Showing papers by "Bechien U. Wu published in 2022"

Predicting pancreatic ductal adenocarcinoma using artificial intelligence analysis of pre-diagnostic computed tomography images.

Abstract: BACKGROUND Early stage diagnosis of Pancreatic Ductal Adenocarcinoma (PDAC) is challenging due to the lack of specific diagnostic biomarkers. However, stratifying individuals at high risk of PDAC, followed by monitoring their health conditions on regular basis, has the potential to allow diagnosis at early stages. OBJECTIVE To stratify high risk individuals for PDAC by identifying predictive features in pre-diagnostic abdominal Computed Tomography (CT) scans. METHODS A set of CT features, potentially predictive of PDAC, was identified in the analysis of 4000 raw radiomic parameters extracted from pancreases in pre-diagnostic scans. The naïve Bayes classifier was then developed for automatic classification of CT scans of the pancreas with high risk for PDAC. A set of 108 retrospective CT scans (36 scans from each healthy control, pre-diagnostic, and diagnostic group) from 72 subjects was used for the study. Model development was performed on 66 multiphase CT scans, whereas external validation was performed on 42 venous-phase CT scans. RESULTS The system achieved an average classification accuracy of 86% on the external dataset. CONCLUSIONS Radiomic analysis of abdominal CT scans can unveil, quantify, and interpret micro-level changes in the pre-diagnostic pancreas and can efficiently assist in the stratification of high risk individuals for PDAC.

Risk Prediction of Pancreatic Cancer in Patients With Abnormal Morphologic Findings Related to Chronic Pancreatitis: A Machine Learning Approach

Abstract: Background and AimsA significant factor contributing to poor survival in pancreatic cancer is the often late stage at diagnosis. We sought to develop and validate a risk prediction model to facilitate the distinction between chronic pancreatitis–related vs potential early pancreatic ductal adenocarcinoma (PDAC)-associated changes on pancreatic imaging.MethodsIn this retrospective cohort study, patients aged 18–84 years whose abdominal computed tomography/magnetic resonance imaging reports indicated duct dilatation, atrophy, calcification, cyst, or pseudocyst between January 2008 and November 2019 were identified. The outcome of interest is PDAC in 3 years. More than 100 potential predictors were extracted. Random survival forests approach was used to develop and validate risk models. Multivariable Cox proportional hazard model was applied to estimate the effect of the covariates on the risk of PDAC.ResultsThe cohort consisted of 46,041 (mean age 66.4 years). The 3-year incidence rate was 4.0 (95% confidence interval CI 3.6–4.4)/1000 person-years of follow-up. The final models containing age, weight change, duct dilatation, and either alkaline phosphatase or total bilirubin had good discrimination and calibration (c-indices 0.81). Patients with pancreas duct dilatation and at least another morphological feature in the absence of calcification had the highest risk (adjusted hazard ratio [aHR] = 14.15, 95% CI 8.7–22.6), followed by patients with calcification and duct dilatation (aHR = 7.28, 95% CI 4.09–12.96), and patients with duct dilation only (aHR = 6.22, 95% CI 3.86–10.03), compared with patients with calcifications alone as the reference group.ConclusionThe study characterized the risk of pancreatic cancer among patients with 5 abnormal morphologic findings based on radiology reports and demonstrated the ability of prediction algorithms to provide improved risk stratification of pancreatic cancer in these patients. A significant factor contributing to poor survival in pancreatic cancer is the often late stage at diagnosis. We sought to develop and validate a risk prediction model to facilitate the distinction between chronic pancreatitis–related vs potential early pancreatic ductal adenocarcinoma (PDAC)-associated changes on pancreatic imaging. In this retrospective cohort study, patients aged 18–84 years whose abdominal computed tomography/magnetic resonance imaging reports indicated duct dilatation, atrophy, calcification, cyst, or pseudocyst between January 2008 and November 2019 were identified. The outcome of interest is PDAC in 3 years. More than 100 potential predictors were extracted. Random survival forests approach was used to develop and validate risk models. Multivariable Cox proportional hazard model was applied to estimate the effect of the covariates on the risk of PDAC. The cohort consisted of 46,041 (mean age 66.4 years). The 3-year incidence rate was 4.0 (95% confidence interval CI 3.6–4.4)/1000 person-years of follow-up. The final models containing age, weight change, duct dilatation, and either alkaline phosphatase or total bilirubin had good discrimination and calibration (c-indices 0.81). Patients with pancreas duct dilatation and at least another morphological feature in the absence of calcification had the highest risk (adjusted hazard ratio [aHR] = 14.15, 95% CI 8.7–22.6), followed by patients with calcification and duct dilatation (aHR = 7.28, 95% CI 4.09–12.96), and patients with duct dilation only (aHR = 6.22, 95% CI 3.86–10.03), compared with patients with calcifications alone as the reference group. The study characterized the risk of pancreatic cancer among patients with 5 abnormal morphologic findings based on radiology reports and demonstrated the ability of prediction algorithms to provide improved risk stratification of pancreatic cancer in these patients.

