Garg, Sushil Kumar

PurposeTo develop a two‐stage three‐dimensional (3D) convolutional neural networks (CNNs) for fully automated volumetric segmentation of pancreas on computed tomography (CT) and to further evaluate its performance in the context of intra‐reader and inter‐reader reliability at full dose and reduced radiation dose CTs on a public dataset.MethodsA dataset of 1994 abdomen CT scans (portal venous phase, slice thickness ≤ 3.75‐mm, multiple CT vendors) was curated by two radiologists (R1 and R2) to exclude cases with pancreatic pathology, suboptimal image quality, and image artifacts (n = 77). Remaining 1917 CTs were equally allocated between R1 and R2 for volumetric pancreas segmentation [ground truth (GT)]. This internal dataset was randomly divided into training (n = 1380), validation (n = 248), and test (n = 289) sets for the development of a two‐stage 3D CNN model based on a modified U‐net architecture for automated volumetric pancreas segmentation. Model’s performance for pancreas segmentation and the differences in model‐predicted pancreatic volumes vs GT volumes were compared on the test set. Subsequently, an external dataset from The Cancer Imaging Archive (TCIA) that had CT scans acquired at standard radiation dose and same scans reconstructed at a simulated 25% radiation dose was curated (n = 41). Volumetric pancreas segmentation was done on this TCIA dataset by R1 and R2 independently on the full dose and then at the reduced radiation dose CT images. Intra‐reader and inter‐reader reliability, model’s segmentation performance, and reliability between model‐predicted pancreatic volumes at full vs reduced dose were measured. Finally, model’s performance was tested on the benchmarking National Institute of Health (NIH)‐Pancreas CT (PCT) dataset.ResultsThree‐dimensional CNN had mean (SD) Dice similarity coefficient (DSC): 0.91 (0.03) and average Hausdorff distance of 0.15 (0.09) mm on the test set. Model’s performance was equivalent between males and females (P = 0.08) and across different CT slice thicknesses (P > 0.05) based on noninferiority statistical testing. There was no difference in model‐predicted and GT pancreatic volumes [mean predicted volume 99 cc (31cc); GT volume 101 cc (33 cc), P = 0.33]. Mean pancreatic volume difference was −2.7 cc (percent difference: −2.4% of GT volume) with excellent correlation between model‐predicted and GT volumes [concordance correlation coefficient (CCC)=0.97]. In the external TCIA dataset, the model had higher reliability than R1 and R2 on full vs reduced dose CT scans [model mean (SD) DSC: 0.96 (0.02), CCC = 0.995 vs R1 DSC: 0.83 (0.07), CCC = 0.89, and R2 DSC:0.87 (0.04), CCC = 0.97]. The DSC and volume concordance correlations for R1 vs R2 (inter‐reader reliability) were 0.85 (0.07), CCC = 0.90 at full dose and 0.83 (0.07), CCC = 0.96 at reduced dose datasets. There was good reliability between model and R1 at both full and reduced dose CT [full dose: DSC: 0.81 (0.07), CCC = 0.83 and reduced dose DSC:0.81 (0.08), CCC = 0.87]. Likewise, there was good reliability between model and R2 at both full and reduced dose CT [full dose: DSC: 0.84 (0.05), CCC = 0.89 and reduced dose DSC:0.83(0.06), CCC = 0.89]. There was no difference in model‐predicted and GT pancreatic volume in TCIA dataset (mean predicted volume 96 cc (33); GT pancreatic volume 89 cc (30), p = 0.31). Model had mean (SD) DSC: 0.89 (0.04) (minimum–maximum DSC: 0.79 −0.96) on the NIH‐PCT dataset.ConclusionA 3D CNN developed on the largest dataset of CTs is accurate for fully automated volumetric pancreas segmentation and is generalizable across a wide range of CT slice thicknesses, radiation dose, and patient gender. This 3D CNN offers a scalable tool to leverage biomarkers from pancreas morphometrics and radiomics for pancreatic diseases including for early pancreatic cancer detection.

Purpose of review Pancreatic cancer is the third leading cause of cancer death and with a dismal 5-year survival of 10%. Poor survival of pancreatic cancer is mostly due to its presentation and diagnosis at a late stage. The present article aims to update clinicians with recent progress in the field of early detection of pancreatic cancer. Recent findings Pancreatic cancer screening is not recommended in the general population due to its low prevalence. In this review, we discuss high-risk groups for pancreatic cancer, including inherited predisposition to pancreatic cancer, new-onset diabetes, mucinous pancreatic cyst, and chronic pancreatitis. We discuss methods of enrichment of high-risk groups with clinical models using electronic health records and biomarkers. We also discuss improvements in imaging modalities and emerging role of machine learning and artificial intelligence in the field of imaging and biomarker to aid in early identification of pancreatic cancer. Summary There are still vast challenges in the field of early detection of pancreatic cancer. We need to develop noninvasive prediagnostic validated biomarkers for longitudinal surveillance of high-risk individuals and imaging modalities that can identify pancreatic cancer early.

