Automated Artificial Intelligence Model Trained on a Large Data Set Can Detect Pancreas Cancer on Diagnostic Computed Tomography Scans As Well As Visually Occult Preinvasive Cancer on Prediagnostic Computed Tomography Scans.

BACKGROUND & AIMS: The aims of our case-control study were (1) to develop an automated 3-dimensional (3D) Convolutional Neural Network (CNN) for detection of pancreatic ductal adenocarcinoma (PDA) on diagnostic computed tomography scans (CTs), (2) evaluate its generalizability on multi-institutional public data sets, (3) its utility as a potential screening tool using a simulated cohort with high pretest probability, and (4) its ability to detect visually occult preinvasive cancer on prediagnostic CTs. METHODS: A 3D-CNN classification system was trained using algorithmically generated bounding boxes and pancreatic masks on a curated data set of 696 portal phase diagnostic CTs with PDA and 1080 control images with a nonneoplastic pancreas. The model was evaluated on (1) an intramural hold-out test subset (409 CTs with PDA, 829 controls); (2) a simulated cohort with a case-control distribution that matched the risk of PDA in glycemically defined new-onset diabetes, and Enriching New-Onset Diabetes for Pancreatic Cancer score ≥3; (3) multi-institutional public data sets (194 CTs with PDA, 80 controls), and (4) a cohort of 100 prediagnostic CTs (i.e., CTs incidentally acquired 3-36 months before clinical diagnosis of PDA) without a focal mass, and 134 controls. RESULTS: Of the CTs in the intramural test subset, 798 (64%) were from other hospitals. The model correctly classified 360 CTs (88%) with PDA and 783 control CTs (94%), with a mean accuracy 0.92 (95% CI, 0.91-0.94), area under the receiver operating characteristic (AUROC) curve of 0.97 (95% CI, 0.96-0.98), sensitivity of 0.88 (95% CI, 0.85-0.91), and specificity of 0.95 (95% CI, 0.93-0.96). Activation areas on heat maps overlapped with the tumor in 350 of 360 CTs (97%). Performance was high across tumor stages (sensitivity of 0.80, 0.87, 0.95, and 1.0 on T1 through T4 stages, respectively), comparable for hypodense vs isodense tumors (sensitivity: 0.90 vs 0.82), different age, sex, CT slice thicknesses, and vendors (all P > .05), and generalizable on both the simulated cohort (accuracy, 0.95 [95% 0.94-0.95]; AUROC curve, 0.97 [95% CI, 0.94-0.99]) and public data sets (accuracy, 0.86 [95% CI, 0.82-0.90]; AUROC curve, 0.90 [95% CI, 0.86-0.95]). Despite being exclusively trained on diagnostic CTs with larger tumors, the model could detect occult PDA on prediagnostic CTs (accuracy, 0.84 [95% CI, 0.79-0.88]; AUROC curve, 0.91 [95% CI, 0.86-0.94]; sensitivity, 0.75 [95% CI, 0.67-0.84]; and specificity, 0.90 [95% CI, 0.85-0.95]) at a median 475 days (range, 93-1082 days) before clinical diagnosis. CONCLUSIONS: This automated artificial intelligence model trained on a large and diverse data set shows high accuracy and generalizable performance for detection of PDA on diagnostic CTs as well as for visually occult PDA on prediagnostic CTs. Prospective validation with blood-based biomarkers is warranted to assess the potential for early detection of sporadic PDA in high-risk individuals.

Multimodal pancreatic cancer detection using methylated DNA biomarkers in pancreatic juice and plasma CA 19-9: A prospective multicenter study.

BACKGROUND AND AIMS: In previous studies methylated DNA markers (MDMs) have been identified in pancreatic juice (PJ) for detecting pancreatic ductal adenocarcinoma (PDAC). In this prospective multicenter study, the sensitivity and specificity characteristics of this panel of PJ-MDMs was evaluated standalone and in combination with plasma CA 19-9. METHODS: Paired PJ and plasma were assayed from 88 biopsy-proven treatment naïve PDAC cases and 134 controls (normal pancreas: 53, chronic pancreatitis (CP): 23, intraductal papillary mucinous neoplasm (IPMN): 58). Bisulfite-converted DNA from buffered PJ was analyzed using long-probe quantitative amplified signal assay targeting 14 MDMs (NDRG4, BMP3, TBX15, C13orf18, PRKCB, CLEC11A, CD1D, ELMO1, IGF2BP1, RYR2, ADCY1, FER1L4, EMX1, and LRRC4) and a reference gene (methylated B3GALT6). Logistic regression was used to fit the previously identified 3-MDM PJ panel (FER1L4, C13orf18 and BMP3). Discrimination accuracy was summarized using area under the receiver operating characteristic curve (AUROC) with corresponding 95% confidence intervals. RESULTS: Methylated FER1L4 had the highest individual AUROC of 0.83 (95% CI: 0.78-0.89). The AUROC for the 3-MDM PJ + Plasma CA 19-9 model (0.95 (0.92-0.98))) was higher than both the 3-MDM PJ panel (0.87 (0.82-0.92)) and plasma CA 19-9 alone ((0.91 (0.87-0.96) (p=0.0002 and 0.0135, respectively). At a specificity of 88% (95% CI: 81-93%), the sensitivity of this model was 89% (80-94%) for all PDAC stages and 83% (64-94%) for stage I/II PDAC. CONCLUSION: A panel combining PJ-MDMs and plasma CA19-9 discriminates PDAC from both healthy and disease control groups with high accuracy. This provides support for combining pancreatic juice and blood-based biomarkers for enhancing diagnostic sensitivity and successful early PDAC detection.

