Automatically Detecting Pancreatic Cysts in Autosomal Dominant Polycystic Kidney Disease on MRI Using Deep Learning.

BACKGROUND: Pancreatic cysts in autosomal dominant polycystic kidney disease (ADPKD) correlate with PKD2 mutations, which have a different phenotype than PKD1 mutations. However, pancreatic cysts are commonly overlooked by radiologists. Here, we automate the detection of pancreatic cysts on abdominal MRI in ADPKD. METHODS: Eight nnU-Net-based segmentation models with 2D or 3D configuration and various loss functions were trained on positive-only or positive-and-negative datasets, comprising axial and coronal T2-weighted MR images from 254 scans on 146 ADPKD patients with pancreatic cysts labeled independently by two radiologists. Model performance was evaluated on test subjects unseen in training, comprising 40 internal, 40 external, and 23 test-retest reproducibility ADPKD patients. RESULTS: Two radiologists agreed on 52% of cysts labeled on training data, and 33%/25% on internal/external test datasets. The 2D model with a loss of combined dice similarity coefficient and cross-entropy trained with the dataset with both positive and negative cases produced an optimal dice score of 0.7 ± 0.5/0.8 ± 0.4 at the voxel level on internal/external validation and was thus used as the best-performing model. In the test-retest, the optimal model showed superior reproducibility (83% agreement between scan A and B) in segmenting pancreatic cysts compared to six expert observers (77% agreement). In the internal/external validation, the optimal model showed high specificity of 94%/100% but limited sensitivity of 20%/24%. CONCLUSIONS: Labeling pancreatic cysts on T2 images of the abdomen in patients with ADPKD is challenging, deep learning can help the automated detection of pancreatic cysts, and further image quality improvement is warranted.

Pancreatico-renal fistula associated with pancreatic cysts caused by type 1 autoimmune pancreatitis.

To our best knowledge, the formation of a pancreatico-renal fistula and the presence of pancreatic fluid collection in the renal subcapsular space have not been reported as autoimmune pancreatitis (AIP) complications. We describe a case of a pancreatico-renal fistula associated with type 1 AIP. The patient presented with abdominal and back pain accompanied by pancreatic cystic lesions during an untreated course of AIP. The diagnosis of pancreatico-renal fistula was based on the presence of a left renal subcapsular fluid collection containing pancreatic amylase, disappearance of pancreatic cysts, and a defect in the partial anterior renal fascia observed on imaging studies. Treatment with steroids and percutaneous drainage resulted in improvement. Pancreatic pseudocysts can affect other organs owing to their digestive action. Similar symptoms may occur in patients with AIP.

Comparison of intra- and inter-reader agreement of abbreviated versus comprehensive MRCP for pancreatic cyst surveillance.

OBJECTIVE: To retrospectively compare inter- and intra-reader agreement of abbreviated MRCP (aMRCP) with comprehensive MRI (cMRCP) protocol for detection of worrisome features, high-risk stigmata, and concomitant pancreatic cancer in pancreatic cyst surveillance. METHODS: 151 patients (104 women, mean age: 69[10] years) with baseline and follow-up contrast-enhanced MRIs were included. This comprised 138 patients under cyst surveillance with 5-year follow-up showing no pancreatic ductal adenocarcinoma (PDAC), 6 with pancreatic cystic lesion-derived malignancy, and 7 with concomitant PDAC. The aMRCP protocol used four sequences (axial and coronal Half-Fourier Single-shot Turbo-spin-Echo, axial T1 fat-saturated pre-contrast, and 3D-MRCP), while cMRCP included all standard sequences, including post-contrast. Three blinded abdominal radiologists assessed baseline cyst characteristics, worrisome features, high-risk stigmata, and PDAC signs using both aMRCP and cMRCP, with a 2-week washout period. Intra- and inter-reader agreement were calculated using Fleiss' multi-rater kappa and Intra-class Correlation Coefficient (ICC). 95% confidence intervals (CI) were calculated. RESULTS: Cyst size, growth, and abrupt main pancreatic duct transition had strong intra- and inter-reader agreement. Intra-reader agreement was ICC = 0.93-0.99 for cyst size, ICC = 0.71-1.00 for cyst growth, and kappa = 0.83-1.00 for abrupt duct transition. Inter-reader agreement for cyst size was ICC = 0.86 (aMRCP) and ICC = 0.83 (cMRCP), and for abrupt duct transition was kappa = 0.84 (aMRCP) and kappa = 0.69 (cMRCP). Thickened cyst wall, mural nodule and cyst-duct communication demonstrated varying intra-reader agreements and poor inter-reader agreements. CONCLUSION: aMRCP showed high intra- and inter-reader agreement for most pancreatic cyst parameters that highly rely on T2-weighted sequences.