INTRODUCTION: The association of Acute Pancreatitis and Celiac Disease (CD) has been sparingly reported in the literature. CD is known to affect approximately 1% of the western population. A Swedish national database study reported a 3-fold increased risk of developing pancreatitis in patients with CD. The exact mechanism of this is unknown, but it could be related to an untreated celiac disease affecting the major papilla causing stenosis or exocrine pancreatic insufficiency that has been reported previously. This study aimed to look at the association between acute pancreatitis and celiac disease. We also looked at whether there was any association between chronic pancreatitis and celiac disease. METHODS: We performed a retrospective cross-sectional cohort study using the 2013 National Inpatient Sample (NIS) database. We included all patients who were hospitalized with either acute pancreatitis (ICD Code 577.0) or Chronic Pancreatitis (577.1). Patients who had diagnoses of both acute and chronic pancreatitis were considered to have acute on chronic pancreatitis and combined with the chronic pancreatitis group. We then analyzed both groups for the presence or absence of celiac disease. Continuous variables were compared using t-tests, and categorical variables were compared using Rao-Scott chi-square tests. All analyses were performed using SAS Survey Procedures (version 9.4, The SAS Institute, Cary, NC). RESULTS: We analyzed more than 35 million hospitalizations in 2013 out of which 378,715 had a primary discharge diagnosis of acute pancreatitis and 155,130 had a primary discharge diagnosis of chronic pancreatitis. The mean age in the acute pancreatitis group was 53.3 yrs and in chronic pancreatitis group was 51.7 yrs. There were more males in chronic pancreatitis group (55.1%) than females (44.9%). There was significantly higher prevalence of acute pancreatitis in CD (2.3%) than non-CD patients (1.2%) (Figure 1, P = 0.001). Similarly, there was higher prevalence of chronic pancreatitis (1.2%) than the control group (0.43%) (Figure 1, P < 0.001). CONCLUSION: There is a higher prevalence of acute pancreatitis in patients with CD. Patients were also found to have much higher prevalence of chronic pancreatitis in CD. This could be related to recurrent acute pancreatitis episodes in CD patients. Further studies are needed to validate these findings.

INTRODUCTION: The association of Acute Pancreatitis and Celiac Disease (CD) has been sparingly reported in the literature. CD is known to affect approximately 1% of the western population. A Swedish national database study reported a 3-fold increased risk of developing pancreatitis in patients with CD. The exact mechanism of this is unknown but it could be related to untreated celiac disease affecting the major papilla causing stenosis or exocrine pancreatic insufficiency that has been reported previously. The aim of this study was to look at the association between acute pancreatitis and celiac disease. We also looked whether there was any association between chronic pancreatitis and celiac disease. METHODS: We performed a retrospective cross-sectional cohort study using the 2013 National Inpatient Sample (NIS) database. We included all patients who were hospitalized with either acute pancreatitis (ICD Code 577.0) or Chronic Pancreatitis (577.1). Patients who had diagnoses of both acute and chronic pancreatitis were considered to have acute on chronic pancreatitis and combined with the chronic pancreatitis group. We then analyzed both groups for presence or absence of celiac disease. Continuous variables were compared using t-tests and categorical variables were compared using Rao-Scott chi-square tests. All analyses were performed using SAS Survey Procedures (version 9.4, The SAS Institute, Cary, NC). RESULTS: We analyzed more than 35 million hospitalizations in 2013 out of which 378,715 had a primary discharge diagnosis of acute pancreatitis and 155,130 had a primary discharge diagnosis of chronic pancreatitis. The mean age in the acute pancreatitis group was 53.3 yrs and in chronic pancreatitis group was 51.7 yrs. There were more males in chronic pancreatitis group (55.1%) than females (44.9%). There was significantly higher prevalence of acute pancreatitis in CD (2.3%) than non-CD patients (1.2%) (Figure 1, P = 0.001). Similarly, there was higher prevalence of chronic pancreatitis (1.2%) than the control group (0.43%) (Figure 1, P < 0.001). CONCLUSION: There is a higher prevalence of acute pancreatitis in patients with CD. Patients were also found to have much higher prevalence of chronic pancreatitis in CD. This could be related to recurrent acute pancreatitis episodes in CD patients. Further studies are needed to validate these findings.