Machine Learning Models for Pancreatic Cancer Risk Prediction Using Electronic Health Record Data-A Systematic Review and Assessment.

INTRODUCTION: Accurate risk prediction can facilitate screening and early detection of pancreatic cancer (PC). We conducted a systematic review to critically evaluate effectiveness of machine learning (ML) and artificial intelligence (AI) techniques applied to electronic health records (EHR) for PC risk prediction. METHODS: Ovid MEDLINE(R), Ovid EMBASE, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, Scopus, and Web of Science were searched for articles that utilized ML/AI techniques to predict PC, published between January 1, 2012, and February 1, 2024. Study selection and data extraction were conducted by 2 independent reviewers. Critical appraisal and data extraction were performed using the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies checklist. Risk of bias and applicability were examined using prediction model risk of bias assessment tool. RESULTS: Thirty studies including 169,149 PC cases were identified. Logistic regression was the most frequent modeling method. Twenty studies utilized a curated set of known PC risk predictors or those identified by clinical experts. ML model discrimination performance (C-index) ranged from 0.57 to 1.0. Missing data were underreported, and most studies did not implement explainable-AI techniques or report exclusion time intervals. DISCUSSION: AI/ML models for PC risk prediction using known risk factors perform reasonably well and may have near-term applications in identifying cohorts for targeted PC screening if validated in real-world data sets. The combined use of structured and unstructured EHR data using emerging AI models while incorporating explainable-AI techniques has the potential to identify novel PC risk factors, and this approach merits further study.

Identification of pancreatic cancer risk factors from clinical notes using natural language processing.

OBJECTIVES: Screening for pancreatic ductal adenocarcinoma (PDAC) is considered in high-risk individuals (HRIs) with established PDAC risk factors, such as family history and germline mutations in PDAC susceptibility genes. Accurate assessment of risk factor status is provider knowledge-dependent and requires extensive manual chart review by experts. Natural Language Processing (NLP) has shown promise in automated data extraction from the electronic health record (EHR). We aimed to use NLP for automated extraction of PDAC risk factors from unstructured clinical notes in the EHR. METHODS: We first developed rule-based NLP algorithms to extract PDAC risk factors at the document-level, using an annotated corpus of 2091 clinical notes. Next, we further improved the NLP algorithms using a cohort of 1138 patients through patient-level training, validation, and testing, with comparison against a pre-specified reference standard. To minimize false-negative results we prioritized algorithm recall. RESULTS: In the test set (n = 807), the NLP algorithms achieved a recall of 0.933, precision of 0.790, and F1-score of 0.856 for family history of PDAC. For germline genetic mutations, the algorithm had a high recall of 0.851, while precision and F1-score were lower at 0.350 and 0.496 respectively. Most false positives for germline mutations resulted from erroneous recognition of tissue mutations. CONCLUSIONS: Rule-based NLP algorithms applied to unstructured clinical notes are highly sensitive for automated identification of PDAC risk factors. Further validation in a large primary-care patient population is warranted to assess real-world utility in identifying HRIs for pancreatic cancer screening.

Systematic review on surveillance for non-resected branch-duct intraductal papillary mucinous neoplasms of the pancreas.