Gastric cancer screening in first-degree relatives: a pilot study in a diverse integrated healthcare system.

Abstract: ABSTRACT BACKGROUNDFamily history of gastric cancer has been shown as an independent risk factor for gastric cancer development and is associated with increased risk of progression to gastric cancer among patients with gastric intestinal metaplasia (GIM).METHODSBetween 2017 and 2020, we conducted a prospective pilot screening program of patients with a confirmed first-degree relative with gastric cancer to evaluate the feasibility of screening and the prevalence of precursor lesions (e.g., GIM or dysplasia) on biopsy.RESULTSA total of 61 patients completed screening by upper endoscopy with mapping biopsy protocol: 27 (44%) were found to have GIM and 4 (7%) were found with low-grade dysplasia.DISCUSSIONOur pilot screening program identified a high prevalence of precursor lesions for gastric cancer among asymptomatic patients with a first-degree relative of gastric cancer. Careful endoscopic inspection and standardized biopsy protocols may aid in prompt identification of these precursor lesions in those at-risk for gastric cancer.

Quantitative Radiomic Features From Computed Tomography Can Predict Pancreatic Cancer up to 36 Months Before Diagnosis

Abstract: Objectives: Pancreatic cancer (PC) is the 3rd leading cause of cancer deaths. We aimed to detect early changes on computed tomography (CT) images associated with pancreatic ductal adenocarcinoma (PDAC) based on quantitative imaging features (QIF). Methods: Adults 18+ years of age diagnosed with PDAC in 2008-2018 were identified. Their CT scans 3 months-3 years prior to the diagnosis date were matched to up to two scans of controls. Pancreas was automatically segmented using a previously developed algorithm. 111 QIF were extracted. The dataset was randomly split for training/validation. Neighborhood and principal component analyses were applied to select the most important features. Conditional support vector machine was used to develop prediction algorithms. The computer labels were compared with manually reviewed CT images 2-3 years prior to the index date in 19 cases and 19 controls. Results: 227 scans from cases (stages: 35% I-II, 44% III-IV, 21% unknown) and 554 matched scans of healthy controls were included (average age 71 years; 51% females). In the validation dataset, accuracy measures were 94%-95%, and area under the curve (AUC) measures were 0.98-0.99. Sensitivity, specificity, positive predictive value, and negative predictive values were in the ranges of 88-91%, 96-98%, 91-95%, and 94-96%. QIF on CT examinations within 2-3 years prior to index date also had very high predictive accuracy (accuracy 95-98%; AUC 0.99-1.00). The QIF-based algorithm outperformed manual re-review of images for determination of PDAC-risk. Conclusions: QIF can accurately predict PDAC on CT imaging and represent promising biomarkers for early detection of pancreatic cancer.