INTRODUCTION: Double balloon enteroscopy (DBE) is an important tool for evaluation of obscure GI bleeding but is a limited resource given time and expertise required. Patient selection is important to help optimize diagnostic yield and subsequent successful management of small bowel bleeding. To maximize the efficiency of DBE performed, our endoscopy group established a screening protocol (Figure) prior to approval of any procedure. We developed a quality improvement project to assess the efficacy of this screening protocol and to evaluate which components of screening were most predictive for procedural success. METHODS: A retrospective chart review was performed on all patients who underwent anterograde or retrograde double balloon enteroscopy for suspected GI bleeding from January 2016 to December 2017. Our screening protocol was implemented on February 1 st , 2017. Data was obtained on patient demographics, clinical presentation, laboratory evaluation, and prior investigation for bleeding evaluation. Endoscopy outcomes were separated into procedural success, defined as lesion identified, and clinical success, defined as resolution of bleeding. Success rates were compared pre and post intervention. Regression analysis was performed to evaluate which components best predicted a successful procedure. RESULTS: We reviewed 114 DBE procedures in 89 patients, 53.5% male (n = 61) with a mean age of 62.8 + 13.3 years. Of these, 86% (n = 98) were performed prior to initiation of the screening protocol. Overall, there was no significant difference in procedural success pre and post intervention as defined by presence of recurrent bleeding, persistent anemia, or transfusion need (Table 1). On univariate analysis of clinical success, defined as lack of recurrent bleeding, gender, anticoagulation use, and presence of lesion on CE were significant (Table 2). Identification and intervention on a lesion during the balloon endoscopy did not impact future anemia, bleeding, or transfusion requirements. CONCLUSION: The patient screening protocol was successful in decreasing the number of procedures performed but did not improve our lesion detection rate on endoscopy and did not improve clinical outcomes as measured by rebleeding. Patients on anticoagulation and those who had prior lesions identified on capsule endoscopy had higher odds of rebleeding, indicating clinical failure. This data may help modify patient selection to maximize success of DBE procedures in the future.

INTRODUCTION: Esophageal cancer accounts for 1% of cancers diagnosed in the United States and the seventh most common cause of cancer-related mortality among men in the United States, with a 5-year survival rate of 19%. With an estimated $1.5 billion in direct medical costs in 2010, we look at ED utilization by this group of patients and analyze the economic burden and trends in ED visits and inpatient admissions over a 9-year period METHODS: The Nationwide Emergency Department Sample (NEDS; 2006-2014) was interrogated to identify patients with a primary diagnosis of esophageal cancer (ICD-9 CM code). The yearly incidence of ED visits, rates of hospitalization, length of stay (LOS), and total charges associated with ED or subsequent inpatient stays were assessed. As secondary analysis, survey logistic regression model was used to determine predictors of hospitalization. RESULTS: A total of 46,431 esophageal cancer-related ED visits were identified during this time period. Overall, early readmission visits increased overall from 6,295 in 2006 to a peak of 6,837 in 2014. The mean ED charges per visit, adjusted to inflation increased significantly from $1,401 in 2006 to $3,168 in 2014, while the mean total charges incurred by patients who were admitted to the hospital also increased from $41,009 in 2006 to $61,809 in 2014. Hospitalization rates decreased from 83.5% in 2006 to 78.8% in 2014 but in-hospital deaths decreased from 11.4% in 2006 to 8.7% in 2014. The average LOS in the hospitals decreased from 7.8 to 7.2 days. CONCLUSION: Increasing ED visits and charges associated with esophageal cancer are concerning. While hospitalization rates are slightly decreased during this 9-year period, there is an alarming uptrend in mortality rates which needs to be explored further. The results of our study suggest that further exploring factors that are associated with the recent upsurge in mortality may lead to the development of interventions that will significantly reduce in-hospital deaths.