BACKGROUND: The management of branch-duct type intraductal papillary mucinous neoplasms (BD-IPMN) varies in existing guidelines. This study investigated the optimal surveillance protocol and safe discontinuation of surveillance considering natural history in non-resected IPMN, by systematically reviewing the published literature. METHODS: This review was guided by PRISMA. Research questions were framed in PICO format "CQ1-1: Is size criteria helpful to determine surveillance period? CQ1-2: How often should surveillance be carried out? CQ1-3: When should surveillance be discontinued? CQ1-4: Is nomogram predicting malignancy useful during surveillance?". PubMed was searched from January-April 2022. RESULTS: The search generated 2373 citations. After screening, 83 articles were included. Among them, 33 studies were identified for CQ1-1, 19 for CQ1-2, 26 for CQ1-3 and 12 for CQ1-4. Cysts <1.5 or 2 cm without worrisome features (WF) were described as more indolent, and most studies advised an initial period of surveillance. The median growth rate of cysts <2 cm ranged from 0.23 to 0.6 mm/year. Patients with cysts <2 cm showing no morphological changes and no WF after 5-years of surveillance have minimal malignancy risk of 0-2%. Two nomograms created with over 1000 patients had AUCs of around 0.8 and appear to be feasible in a real-world practice. CONCLUSIONS: For patients with suspected BD-IPMN <2 cm and no other WF, less frequent surveillance is recommended. Surveillance may be discontinued for cysts that remain stable during 5-year surveillance, with consideration of patient condition and life expectancy. With this updated surveillance strategy, patients with non-worrisome BD-IPMN should expect more streamlined management and decreased healthcare utilization.

The role of MRCP for preventing pancreatic fluid collection recurrence after EUS-guided drainage of walled-off necrosis.

BACKGROUND AND AIMS: Pancreatic fluid collections (PFCs) may recur after initial successful endoscopic drainage of walled-off necrosis (WON), most commonly due to disconnected pancreatic duct syndrome (DPDS). The primary aim of this study was to assess the role of MRCP for identifying DPDS to guide appropriate management and prevent PFC recurrence. METHODS: Patients with WON undergoing lumen-apposing metal stent drainage of a PFC were retrospectively identified and categorized as those with MRCP versus those without MRCP before removal of transmural stents. Data on patient demographic characteristics, procedural details, cross-sectional imaging, and recurrence rates were collected through chart review. RESULTS: A total of 121 patients with WON were identified, of whom 44 (36.4%) had an MRCP before transmural stent removal. In patients without MRCP, 13 (16.8%) of 77 had PFC recurrence versus 0 of 44 (0%; P = .003) in those with MRCP. MRCP identified DPDS in 12 (27.2%) patients, all of whom were managed with indefinite drainage with double-pigtail plastic stents without recurrence. In the group without MRCP, PFCs recurred at a median interval of 284 days (interquartile range, 182-618 days) after transmural stent removal. Among the 13 patients with PFC recurrence, 11 (85%) had undiagnosed DPDS detected on subsequent imaging, of whom 9 were subsequently managed with indefinite double-pigtail plastic stents, with no further PFC recurrence. CONCLUSIONS: Patients with WON who underwent MRCP before transmural stent removal had a lower rate of PFC recurrence largely due to the identification of DPDS with appropriate endoscopic management.

Progression of Intraductal Papillary Mucinous Neoplasms of the Pancreas After Liver Transplantation.

BACKGROUND: Intraductal papillary mucinous neoplasms (IPMNs) are the most commonly identified pancreatic cystic neoplasms. Incidentally detected IPMNs are common among liver transplant recipients. The risk of IPMN progression to pancreatic cancer in transplant recipients and the impact of immunosuppression on the risk of malignant transformation of IPMN are unclear. METHODS: In this retrospective study of consecutive liver transplant recipients across Mayo Clinic over a 13-year period, patients were assessed for possible IPMN by automated chart review. Pancreatic cystic lesions were characterized as suspected IPMNs based on imaging criteria. Cox proportional hazards models were used to determine the association between IPMN progression (the development of cancer or worrisome features) and clinical and immunosuppression regimen characteristics. RESULTS: Of 146 patients with suspected IPMNs, progression occurred in 7 patients (2 cases of IPMN-associated cancer and 5 cases of worrisome features) over an average follow-up of 66.6 months. Immunosuppression type, medication number, and tacrolimus trough levels were not associated with IPMN progression (p > 0.05). Combined kidney and liver transplantation (p = 0.005) and pretransplant cholangiocarcinoma (p = 0.012) were associated with IPMN progression. CONCLUSION: IPMN progression is rare after liver transplantation even over an extended follow-up period. The findings were notable for the absence of an association between IPMN progression and immunosuppression regimen. Larger studies are needed given the low incidence.

Radiomics-based Machine-learning Models Can Detect Pancreatic Cancer on Prediagnostic Computed Tomography Scans at a Substantial Lead Time Before Clinical Diagnosis.

