB 0|. The authors compared the maps with images obtained using a clinical MR-sequence. Finally, for the best visible marker as well as the marker that showed the smallest artifact, the authors compared the phantom data with in vivo MR-images in four pancreatic cancer patients. Results: The authors found a strong correlation (Rs = 1.00, p < 0.01) between the volume of low T 2 ∗ and the volume with high |ΔB 0|. Visibility in clinical MR-images increased with lower T 2 ∗. Signal shift artifacts became worse for markers with high |ΔB 0|. The marker that was best visible in the phantom, a folded marker with 0.5% iron content, was also visible in vivo, but showed artifacts on diffusion weighted images. The marker with the smallest artifact in the phantom, a small, stretched, ironless marker, was indiscernible on in vivo MR-images. Conclusions: Changes in T 2 ∗ and ΔB 0 are sequence-independent measures for potential visibility and artifact size, respectively. Improved visibility of markers correlates strongly to signal shift artifacts; therefore, marker choice will depend on the clinical purpose. When visibility of the markers is most important, markers that contain iron are optimal, preferably in a folded configuration. For artifact sensitive imaging, small ironless markers are best, preferably in a stretched configuration.