INTRODUCTION: Immunoglobulin G4-related sclerosing cholangitis (IgG4-SC) has been recognized worldwide. However, its natural history and long-term outcomes are not well known. METHODS: In this retrospective, single center, cohort study, we identified patients who received a diagnosis of IgG4-SC at Mayo Clinic Rochester from 1999 to 2018. The diagnosis of IgG4-SC was rigorously ascertained by review of medical records and using the validated HISORt criteria. Comprehensive clinical data was extracted from patients' charts. Long-term outcomes of interest included cirrhosis, cholangiocarcinoma (CCA), and death. To compare outcomes, a random sample of PSC patients was selected in a 1:1 ratio matched for age of diagnosis and gender. Survival analyses were performed for comparisons using the Kaplan-Meier method, and log-rank test, and the cumulative incidence of cirrhosis and CCA was assessed using the competing risks regression method. RESULTS: A total of 61 patients were diagnosed with IgG4-SC during the study period. The median age of diagnosis was 66 years (IQR: 56–71 years), 85% were male, and the median follow-up was 6.1 years (IQR: 3.2–8.3 years). The pancreas was the most commonly involved extra-biliary organ (93%). Fifty-five patients received prednisone of whom 96% experienced complete or partial remission. Sixteen patients received rituximab as maintenance therapy, of whom 88% maintained remission. Three patients (5%) developed cirrhosis, and 2 patients (3%) developed CCA. Nine patients (15%) died; 2 due to cirrhosis, 2 due to CCA, 1 due to pancreatic cancer, and 4 due to non-hepatobiliary causes. Patients with IgG4-SC had significantly lower 5- and 10-year cumulative incidence of developing cirrhosis (5 y: 4% vs. 34%; 10 y: 6% vs. 43%; P < 0.001) and CCA (5 y: 3% vs. 17%; 10 y: 4% vs. 22%; P = 0.012) compared with PSC patients. Two patients in the PSC group underwent liver transplantation, while none in the IgG4-SC group. Overall 5- and 10-year survival was greater in the IgG4-SC cohort (5 y: 92% vs. 70%; 10 y: 77% vs. 57%; P = 0.02). CONCLUSION: Patients with IgG4-SC, the majority of whom received treatment, have a small potential risk of long-term hepatobiliary complications. Their risk of cirrhosis, CCA and death, was lower than in patients with PSC. The better outcome in the IgG4-SC group is likely due to the benefit of treatment. Surveillance in this population needs further assessment, especially the subset that does receive or respond to treatment.

INTRODUCTION: The average incidence of acute pancreatitis (AP) in patients with a history of solid organ transplant ranges from 2 to 8%. Patients with a history of bone marrow transplant are reported to have an incidence of 4%. Presentations include “chemical” pancreatitis with enzymes elevation only and clinical pancreatitis according to the revised Atlanta criteria. A few previous studies have shown that multi-organ failure and death are frequent complications of AP in transplanted patients. Most of the studies individualize their findings based on a specific organ transplanted (such as kidney or heart or liver, etc.) The aim of our study was to evaluate the outcomes of patients with acute pancreatitis with a history of any solid organ or bone marrow transplant using a national database. METHODS: We performed a retrospective cross-sectional analysis using the National Inpatient Sample (NIS) database from 2014. We included all patients with acute pancreatitis (ICD 9 Code: 577.0). Next, we divided the AP patients into two groups - one with history of a transplant and those without a history of transplant. We compared the outcomes of both groups. Both univariable and multivariable analyses were performed to assess the risk factors and outcomes of patients with acute pancreatitis in post transplant patients. RESULTS: We analyzed more than 35 million US hospitalizations in 2014 out of which 279,145 were acute pancreatitis related hospitalizations. There were 1,265 hospitalizations for acute pancreatitis in previously transplanted patients, an incidence of 4.53%. The mortality rate of AP was similar in non-transplanted patients (0.66%) than in transplanted patients (0.4%). The mean length of stay (4.62 days vs. 4.61 days) and mean total charges ($35,723 vs. $31,202) were similar in acute pancreatitis for transplant and non-transplant patients. Multivariable regression analysis showed that multiorgan dysfunction risk among AP patients was significantly higher if they had a history of transplant (OR of 2.27; P < 0.01). CONCLUSION: Our study showed that AP generally has similar outcomes in transplant and non-transplant patients even with a higher risk of multiorgan failure. More data is needed to determine if this holds true for subjects who develop AP in the immediate post transplant period.

Introduction: Clostridioides difficile infection (CDI) is associated with substantial emergency department (ED) and inpatient burden. To date, few studies have evaluated the ED burden of CDI. Using the Nationwide Emergency Department Sample, we evaluated trends in ED use, ED and inpatient charges, admission and mortality rates, length of stay, and independent risk factors for hospital admission and mortality after an ED visit. Methods: Using Nationwide Emergency Department Sample for 2006 through 2014, we identified all patients with the primary diagnosis of CDI (using diagnostic codes). We determined the trends in ED visits and used survey logistic regression analysis to identify factors associated with hospital admission. Results: Overall, 909,236 ED visits for CDI resulted in 817,935 admissions (90%) to the hospital. The number of visits increased from 76,709 in 2006 to 106,869 in 2014, and the admission rate decreased from 92.4% to 84.4%. ED charges adjusted for inflation went up from US$1433.0 to 2900, a significant rise even accounting for inflation. The overall length of hospital stay decreased from 7 to 5.8 days. Independent predictors of admission after ED visits included smoking, use of alcohol, and presence of multiple comorbidities. Independent risk factors for mortality in admitted patients include increasing age and presence of comorbidities. Conclusions: Although ED use for CDI increased, rates of hospital admission decreased over 9 years. Identification of predictors of admission and in-hospital mortality will help guide policies and interventions to reduce the burden on health care resources.