Imaging of the Pancreas in New-Onset Diabetes: A Prospective Pilot Study

Abstract: INTRODUCTION: The aim of this study was to assess the feasibility of cross-sectional imaging for detection of pancreatic cancer (PDAC) in patients with new-onset hyperglycemia and diabetes (NOD). METHODS: We conducted a prospective pilot study from November 2018 to March 2020 within an integrated health system. Patients aged 50–85 years with newly elevated glycemic parameters without a history of diabetes were invited to complete a 3-phase contrast-enhanced computed tomography pancreas protocol scan while participating in the Prospective Study to Establish a NOD Cohort. Abnormal pancreatic findings, incidental extrapancreatic findings, and subsequent clinical evaluation were identified. Variability in clinical reporting between medical centers based on descriptors of pancreatic duct and parenchyma was assessed. RESULTS: A total of 130 of 147 participants (88.4%) consented to imaging; 93 scans were completed (before COVID-19 stay-at-home order). The median age was 62.4 years (interquartile range 56.3–68.8), 37.6% women; Hispanic (39.8%), White (29.0%), Black (14.0%), and Asian (13.3%). One (1.1%) case of PDAC (stage IV) was diagnosed, 12 of 93 participants (12.9%) had additional pancreatic findings: 5 fatty infiltration, 3 cysts, 2 atrophy, 1 divisum, and 1 calcification. There were 57 extrapancreatic findings among 52 of 93 (56%) unique patients; 12 of 57 (21.1%) prompted clinical evaluation with 2 additional malignancies diagnosed (nonsmall cell lung and renal oncocytoma). Reports from 1 participating medical center more frequently provided description of pancreatic parenchyma and ducts (92.9% vs 18.4%), P < 0.0001. DISCUSSION: High proportion of incidental findings and variability in clinical reports are challenges to be addressed for a successful NOD-based early detection strategy for PDAC.

Clinical and imaging predictors of pancreatic cancer in patients hospitalized for acute pancreatitis

Abstract: Background and AimsIdentifying factors associated with increased short-term risk of pancreatic cancer in the setting of acute pancreatitis (AP) can inform clinical care decisions and expedite cancer diagnosis.MethodsA retrospective cohort study of patients hospitalized for AP between 2007 and 2017 in an integrated health-care system in Southern California. AP cases were identified by diagnosis code with laboratory confirmation. Multivariable Cox proportional hazards regression model was used to assess risk of pancreatic cancer within 3 years of AP, adjusting for patient demographics, clinical parameters (body mass index, AP etiology, chronic pancreatitis, diabetes) and radiographic imaging features.ResultsAmong 9,490 patients hospitalized with AP, the mean (standard deviation) age was 55.8 (17.8) years, 55% were women, and 42% were Hispanic. Majority of AP cases were biliary (64%), 12% were alcohol-related, 5% were hypertriglyceridemia-induced, and 19% were other/unknown etiology. Ninety-five (1%) patients were diagnosed with pancreatic cancer within 3 years of AP (4.2 cases/1000 person-years). Risk factors for pancreatic cancer were age ≥65 years (hazard risk [HR]: 2.5, 95% confidence interval [CI]: 1.2–5.3), male sex (HR: 1.9, 95% CI: 1.2–2.8), Asian/Pacific Islander race (HR: 2.0, 95% CI: 1.1–3.6), and underweight body mass index (HR: 2.6, 95% CI: 1.1–6.5). In addition, other/unknown AP etiology (HR: 2.0, 95% CI: 1.3–3.1) and dilatation of the main pancreatic duct (HR: 6.6, 95% CI: 4.2–10.5) were independently associated with increased risk of pancreatic cancer.ConclusionIn addition to older age, the lack of well-established etiology, underweight body habitus, and main pancreatic duct dilatation were independently associated with increased short-term risk of pancreatic cancer among patients hospitalized for AP. Identifying factors associated with increased short-term risk of pancreatic cancer in the setting of acute pancreatitis (AP) can inform clinical care decisions and expedite cancer diagnosis. A retrospective cohort study of patients hospitalized for AP between 2007 and 2017 in an integrated health-care system in Southern California. AP cases were identified by diagnosis code with laboratory confirmation. Multivariable Cox proportional hazards regression model was used to assess risk of pancreatic cancer within 3 years of AP, adjusting for patient demographics, clinical parameters (body mass index, AP etiology, chronic pancreatitis, diabetes) and radiographic imaging features. Among 9,490 patients hospitalized with AP, the mean (standard deviation) age was 55.8 (17.8) years, 55% were women, and 42% were Hispanic. Majority of AP cases were biliary (64%), 12% were alcohol-related, 5% were hypertriglyceridemia-induced, and 19% were other/unknown etiology. Ninety-five (1%) patients were diagnosed with pancreatic cancer within 3 years of AP (4.2 cases/1000 person-years). Risk factors for pancreatic cancer were age ≥65 years (hazard risk [HR]: 2.5, 95% confidence interval [CI]: 1.2–5.3), male sex (HR: 1.9, 95% CI: 1.2–2.8), Asian/Pacific Islander race (HR: 2.0, 95% CI: 1.1–3.6), and underweight body mass index (HR: 2.6, 95% CI: 1.1–6.5). In addition, other/unknown AP etiology (HR: 2.0, 95% CI: 1.3–3.1) and dilatation of the main pancreatic duct (HR: 6.6, 95% CI: 4.2–10.5) were independently associated with increased risk of pancreatic cancer. In addition to older age, the lack of well-established etiology, underweight body habitus, and main pancreatic duct dilatation were independently associated with increased short-term risk of pancreatic cancer among patients hospitalized for AP.