Dilated common bile duct is commonly associated with main duct Intraductal Papillary Mucinous Neoplasm of the pancreas.

BACKGROUND: Dilatation of common bile duct (CBD) is mostly pathological and mainly occurs secondary to mechanical causes. We aimed to explore the prevalence of CBD dilatation in Intraductal Papillary Mucinous Neoplasms of the pancreas (IPMN) among patients referred to EUS. METHODS: A retrospective study of all patients who had an EUS diagnosis of IPMN from 2011 to 2019 at Galilee Medical Center were extracted. Control group including patients with other types of pancreatic cysts. RESULTS: Overall, 2400 patients were included in the study, of them 158 patients were diagnosed with pancreatic cysts, 117 patients (74%) diagnosed with IPMN (group A), and 41 patients (26%) diagnosed with other pancreatic cysts (group B). Univariate analysis showed significant association of IPMN (OR 3.8, 95% CI 1.3-11.5), resected gallbladder (GB) (OR 7.75, 95% CI 3.19-18.84), and age (OR 1, 95% CI 1.01-1.08) with CBD dilatation. Classifying IPMN to sub-groups using adjusted multivariate logistic regression analysis, only main duct-IPMN (MD-IPMN) significantly correlated with CBD dilatation compared to branch duct (BD)-IPMN and mixed type-IPMN (OR 19.6, 95% CI 4.57-83.33, OR 16.3, 95% CI 3.02-88.08). CONCLUSION: MD-IPMN was significantly correlated with dilated CBD. Assessment of the pancreas is warranted in encountered cases of dilated CBD without obvious mechanical cause.

Endoscopic Ultrasound-Guided Ablation of Pancreatic Mucinous Cysts.

Endoscopic ultrasound (EUS) has rapidly evolved from a diagnostic to a therapeutic tool with applications for various pancreaticobiliary diseases. As part of this evolution, EUS-guided chemoablation for neoplastic pancreatic cysts is developing as a minimally invasive treatment option for appropriately selected mucinous cysts, which can spare patients major resective surgery and may reduce progression to pancreatic cancer. Chemotherapeutic cyst ablation has demonstrated encouraging complete resolution rates, while an alcohol-free chemoablation protocol has demonstrated a significant decrease in adverse events without a compromise to complete ablation rates when compared with previous alcohol-based protocols. Most pancreatic cysts are small, low risk, and best managed by surveillance per accepted guidelines. Cysts with features suggestive of overt malignancy are best discussed by a multidisciplinary committee, and surgery is considered if appropriate. However, for patients in the middle ground with cysts that are structurally suitable for chemoablation, alcohol-free chemoablation has been shown to allow effective, safe, and durable results especially for those who are not ideal operative candidates. EUS-guided alcohol-free chemoablation is promising and continues to evolve; however, as a relatively novel treatment option it has areas of uncertainty that will require further investigation and development.

The Incidental Pancreatic Cyst: When to Worry About Cancer.