BACKGROUND & AIMS: Our purpose was to detect pancreatic ductal adenocarcinoma (PDAC) at the prediagnostic stage (3-36 months before clinical diagnosis) using radiomics-based machine-learning (ML) models, and to compare performance against radiologists in a case-control study. METHODS: Volumetric pancreas segmentation was performed on prediagnostic computed tomography scans (CTs) (median interval between CT and PDAC diagnosis: 398 days) of 155 patients and an age-matched cohort of 265 subjects with normal pancreas. A total of 88 first-order and gray-level radiomic features were extracted and 34 features were selected through the least absolute shrinkage and selection operator-based feature selection method. The dataset was randomly divided into training (292 CTs: 110 prediagnostic and 182 controls) and test subsets (128 CTs: 45 prediagnostic and 83 controls). Four ML classifiers, k-nearest neighbor (KNN), support vector machine (SVM), random forest (RM), and extreme gradient boosting (XGBoost), were evaluated. Specificity of model with highest accuracy was further validated on an independent internal dataset (n = 176) and the public National Institutes of Health dataset (n = 80). Two radiologists (R4 and R5) independently evaluated the pancreas on a 5-point diagnostic scale. RESULTS: Median (range) time between prediagnostic CTs of the test subset and PDAC diagnosis was 386 (97-1092) days. SVM had the highest sensitivity (mean; 95% confidence interval) (95.5; 85.5-100.0), specificity (90.3; 84.3-91.5), F1-score (89.5; 82.3-91.7), area under the curve (AUC) (0.98; 0.94-0.98), and accuracy (92.2%; 86.7-93.7) for classification of CTs into prediagnostic versus normal. All 3 other ML models, KNN, RF, and XGBoost, had comparable AUCs (0.95, 0.95, and 0.96, respectively). The high specificity of SVM was generalizable to both the independent internal (92.6%) and the National Institutes of Health dataset (96.2%). In contrast, interreader radiologist agreement was only fair (Cohen's kappa 0.3) and their mean AUC (0.66; 0.46-0.86) was lower than each of the 4 ML models (AUCs: 0.95-0.98) (P < .001). Radiologists also recorded false positive indirect findings of PDAC in control subjects (n = 83) (7% R4, 18% R5). CONCLUSIONS: Radiomics-based ML models can detect PDAC from normal pancreas when it is beyond human interrogation capability at a substantial lead time before clinical diagnosis. Prospective validation and integration of such models with complementary fluid-based biomarkers has the potential for PDAC detection at a stage when surgical cure is a possibility.

Artificial intelligence assisted whole organ pancreatic fat estimation on magnetic resonance imaging and correlation with pancreas attenuation on computed tomography.

BACKGROUND: Fatty pancreas is associated with inflammatory and neoplastic pancreatic diseases. Magnetic resonance imaging (MRI) is the diagnostic modality of choice for measuring pancreatic fat. Measurements typically use regions of interest limited by sampling and variability. We have previously described an artificial intelligence (AI)-aided approach for whole pancreas fat estimation on computed tomography (CT). In this study, we aimed to assess the correlation between whole pancreas MRI proton-density fat fraction (MR-PDFF) and CT attenuation. METHODS: We identified patients without pancreatic disease who underwent both MRI and CT between January 1, 2015 and June 1, 2020. 158 paired MRI and CT scans were available for pancreas segmentation using an iteratively trained convolutional neural network (CNN) with manual correction. Boxplots were generated to visualize slice-by-slice variability in 2D-axial slice MR-PDFF. Correlation between whole pancreas MR-PDFF and age, BMI, hepatic fat and pancreas CT-Hounsfield Unit (CT-HU) was assessed. RESULTS: Mean pancreatic MR-PDFF showed a strong inverse correlation (Spearman -0.755) with mean CT-HU. MR-PDFF was higher in males (25.22 vs 20.87; p = 0.0015) and in subjects with diabetes mellitus (25.95 vs 22.17; p = 0.0324), and was positively correlated with age and BMI. The pancreatic 2D-axial slice-to-slice MR-PDFF variability increased with increasing mean whole pancreas MR-PDFF (Spearman 0.51; p < 0.0001). CONCLUSION: Our study demonstrates a strong inverse correlation between whole pancreas MR-PDFF and CT-HU, indicating that both imaging modalities can be used to assess pancreatic fat. 2D-axial pancreas MR-PDFF is variable across slices, underscoring the need for AI-aided whole-organ measurements for objective and reproducible estimation of pancreatic fat.

Bounding box-based 3D AI model for user-guided volumetric segmentation of pancreatic ductal adenocarcinoma on standard-of-care CTs.