Risk Prediction of Pancreatic Cancer in Patients With Recent-onset Hyperglycemia

Abstract: Background: New-onset diabetes (NOD) has been suggested as an early indicator of pancreatic cancer. However, the definition of NOD by the American Diabetes Association requires 2 simultaneous or consecutive elevated glycemic measures. We aimed to apply a machine-learning approach using electronic health records to predict the risk in patients with recent-onset hyperglycemia. Materials and Methods: In this retrospective cohort study, health plan enrollees 50 to 84 years of age who had an elevated (6.5%+) glycated hemoglobin (HbA1c) tested in January 2010 to September 2018 with recent-onset hyperglycemia were identified. A total of 102 potential predictors were extracted. Ten imputation datasets were generated to handle missing data. The random survival forests approach was used to develop and validate risk models. Performance was evaluated by c-index, calibration plot, sensitivity, specificity, and positive predictive value. Results: The cohort consisted of 109,266 patients (mean age: 63.6 y). The 3-year incidence rate was 1.4 (95% confidence interval: 1.3-1.6)/1000 person-years of follow-up. The 3 models containing age, weight change in 1 year, HbA1c, and 1 of the 3 variables (HbA1c change in 1 y, HbA1c in the prior 6 mo, or HbA1c in the prior 18 mo) appeared most often out of the 50 training samples. The c-indexes were in the range of 0.81 to 0.82. The sensitivity, specificity, and positive predictive value in patients who had the top 20% of the predicted risks were 56% to 60%, 80%, and 2.5% to 2.6%, respectively. Conclusion: Targeting evaluation at the point of recent hyperglycemia based on elevated HbA1c could offer an opportunity to identify pancreatic cancer early and possibly impact survival in cancer patients.

Satisfaction With Telehealth Among Patients With Irritable Bowel Syndrome During COVID-19: Cross-Sectional Survey Findings