Incidental pancreatic cystic lesions are a common challenge encountered by diagnostic radiologists. Specifically, given the prevalence of benign pancreatic cystic lesions, determining when to recommend aggressive actions such as surgical resection or endoscopic ultrasound with sampling is difficult. In this article, we review the common types of cystic pancreatic lesions including serous cystadenoma, intraductal papillary mucinous neoplasm, and mucinous cystic neoplasm with imaging examples of each. We also discuss high-risk or worrisome imaging features that warrant a referral to a surgeon or endoscopist and provid several examples of these features. These imaging features adhere to the latest guidelines from the International Consensus Guidelines, American Gastroenterological Association (2015), American College of Gastroenterology (2018), American College of Radiology (2010, 2017), and European Guidelines (2013, 2018). Our focused article addresses the imaging dilemma of managing incidental cystic pancreatic lesions, weighing the options between imaging follow-up and aggressive interventions.

DenseNet model incorporating hybrid attention mechanisms and clinical features for pancreatic cystic tumor classification.

PURPOSE: The aim of this study is to develop a deep learning model capable of discriminating between pancreatic plasma cystic neoplasms (SCN) and mucinous cystic neoplasms (MCN) by leveraging patient-specific clinical features and imaging outcomes. The intent is to offer valuable diagnostic support to clinicians in their clinical decision-making processes. METHODS: The construction of the deep learning model involved utilizing a dataset comprising abdominal magnetic resonance T2-weighted images obtained from patients diagnosed with pancreatic cystic tumors at Changhai Hospital. The dataset comprised 207 patients with SCN and 93 patients with MCN, encompassing a total of 1761 images. The foundational architecture employed was DenseNet-161, augmented with a hybrid attention mechanism module. This integration aimed to enhance the network's attentiveness toward channel and spatial features, thereby amplifying its performance. Additionally, clinical features were incorporated prior to the fully connected layer of the network to actively contribute to subsequent decision-making processes, thereby significantly augmenting the model's classification accuracy. The final patient classification outcomes were derived using a joint voting methodology, and the model underwent comprehensive evaluation. RESULTS: Using the five-fold cross validation, the accuracy of the classification model in this paper was 92.44%, with an AUC value of 0.971, a precision rate of 0.956, a recall rate of 0.919, a specificity of 0.933, and an F1-score of 0.936. CONCLUSION: This study demonstrates that the DenseNet model, which incorporates hybrid attention mechanisms and clinical features, is effective for distinguishing between SCN and MCN, and has potential application for the diagnosis of pancreatic cystic tumors in clinical practice.

The multidisciplinary pancreatic cyst clinic: A collaborative approach to streamlined management.

BACKGROUND: Pancreatic cysts are often incidentally detected on routine imaging studies. Of these, mucinous cysts have a malignant potential. Several guidelines propose different management strategies, and implementation in patient care is inconsistent in the absence of dedicated infrastructure. METHODS: To address the challenges of pancreatic cyst diagnosis and management, we established a multidisciplinary pancreas cyst clinic (PCC) within our health system. This clinic encompasses both tertiary care academic centers and community hospitals, with leadership from surgical oncology, gastroenterology, and radiology. Our PCC's primary goal is to provide accurate diagnosis and tailored management recommendations for all patients with pancreatic cysts. Additionally, we maintain a prospective database to study the disease's natural history and the outcomes of various treatment strategies. CLINIC INFRASTRUCTURE: The clinic meets once per week for 45 min virtually via Zoom in the mornings. Patients are referred via electronic medical record (EMR) order, telephone call, or email from patient or referring provider. A dedicated advanced practice provider reviews referrals several times per day, calls patients to gather clinical data, ensures imaging is uploaded, and coordinates logistical aspects of the meeting during the dedicated time. Conferences are attended by representatives from surgery, radiology, medical pancreatology, and interventional gastroenterology. Each patient case is reviewed in detail and recommendations are submitted to referring providers and patients via an EMR message and letter. For patients requiring imaging surveillance, patients are followed longitudinally by the referring provider, gastroenterology team, or surgical team. For patients requiring endoscopic ultrasound (EUS) or surgical consultation, expedited referral to these services is made with prompt subsequent evaluation. RESULTS: A total of 1052 patients from our health system were evaluated between 2020 and 2021. Of these, 196 (18.6 %) underwent EUS, 41 (3.9 %) underwent upfront surgical resection, and the remainder were referred to gastroenterology (141-13.4 %), surgery (314-29.8 %), or back to their referring provider (597-56.7 %) for ongoing surveillance in collaboration with their primary care provider (PCP). Of cysts under surveillance, 61.3 % remained stable, 13.2 % increased in size, and 2 % decreased in size. A total of 2.3 % of patients were recommended to discontinue surveillance. CONCLUSIONS: The PCC provides infrastructure that has served to provide multidisciplinary review and consensus recommendations to patients with pancreatic cysts. This has served to improve the application of guidelines while providing individualized recommendations to each patient, while aiding non-expert referring providers throughout the region.