OBJECTIVES: To develop a bounding-box-based 3D convolutional neural network (CNN) for user-guided volumetric pancreas ductal adenocarcinoma (PDA) segmentation. METHODS: Reference segmentations were obtained on CTs (2006-2020) of treatment-naïve PDA. Images were algorithmically cropped using a tumor-centered bounding box for training a 3D nnUNet-based-CNN. Three radiologists independently segmented tumors on test subset, which were combined with reference segmentations using STAPLE to derive composite segmentations. Generalizability was evaluated on Cancer Imaging Archive (TCIA) (n = 41) and Medical Segmentation Decathlon (MSD) (n = 152) datasets. RESULTS: Total 1151 patients [667 males; age:65.3 ± 10.2 years; T1:34, T2:477, T3:237, T4:403; mean (range) tumor diameter:4.34 (1.1-12.6)-cm] were randomly divided between training/validation (n = 921) and test subsets (n = 230; 75% from other institutions). Model had a high DSC (mean ± SD) against reference segmentations (0.84 ± 0.06), which was comparable to its DSC against composite segmentations (0.84 ± 0.11, p = 0.52). Model-predicted versus reference tumor volumes were comparable (mean ± SD) (29.1 ± 42.2-cc versus 27.1 ± 32.9-cc, p = 0.69, CCC = 0.93). Inter-reader variability was high (mean DSC 0.69 ± 0.16), especially for smaller and isodense tumors. Conversely, model's high performance was comparable between tumor stages, volumes and densities (p > 0.05). Model was resilient to different tumor locations, status of pancreatic/biliary ducts, pancreatic atrophy, CT vendors and slice thicknesses, as well as to the epicenter and dimensions of the bounding-box (p > 0.05). Performance was generalizable on MSD (DSC:0.82 ± 0.06) and TCIA datasets (DSC:0.84 ± 0.08). CONCLUSION: A computationally efficient bounding box-based AI model developed on a large and diverse dataset shows high accuracy, generalizability, and robustness to clinically encountered variations for user-guided volumetric PDA segmentation including for small and isodense tumors. CLINICAL RELEVANCE: AI-driven bounding box-based user-guided PDA segmentation offers a discovery tool for image-based multi-omics models for applications such as risk-stratification, treatment response assessment, and prognostication, which are urgently needed to customize treatment strategies to the unique biological profile of each patient's tumor.

Utility of Urgent Endoscopic Retrograde Cholangiopancreatography in Patients with Predicted Mild Acute Pancreatitis and Cholestasis.

BACKGROUND: Endoscopic retrograde cholangiopancreatography (ERCP) within 72 h is suggested for patients presenting with acute biliary pancreatitis (ABP) and biliary obstruction without cholangitis. This study aimed to identify if urgent ERCP (within 24 h) improved outcomes compared to early ERCP (24-72 h) in patients admitted with predicted mild ABP. METHODS: Patients admitted for predicted mild ABP defined as a bedside index of severity in acute pancreatitis score < 3 and underwent ERCP for biliary obstruction within 72 h of presentation during the study period were included. Patients with prior biliary sphincterotomy or surgically altered anatomy preventing conventional ERCP were excluded. The primary outcome was the development of moderately severe or severe pancreatitis based on the revised Atlanta classification. Secondary outcomes were the length of hospital stay, the need for ICU admission, and ERCP-related adverse events (AEs). RESULTS: Of the identified 166 patients, baseline characteristics were similar between both the groups except for the WBC count (9.4 vs. 8.3/µL; p < 0.044) and serum bilirubin level (3.0 vs. 1.6 mg/dL; p < 0.0039). Biliary cannulation rate and technical success were both high in the overall cohort (98.8%). Urgent ERCP was not associated with increased development of moderately severe pancreatitis (10.4% vs. 15.7%; p = 0.3115). The urgent ERCP group had a significantly shorter length of hospital stay [median 3 (IQR 2-3) vs. 3 days (IQR 3-4), p < 0.01]. CONCLUSION: Urgent ERCP did not impact the rate of developing more severe pancreatitis in patients with predicted mild ABP but was associated with a shorter length of hospital stay and a lower rate of hospital readmission.

Prospective Study Assessing Impact of Ethylene Oxide Sterilization on Endoscopic Ultrasound Image Quality.

BACKGROUND & AIMS: Duodenoscope-associated transmission of infections has raised questions about efficacy of endoscope reprocessing using high-level disinfection (HLD). Although ethylene oxide (ETO) gas sterilization is effective in eradicating microbes, the impact of ETO on endoscopic ultrasound (EUS) imaging equipment remains unknown. In this study, we aimed to compare the changes in EUS image quality associated with HLD vs HLD followed by ETO sterilization. METHODS: Four new EUS instruments were assigned to 2 groups: Group 1 (HLD) and Group 2 (HLD + ETO). The echoendoscopes were assessed at baseline, monthly for 6 months, and once every 3 to 4 months thereafter, for a total of 12 time points. At each time point, review of EUS video and still image quality was performed by an expert panel of reviewers along with phantom-based objective testing. Linear mixed effects models were used to assess whether the modality of reprocessing impacted image and video quality. RESULTS: For clinical testing, mixed linear models showed minimal quantitative differences in linear analog score (P = .04; estimated change, 3.12; scale, 0-100) and overall image quality value (P = .007; estimated change, -0.12; scale, 1-5) favoring ETO but not for rank value (P = .06). On phantom testing, maximum depth of penetration was lower for ETO endoscopes (P < .001; change in depth, 0.49 cm). CONCLUSIONS: In this prospective study, expert review and phantom-based testing demonstrated minimal differences in image quality between echoendoscopes reprocessed using HLD vs ETO + HLD over 2 years of clinical use. Further studies are warranted to assess the long-term clinical impact of these findings. In the interim, these results support use of ETO sterilization of EUS instruments if deemed clinically necessary.