Abstract: INTRODUCTION: Coronavirus disease 2019 rapidly shifted health care toward telehealth. We assessed satisfaction with and preferences for telehealth among patients with irritable bowel syndrome (IBS). METHODS: We conducted a cross-sectional survey in an integrated healthcare system in Southern California with members aged 18–90 years with an International Classification of Diseases 9 and 10 codes for IBS from office-based encounters between June 1, 2018, and June 1, 2020. Eligible patients were emailed a survey assessing telehealth satisfaction overall and by patient-related factors, IBS characteristics, health and technologic literacy, utilization, and coronavirus disease 2019 perceptions. We identified perceived telehealth benefits and challenges. Multivariable logistic regression identified predictors of telehealth dissatisfaction. RESULTS: Of 44,789 surveys sent, 5,832 (13.0%) patients responded and 1,632 (3.6%) had Rome IV IBS. Among 1,314 (22.5%) patients with IBS and prior telehealth use (mean age 52.6 years [17.4]; 84.9% female; and 59.4% non-Hispanic White, 29.0% Hispanic, and 5.6% non-Hispanic Black), 898 (68.3%) were satisfied, 130 (9.9%) were dissatisfied, and 286 (21.8%) felt neutral. In addition, 78.6% would use telehealth again. Independent predictors of telehealth dissatisfaction include social media use of once a week or less (adjusted odds ratio [OR] = 2.1; 1.3–3.5), duration of IBS for <1 year (adjusted OR = 8.2; 1.9–35.8), and willingness to travel 60 plus minutes for face-to-face visits (adjusted OR = 2.6; 1.4–3.7). Patients' main concern with telehealth was a lack of physical examination. DISCUSSION: Most of the patients with IBS are satisfied with telehealth. Shorter duration of IBS diagnosis, comfort with technology, and increased willingness to travel were associated with telehealth dissatisfaction. These predictors may help identify a target population for a focused IBS-telehealth program.

Abstract A028: The Early Detection Initiative trial version 2: A platform trial to test novel approaches to pancreatic cancer screening in patients with new onset hyperglycemia and diabetes

Abstract: The Early Detection Initiative (EDI) is an innovative study designed to build a platform that will evaluate novel approaches to early detection of resectable pancreatic ductal adenocarcinoma (PDAC) in patients over age 50 with new onset hyperglycemia and diabetes (NOD). This randomized controlled trial (RCT) of algorithm-based screening for PDAC uses a modified Zelen’s design with post-randomization consent. Eligible subjects are identified through electronic medical record (EMR) surveillance and randomized 1:1 to the Observation or Intervention Arm. The Enriching New-onset Diabetes for Pancreatic Cancer (ENDPAC) score, an algorithm that further risk stratifies NOD based on age and changes in weight and glycemic parameters, is calculated in the Intervention Arm. Consenting subjects with ENDPAC >0 have Computerized Tomography (CT) scan imaging for PDAC detection; potential harm, including over-diagnosis, is also estimated. At the time of abstract submission >2000 individuals have been randomized and >50 subjects consented for CT imaging. EDI version 2 adds several innovations in response to challenges encountered and experience gained so far. Most significantly, EDI v2 has a platform design with a common control arm and allows for multiple intervention arms. This opens the door for evaluating future new biomarkers for PDAC early detection in a high–risk NOD population, including new blood-based biomarkers for the early detection of cancer that are now commercially available. EDI v2 focuses primarily on co-efficacy endpoints (PDAC stage shift in ENDPAC >0 group, and in all patients, excluding non-consented patients); effectiveness evaluation becomes a secondary objective. This addresses the challenges of low consent rate (~25%) due to lack of awareness of connection between NOD and PDAC and low acceptance for a novel research screening modality. The analyses of primary endpoints were adjusted to accommodate potential over-diagnosis due to screening and imbalance between ENDPAC >0 and ENDPAC ≤0 groups due to low consent rate and likely imperfect enrichment by ENDPAC, when evaluating all patients. Electronic eligibility checks and consent shorten the recruitment time, a critical factor for stage shift due to rapid progression of PDAC. A pre-randomization screening pilot study is planned to enhance the post-randomization consent rate. Intervention to Control randomization ratio is changed to 2:1, allowing more rapid recruitment and, when factoring the consent rate, doubling the sample size of the observation arm. Through these innovations, EDI v2 maintains the full advantage of Zelen’s design that made the trial feasible in sample size and cost. Simulation studies demonstrate that our approach correctly controls the type-1 error and the trial has adequate statistical power. The EDI trial represents an innovative and contemporary approach to developing and refining an early detection strategy for pancreatic cancer. Citation Format: Ying-Qi Zhao, Suresh T. Chari, Anirban Maitra, Lynn M. Matrisian, Eva E. Shrader, Bechien U. Wu, Avinash Kambadakone, Barbara Kenner, Jo Ann S. Rinaudo, Sudhir Srivastava, Ying Huang, Ziding Feng. The Early Detection Initiative trial version 2: A platform trial to test novel approaches to pancreatic cancer screening in patients with new onset hyperglycemia and diabetes [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer; 2022 Sep 13-16; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2022;82(22 Suppl):Abstract nr A028.