[Applications of Artificial Intelligence in Pancreatic Cystic Lesion Imaging].

As the detection rate of pancreatic cystic lesions(PCL)increases,artificial intelligence(AI)has made breakthroughs in the imaging workflow of PCL,including image post-processing,lesion detection,segmentation,diagnosis and differential diagnosis.AI-based image post-processing can optimize the quality of medical images and AI-assisted models for lesion detection,segmentation,diagnosis and differential diagnosis significantly enhance the work efficiency of radiologists.This article reviews the application progress of AI in PCL imaging and provides prospects for future research directions.

Investigating morphological changes in a simple mucinous cyst during the follow-up period: a case report.

A 64-year-old man was referred to our department after a small pancreatic cystic lesion was discovered on computed tomography performed to assess choledocholithiasis. Multiple standard imaging modalities, including endoscopic ultrasound (EUS), failed to reveal pancreatic masses, wall thickening, or mural nodules. Consequently, a benign pancreatic cystic lesion was suspected, and the patient underwent biannual imaging studies including rotating magnetic resonance imaging and EUS. Six years after the initial detection of the pancreatic cyst, wall thickening was observed, leading to a shortened observation period of once every 3 months. After 6.5 years, hypoechoic area surrounding the cyst, which could be interpreted as thickening of the cyst wall was observed, prompting distal pancreatectomy due to the suspicion of malignant disease. The histopathological examination revealed a unilocular mucinous cyst with a single layer of cuboidal cells and low-grade dysplasia. A fibrous proliferation of the polycystic stroma and no ovarian-type stroma was observed. Malignant cells were absent from the cystic epithelium and stroma. The final histopathological diagnosis was a simple mucinous cyst of the pancreatic tail.

A unique case of a typical pancreatic ductal adenocarcinoma that initially presented with a cystic component but underwent morphological changes.

A 66-year-old man was initially suspected of having a microcystic serous cystic neoplasm based on magnetic resonance imaging findings of a multifocal mass measuring 46 mm in the pancreatic head, with a cystic component showing a high signal on T2-weighted images. The tumor marker levels were within normal limits. However, contrast-enhanced computed tomography revealed thick cyst walls with delayed staining, which was atypical for serous cystic neoplasms; therefore, the patient was followed up closely. Twenty-two months later, the delayed contrast area was enlarged, carbohydrate antigen 19-9 levels were elevated, and 18 F-fluorodeoxyglucose-positron emission tomography revealed increased accumulation, indicating a potentially malignant lesion. Pancreatoduodenectomy was performed and histopathological examination confirmed the diagnosis of normal-type pancreatic carcinoma with predominantly poorly differentiated cells. Based on the pathological findings and a literature review, it is highly likely that this case represents pancreatic ductal adenocarcinoma with a cystic structure from the beginning. While distinguishing pancreatic ductal adenocarcinoma from other pancreatic cystic tumors, such as serous cystic neoplasms, is critical owing to differing treatments and prognoses, caution is warranted as they may exhibit similar imaging features, as observed in our patient.