Pancreatic cancer risk to siblings of probands in bilineal cancer settings.

PURPOSE: Pancreatic cancer (PC) risk is increased in families, but PC risk and risk perception have been understudied when both parents have cancer. METHODS: An unbiased method defining cancer triads (proband with PC and both parents with cancer) in a prospective registry estimated risk of PC to probands' siblings in triad group 1 (no parent with PC), group 2 (1 parent with PC), and group 3 (both parents with PC). We estimated standardized incidence ratios (SIRs) using a Surveillance, Epidemiology, and End Results (SEER) reference. We also estimated the risk when triad probands carried germline pathogenic/likely pathogenic variants in any of the 6 PC-associated genes (ATM, BRCA1, BRCA2, CDKN2A, MLH1, and TP53). PC risk perception/concern was surveyed in siblings and controls. RESULTS: Risk of PC was higher (SIR = 3.5; 95% CI = 2.2-5.2) in 933 at-risk siblings from 297 triads. Risk increased by triad group: 2.8 (95% CI = 1.5-4.5); 4.5 (95% CI = 1.6-9.7); and 21.2 (95% CI = 4.3-62.0). SIR in variant-negative triads was 3.0 (95% CI = 1.6-5.0), whereas SIR in variant-positive triads was 10.0 (95% CI = 3.2-23.4). Siblings' perceived risk/concern of developing PC increased by triad group. CONCLUSION: Sibling risks were 2.8- to 21.2-fold higher than that of the general population. Positive variant status increased the risk in triads. Increasing number of PC cases in a triad was associated with increased concern and perceived PC risk.

Tissue methylated DNA markers for sporadic pancreatic cancer are strongly associated with familial and genetically predisposed pancreatic cancer.

High-risk individuals (HRIs) with familial and genetic predisposition to pancreatic ductal adenocarcinoma (PDAC) are eligible for screening. There is no accurate biomarker for detecting early-stage PDAC. We previously demonstrated that a panel of methylated DNA markers (MDMs) accurately detect sporadic PDAC. In this study we compared the distribution of MDMs in DNA extracted from tissue of PDAC cases who carry germline mutations and non-carriers with family history, with control tissue and demonstrate high discrimination like that seen in sporadic PDAC. These results provide scientific rationale for examining plasma MDMs in HRIs with the goal of developing a minimally-invasive early detection test.

Volumetric Pancreas Segmentation on Computed Tomography: Accuracy and Efficiency of a Convolutional Neural Network Versus Manual Segmentation in 3D Slicer in the Context of Interreader Variability of Expert Radiologists.

PURPOSE: This study aimed to compare accuracy and efficiency of a convolutional neural network (CNN)-enhanced workflow for pancreas segmentation versus radiologists in the context of interreader reliability. METHODS: Volumetric pancreas segmentations on a data set of 294 portal venous computed tomographies were performed by 3 radiologists (R1, R2, and R3) and by a CNN. Convolutional neural network segmentations were reviewed and, if needed, corrected ("corrected CNN [c-CNN]" segmentations) by radiologists. Ground truth was obtained from radiologists' manual segmentations using simultaneous truth and performance level estimation algorithm. Interreader reliability and model's accuracy were evaluated with Dice-Sorenson coefficient (DSC) and Jaccard coefficient (JC). Equivalence was determined using a two 1-sided test. Convolutional neural network segmentations below the 25th percentile DSC were reviewed to evaluate segmentation errors. Time for manual segmentation and c-CNN was compared. RESULTS: Pancreas volumes from 3 sets of segmentations (manual, CNN, and c-CNN) were noninferior to simultaneous truth and performance level estimation-derived volumes [76.6 cm 3 (20.2 cm 3 ), P < 0.05]. Interreader reliability was high (mean [SD] DSC between R2-R1, 0.87 [0.04]; R3-R1, 0.90 [0.05]; R2-R3, 0.87 [0.04]). Convolutional neural network segmentations were highly accurate (DSC, 0.88 [0.05]; JC, 0.79 [0.07]) and required minimal-to-no corrections (c-CNN: DSC, 0.89 [0.04]; JC, 0.81 [0.06]; equivalence, P < 0.05). Undersegmentation (n = 47 [64%]) was common in the 73 CNN segmentations below 25th percentile DSC, but there were no major errors. Total inference time (minutes) for CNN was 1.2 (0.3). Average time (minutes) taken by radiologists for c-CNN (0.6 [0.97]) was substantially lower compared with manual segmentation (3.37 [1.47]; savings of 77.9%-87% [ P < 0.0001]). CONCLUSIONS: Convolutional neural network-enhanced workflow provides high accuracy and efficiency for volumetric pancreas segmentation on computed tomography.