Abstract A037: Predicting pancreatic cancer using artificial intelligence analysis of pancreatic subregions using computed tomography images

Abstract: Study background: Early detection of pancreatic ductal adenocarcinoma (PDAC) can elevate the current ~10% five-years survival rate of PDAC up to 50%. Accurate stratification of high-risk individuals for PDAC can improve early detection as follow-up screening may assist diagnosis at an early stage. Studies show that the pancreas adopts changes prior to or during the development of cancer due to the underlying biological variations. This study aimed to examine the precancerous changes that occurred within and across pancreatic subregions to help stratify individuals at high risk of developing PDAC. Dataset: In a multi-institute retrospective study, 108 contrast-enhanced CT abdominal scans were collected, consisting of 36 diagnostic scans with established PDAC and observable tumor, 36 pre-diagnostic scans of the same subjects as in the diagnostic group but were obtained up to 3 years before PDAC diagnosis and were deemed ‘normal’ by radiologists, and 36 healthy scans reported with no PDAC signs. Trained radiologists outlined 3 subregions (head, body, tail) in all scans. Also, the subregions in pre-diagnostic scans were classified into high-risk (with cancer underdevelopment) and low-risk (no cancer development) groups by exploring the tumor signs in their corresponding subregions in the diagnostic scans. Experiments and results: Radiomic analysis was performed on all 324 subregions by extracting and analyzing hundreds of morphological and textural features. In a pairwise feature analysis (i.e. between corresponding subregions), the texture of the high-risk subregions in pre-diagnostic scans was found significantly unique and statistically different than that of the low-risk subregions, supporting the study hypothesis. Such textural features are usually too minute and remain obscured when the pancreas is observed as a single structure. The analysis showed that AI can efficiently identify and quantify such predictors. A Naïve Bayes model was then trained using the same data to automatically predict PDAC using the textural features of the pancreatic subregions. In four-fold cross-validation, the model obtained prediction accuracy by correctly classifying pre-diagnostic and healthy CT scans by 88.2% on average, with sensitivity (true positive rate) and specificity (true negative rate) reaching 82.5% and 94.0%, respectively. The results of this preliminary study are promising and encouraging to further validate the model on a larger dataset. The model showed improved results over those produced in our recent study [1] in which the pancreas as a single structure was examined. The prediction based on the proposed model can potentially assist clinicians to undertake specialized screening, diagnosis, and treatment planning accordingly as the tumor structure, symptoms, and drug response for each pancreatic subregion differs a lot. 1. Qureshi et. al, Predicting pancreatic ductal adenocarcinoma using artificial intelligence analysis of pre-diagnostic computed tomography images. Cancer Biomarkers, 33(2), pp.211-217, 2022. Citation Format: Sehrish Javed, Touseef Ahmad Qureshi, Srinivas Gaddam, Ashley Wachsman, Linda Azab, Vahid Asadpour, Wansu Chen, Bechien Wu, Yibin Xie, Stephen Pandol, Debiao Li. Predicting pancreatic cancer using artificial intelligence analysis of pancreatic subregions using computed tomography images [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer; 2022 Sep 13-16; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2022;82(22 Suppl):Abstract nr A037.