Pancreatic cyst prevalence and detection with photon counting CT compared with conventional energy integrating detector CT.

PURPOSE: To calculate the prevalence of pancreatic cysts on photon counting CT (PCCT) and compare with that of 128-slice conventional energy-integrating detector CT (EIDCT). METHOD: A retrospective single institution database search identified all contrast-enhanced abdominal CT examinations performed at an outpatient facility that has both a PCCT and EIDCT between 4/11/2022 and 7/26/2022. The presence and size of pancreatic cysts were recorded. In patients with PCCT reported pancreatic cysts, prior CT imaging (EIDCT) was reviewed for reported pancreatic cysts. Fisher's exact test was used to compare the pancreatic cyst detection rate for PCCT and EIDCT. Wilcoxon rank sum test was used to compare cyst size and patient age. A p <.05 indicated statistical significance. RESULTS: 2494 patients were included. Our pancreatic cyst detection rate was 4.9 % (49/1009) with PCCT and 3.0 % (44/1485) for EIDCT (p =.017). For CT angiograms, pancreatic cysts were detected in 6.6 % (21/319) with PCCT and 0.0 % (0/141) with EIDCT (p <.001). Pancreatic cyst detection rate was not statistically different for portal venous, enterography, renal mass, pancreas, 3-phase liver, or venogram protocols (all p >.05). Mean[SD] pancreatic cyst size was 13.7[9.7]mm for PCCT and 15.3[14.7] for EIDCT (p =.95). 55.1 % (27/49) of PCCT and 61.4 % (27/44) of EIDCT that described pancreatic cysts had prior contrast-enhanced EIDCTs. Of these, 40.7 % (11/27) of PCCT and 14.8 % (4/27) of EIDCT described pancreatic cysts were not previously reported (p =.027). CONCLUSIONS: Photon-counting CT afforded greater pancreatic cyst detection than conventional energy-integrating detector CT, particularly with CT angiograms.

Detection Rate of Small Pancreas Cysts and Intrareader Reliability of the Cysts Size Measurements on Transabdominal Ultrasonography.

ABSTRACT: This study aimed to assess the detection rate of small (<10 mm) pancreas cyst and intrareader reliability for cyst size measurements on transabdominal ultrasonography (US). From 2020 to 2022, 194 pancreas cysts in 173 patients, incidentally detected on computed tomography or magnetic resonance imaging, were evaluated on US by 1 of 2 radiologists (readers 1 and 2). Intrareader agreements of cyst size measurements on US were assessed by intraclass correlation coefficient (ICC). Bland-Altman plot was used to visualize the differences between the first and second size measurements in each reader. In this study, readers 1 and 2 evaluated 86 cysts in 76 patients and 108 cysts in 97 patients, respectively. Most of the cysts (191 of 194) were located in the nontail portion of the pancreas. Overall detection rate of pancreas cysts by US was 92.3% (179 of 194). The mean size of measured 179 pancreas cysts was 4.7 ± 1.5 mm. The readers showed excellent intrareader agreements (ICC = 0.925 and 0.960) for cyst size measurements, except for the cysts with size ≤5 mm, where both readers showed good intrareader agreements (ICC = 0.848 and 0.873). The 95% limits of agreement of readers 1 and 2 were 13.8% and 14.9% of the mean, respectively. Therefore, transabdominal US could be a reliable follow-up imaging modality for small (<10 mm) nontail pancreas cysts incidentally detected on computed tomography or magnetic resonance imaging, especially for the cysts with size between 5 and 10 mm. Size changes of the pancreas cysts approximately less than 15% may be within the measurement error.

Performance of Intracystic Glucose Measurement for the Characterization of Pancreatic Cystic Lesions.