Utilisation of artificial intelligence for the development of an EUS-convolutional neural network model trained to enhance the diagnosis of autoimmune pancreatitis.

OBJECTIVE: The diagnosis of autoimmune pancreatitis (AIP) is challenging. Sonographic and cross-sectional imaging findings of AIP closely mimic pancreatic ductal adenocarcinoma (PDAC) and techniques for tissue sampling of AIP are suboptimal. These limitations often result in delayed or failed diagnosis, which negatively impact patient management and outcomes. This study aimed to create an endoscopic ultrasound (EUS)-based convolutional neural network (CNN) model trained to differentiate AIP from PDAC, chronic pancreatitis (CP) and normal pancreas (NP), with sufficient performance to analyse EUS video in real time. DESIGN: A database of still image and video data obtained from EUS examinations of cases of AIP, PDAC, CP and NP was used to develop a CNN. Occlusion heatmap analysis was used to identify sonographic features the CNN valued when differentiating AIP from PDAC. RESULTS: From 583 patients (146 AIP, 292 PDAC, 72 CP and 73 NP), a total of 1 174 461 unique EUS images were extracted. For video data, the CNN processed 955 EUS frames per second and was: 99% sensitive, 98% specific for distinguishing AIP from NP; 94% sensitive, 71% specific for distinguishing AIP from CP; 90% sensitive, 93% specific for distinguishing AIP from PDAC; and 90% sensitive, 85% specific for distinguishing AIP from all studied conditions (ie, PDAC, CP and NP). CONCLUSION: The developed EUS-CNN model accurately differentiated AIP from PDAC and benign pancreatic conditions, thereby offering the capability of earlier and more accurate diagnosis. Use of this model offers the potential for more timely and appropriate patient care and improved outcome.

Predicting the Need for Step-Up Therapy After EUS-Guided Drainage of Pancreatic Fluid Collections With Lumen-Apposing Metal Stents.

BACKGROUND & AIMS: A significant proportion of individuals with pancreatic fluid collections (PFCs) require step-up therapy after endoscopic drainage with lumen-apposing metal stents. The aim of this study is to identify factors associated with PFCs that require step-up therapy. METHODS: A retrospective cohort study of patients undergoing endoscopic ultrasound-guided drainage of PFCs with lumen-apposing metal stents from April 2014 to October 2019 at a single center was performed. Step-up therapy included direct endoscopic necrosectomy, additional drainage site (endoscopic or percutaneous), or surgical intervention after the initial drainage procedure. Multivariable logistic regression was performed using a backward stepwise approach with a P ≤ .2 threshold for variable retention to identify factors predictive for the need for step-up therapy. RESULTS: One hundred thirty-six patients were included in the final study cohort, of whom 69 (50.7%) required step-up therapy. Independent predictors of step-up therapy included: collection size measuring ≥10 cm (odds ratio [OR], 8.91; 95% confidence interval [CI], 3.36-23.61), paracolic extension of the PFC (OR, 4.04; 95% CI, 1.60-10.23), and ≥30% solid necrosis (OR, 4.24; 95% CI, 1.48-12.16). In a sensitivity analysis of 81 patients with walled-off necrosis, 51 (63.0%) required step-up therapy. Similarly, factors predictive of the need for step-up therapy for walled-off necrosis included: collection size measuring ≥10 cm (OR, 6.94; 95% CI, 1.76-27.45), paracolic extension of the PFC (OR, 3.79; 95% CI, 1.18-12.14), and ≥30% solid necrosis (OR, 7.10; 95% CI, 1.16-43.48). CONCLUSIONS: Half of all patients with PFCs drained with lumen-apposing metal stents required step-up therapy, most commonly direct endoscopic necrosectomy. Individuals with PFCs ≥10 cm in size, paracolic extension, or ≥30% solid necrosis are more likely to require step-up therapy and should be considered for early endoscopic reintervention.

Early detection of pancreatic cancer: current state and future opportunities.

PURPOSE OF REVIEW: Pancreatic ductal adenocarcinoma (PDAC) is third leading cause of cancer death in the United States, a lethal disease with no screening strategy. Although diagnosis at an early stage is associated with improved survival, clinical detection of PDAC is typically at an advanced symptomatic stage when best in class therapies have limited impact on survival. RECENT FINDINGS: In recent years this status quo has been challenged by the identification of novel risk factors, molecular markers of early-stage disease and innovations in pancreatic imaging. There is now expert consensus that screening may be pursued in a cohort of individuals with increased likelihood of developing PDAC based on genetic and familial risk. SUMMARY: The current review summarizes the known risk factors of PDAC, current knowledge and recent observations pertinent to early detection of PDAC in these risk groups and outlines future approaches that will potentially advance the field.

Quality gaps in public pancreas imaging datasets: Implications & challenges for AI applications.

OBJECTIVE: Quality gaps in medical imaging datasets lead to profound errors in experiments. Our objective was to characterize such quality gaps in public pancreas imaging datasets (PPIDs), to evaluate their impact on previously published studies, and to provide post-hoc labels and segmentations as a value-add for these PPIDs. METHODS: We scored the available PPIDs on the medical imaging data readiness (MIDaR) scale, and evaluated for associated metadata, image quality, acquisition phase, etiology of pancreas lesion, sources of confounders, and biases. Studies utilizing these PPIDs were evaluated for awareness of and any impact of quality gaps on their results. Volumetric pancreatic adenocarcinoma (PDA) segmentations were performed for non-annotated CTs by a junior radiologist (R1) and reviewed by a senior radiologist (R3). RESULTS: We found three PPIDs with 560 CTs and six MRIs. NIH dataset of normal pancreas CTs (PCT) (n = 80 CTs) had optimal image quality and met MIDaR A criteria but parts of pancreas have been excluded in the provided segmentations. TCIA-PDA (n = 60 CTs; 6 MRIs) and MSD(n = 420 CTs) datasets categorized to MIDaR B due to incomplete annotations, limited metadata, and insufficient documentation. Substantial proportion of CTs from TCIA-PDA and MSD datasets were found unsuitable for AI due to biliary stents [TCIA-PDA:10 (17%); MSD:112 (27%)] or other factors (non-portal venous phase, suboptimal image quality, non-PDA etiology, or post-treatment status) [TCIA-PDA:5 (8.5%); MSD:156 (37.1%)]. These quality gaps were not accounted for in any of the 25 studies that have used these PPIDs (NIH-PCT:20; MSD:1; both: 4). PDA segmentations were done by R1 in 91 eligible CTs (TCIA-PDA:42; MSD:49). Of these, corrections were made by R3 in 16 CTs (18%) (TCIA-PDA:4; MSD:12) [mean (standard deviation) Dice: 0.72(0.21) and 0.63(0.23) respectively]. CONCLUSION: Substantial quality gaps, sources of bias, and high proportion of CTs unsuitable for AI characterize the available limited PPIDs. Published studies on these PPIDs do not account for these quality gaps. We complement these PPIDs through post-hoc labels and segmentations for public release on the TCIA portal. Collaborative efforts leading to large, well-curated PPIDs supported by adequate documentation are critically needed to translate the promise of AI to clinical practice.

Determining age and sex-specific distribution of pancreatic whole-gland CT attenuation using artificial intelligence aided image segmentation: Associations with body composition and pancreatic cancer risk.

BACKGROUND & AIMS: Increased intrapancreatic fat is associated with pancreatic diseases; however, there are no established objective diagnostic criteria for fatty pancreas. On non-contrast computed tomography (CT), adipose tissue shows negative Hounsfield Unit (HU) attenuations (-150 to -30 HU). Using whole organ segmentation on non-contrast CT, we aimed to describe whole gland pancreatic attenuation and establish 5th and 10th percentile thresholds across a spectrum of age and sex. Subsequently, we aimed to evaluate the association between low pancreatic HU and risk of pancreatic ductal adenocarcinoma (PDAC). METHODS: The whole pancreas was segmented in 19,456 images from 469 non-contrast CT scans. A convolutional neural network was trained to assist pancreas segmentation. Mean pancreatic HU, volume, and body composition metrics were calculated. The lower 5th and 10th percentile for mean pancreatic HU were identified, examining the association with age and sex. Pre-diagnostic CT scans from patients who later developed PDAC were compared to cancer-free controls. RESULTS: Less than 5th percentile mean pancreatic HU was significantly associated with increase in BMI (OR 1.07; 1.03-1.11), visceral fat (OR 1.37; 1.15-1.64), total abdominal fat (OR 1.12; 1.03-1.22), and diabetes mellitus type 1 (OR 6.76; 1.68-27.28). Compared to controls, pre-diagnostic scans in PDAC cases had lower mean whole gland pancreatic HU (-0.2 vs 7.8, p = 0.026). CONCLUSION: In this study, we report age and sex-specific distribution of pancreatic whole-gland CT attenuation. Compared to controls, mean whole gland pancreatic HU is significantly lower in the pre-diagnostic phase of PDAC.