BACKGROUND AND AIMS: Endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA) is essential for the classification of pancreatic cystic lesions (PCLs). Recently, intracystic glucose has been suggested as an alternative to carcinoembryonic antigen (CEA) level as a predictor of mucinous cystic lesions (M-PCLs). This study aims to evaluate the diagnostic performance of intra-cystic glucose in distinguishing between M-PCLs and non M-PCLs (NM-PCLs) and to analyze the possibility of on-site glucose measurement with a standard glucometer. METHODS: Patients with PCLs submitted to EUS-FNA with simultaneous intracystic glucose measurement between 2017 and 2022 were included. The diagnostic performance of glucose versus CEA for the differentiation between M-PCLs and NM-PCLs was compared to a final diagnosis based on the analysis of surgical specimen, intracystic biopsy or, if this data was unavailable, multidisciplinary evaluation. A cut-off of <50 mg/dL was used for the diagnosis of MCLs. Additionally, the agreement between on-site glucose determination with a standard glucometer and laboratory glucose measurement was assessed. RESULTS: Mucinous lesions accounted for 56% of all PCLs. The median values of glucose and CEA for M-PCLs were 18 mg/dL and 286 ng/mL, respectively. Intracystic glucose had a sensitivity and specificity of 93.2% and 76.5%, respectively, for the diagnosis of MCLs (versus 55.6% and 87.5%, respectively, for CEA). The area under the curve was 0.870 for on-site glucose (versus 0.806 for CEA). An excellent correlation was observed between on-site and laboratory glucose measurement (ρ=0.919). CONCLUSIONS: The measurement of intracystic glucose showed superior performance compared with CEA in distinguishing between M-PCLs and NM-PCLs, with excellent correlation between on-site and conventional lab glucose measurement. Thus, on-site intracystic glucose appears to be an excellent biomarker for the characterization of PCLs due to its low cost, high availability, and the need for a minimal cyst fluid volume for its determination.

Global Prevalence of Pancreatic Cystic Lesions in the General Population on Magnetic Resonance Imaging: A Systematic Review and Meta-analysis.

BACKGROUND & AIMS: Understanding the burden of pancreatic cystic lesions (PCLs) in the general population is important for clinicians and policymakers. In this systematic review, we sought to estimate the global prevalence of PCLs using magnetic resonance imaging (MRI) and to investigate factors that contribute to its variation. METHODS: We searched MEDLINE, EMBASE, and Cochrane Central, from database inception through February 2023. We included full-text articles that reported the prevalence of PCLs using MRI in the general population. A proportional meta-analysis was performed, and the prevalence of PCLs was pooled using a random-effects model. RESULTS: Fifteen studies with 65,607 subjects were identified. The pooled prevalence of PCLs was 16% (95% confidence interval [CI], 13%-18%; I2 = 99%), most of which were under 10 mm. Age-specific prevalence of PCLs increased from 9% (95% CI, 7%-12%) at 50 to 59 years, to 18% (95% CI, 14%-22%) at 60 to 69 years, 26% (95% CI, 20%-33%) at 70 to 79 years, and 38% at 80 years and above (95% CI, 25%-52%). There was no difference in prevalence between sexes. Subgroup analysis showed higher PCL prevalence when imaging findings were confirmed by independent radiologist(s) (25%; 95% CI, 16%-33%) than when chart review alone was used (5%; 95% CI, 4%-7%; P < .01). There was no independent association of PCL prevalence with geographic location (Europe, North America, or Asia), MRI indication (screening vs evaluation of non-pancreatic pathology), enrollment period, sample size, magnet strength (1.5 vs 3 tesla), and MRI sequence (magnetic resonance cholangiopancreatography vs no magnetic resonance cholangiopancreatography). CONCLUSION: In this systematic review, the global prevalence of PCLs using a highly sensitive noninvasive imaging modality ranged between 13% and 18%.