Research on machine learning hybrid framework by coupling grid-based runoff generation model and runoff process vectorization for flood forecasting.

One of the important non-engineering measures for flood forecasting and disaster reduction in watersheds is the application of machine learning flood prediction models, with Long Short-Term Memory (LSTM) being one of the most representative time series prediction models. However, the LSTM model has issues of underestimating peak flows and poor robustness in flood forecasting applications. Therefore, based on a thorough analysis of complex underlying surface attributes, this study proposes a framework for distinguishing runoff models and integrates a Grid-based Runoff Generation Model (GRGM). Simultaneously considering the time series characteristics of runoff processes, including rising, peak, and recession, a runoff process vectorization (RPV) method is proposed. In this study, a hybrid deep learning flood forecasting framework, GRGM-RPV-LSTM, is constructed by coupling the GRGM, RPV, and LSTM neural network models. Taking the Jialu River in the Zhongmu station control basin as an example, the model is validated using 18 instances of measured floods and compared with the LSTM and GRGM-LSTM models. The study shows that the GRGM model has a relative error and average coefficient of determination for simulating runoff of 8.41% and 0.976, respectively, indicating that considering the spatial distribution of runoff patterns leads to more accurate runoff calculations. Under the same lead time conditions, the GRGM-RPV-LSTM hybrid forecasting model has a Nash efficiency coefficient greater than 0.9, demonstrating better simulation performance compared to the GRGM-LSTM and LSTM models. As the lead time increases, the GRGM-RPV-LSTM model provides more accurate peak flow predictions and exhibits better robustness. The research findings can provide scientific basis for coordinated management of flood control and disaster reduction in watersheds.

Deep Learning for Fully Automated Prediction of Overall Survival in Patients Undergoing Resection for Pancreatic Cancer: A Retrospective Multicenter Study.

OBJECTIVE: To develop an imaging-derived biomarker for prediction of overall survival (OS) of pancreatic cancer by analyzing preoperative multiphase contrast-enhanced computed topography (CECT) using deep learning. BACKGROUND: Exploiting prognostic biomarkers for guiding neoadjuvant and adjuvant treatment decisions may potentially improve outcomes in patients with resectable pancreatic cancer. METHODS: This multicenter, retrospective study included 1516 patients with resected pancreatic ductal adenocarcinoma (PDAC) from 5 centers located in China. The discovery cohort (n=763), which included preoperative multiphase CECT scans and OS data from 2 centers, was used to construct a fully automated imaging-derived prognostic biomarker-DeepCT-PDAC-by training scalable deep segmentation and prognostic models (via self-learning) to comprehensively model the tumor-anatomy spatial relations and their appearance dynamics in multiphase CECT for OS prediction. The marker was independently tested using internal (n=574) and external validation cohorts (n=179, 3 centers) to evaluate its performance, robustness, and clinical usefulness. RESULTS: Preoperatively, DeepCT-PDAC was the strongest predictor of OS in both internal and external validation cohorts [hazard ratio (HR) for high versus low risk 2.03, 95% confidence interval (CI): 1.50-2.75; HR: 2.47, CI: 1.35-4.53] in a multivariable analysis. Postoperatively, DeepCT-PDAC remained significant in both cohorts (HR: 2.49, CI: 1.89-3.28; HR: 2.15, CI: 1.14-4.05) after adjustment for potential confounders. For margin-negative patients, adjuvant chemoradiotherapy was associated with improved OS in the subgroup with DeepCT-PDAC low risk (HR: 0.35, CI: 0.19-0.64), but did not affect OS in the subgroup with high risk. CONCLUSIONS: Deep learning-based CT imaging-derived biomarker enabled the objective and unbiased OS prediction for patients with resectable PDAC. This marker is applicable across hospitals, imaging protocols, and treatments, and has the potential to tailor neoadjuvant and adjuvant treatments at the individual level.

Feasibility and Reproducibility of Multifrequency Magnetic Resonance Elastography in Healthy and Diseased Pancreases.

BACKGROUND: The feasibility and reproducibility of multifrequency MR elastography (MRE) for diagnosing pancreatic ductal adenocarcinoma (PDAC) have not been reported. PURPOSE: To determine the feasibility and reproducibility of multifrequency MRE for assessing pancreatic stiffness in healthy and diseased pancreases. STUDY TYPE: Prospective. SUBJECTS: A total of 40 healthy volunteers and 10 patients with PDAC were prospectively recruited between March 2018 and October 2021. FIELD STRENGTH/SEQUENCE: A 3.0-T pancreatic MRE at frequencies in the order of 30, 40, 60, 80, and 100 Hz. ASSESSMENT: Body mass index (BMI) and wave distance of the healthy pancreas and PDAC were measured. Image quality was assessed using the image quality score (IQS: 1-4, ≥3 were considered diagnostic quality). Three readers independently performed the pancreatic stiffness and IQS assessments to evaluate reproducibility. STATISTICAL TESTS: Logistic regression analyses were performed to determine variables that influenced IQS. Statistical significance was set at P <0.05. Levels of inter- and intrarater agreement were assessed using intraclass correlation coefficients (ICC) and Cohen's kappa coefficient (κ). Good reproducibility was set at ICC and κ ≥ 0.8. RESULTS: In logistic regression analysis, a diagnostic IQS in healthy volunteers was independently associated with a lower BMI (odds ratio [OR] = 0.89 kg/m-2 ), shorter wave distance (OR = 0.70 cm-1 ), and lower frequency (30 and 40 Hz: OR = 170.01 and 96.02). In PDAC, frequency was the only independent factor for diagnostic IQS (30-60 Hz: OR = 46.18, 46.18, and 17.20, respectively) with 100 Hz as a reference. In healthy volunteers, good reproducibility was observed at 30 and 40 Hz. In PDAC, good reproducibility was observed at 30-60 Hz. DATA CONCLUSION: MRE at 30 and 40 Hz provides diagnostic wave images and reliable measurements of pancreatic stiffness in healthy volunteers. MRE at 30-60 Hz is acceptable for PDACs (IQS ≥ 3, ICC and κ ≥ 0.80). EVIDENCE LEVEL: 1 TECHNICAL EFFICACY: Stage 2.

Clinical utility of 3D magnetic resonance elastography in patients with biliary obstruction.

OBJECTIVES: Three-dimensional magnetic resonance elastography (3D-MRE) allows for multiparametric modeling of both elastic and viscous tissue characteristics. Our aim was to compare 3D-MRE with conventional liver shear stiffness assessment in gauging obstructive jaundice (OJ), predicting the adequacy of biliary decompression after drainage, and discriminating OJ from liver fibrosis. METHODS: Patients with no histories of liver disease (n = 201) were studied in retrospect, grouped by bilirubin levels as no jaundice (NJ ≤ 2 mg/dL; n = 75), mild OJ (>2 mg/dL and ≤ 4 mg/dL; n = 56), and severe OJ (> 4 mg/dL; n = 70). For comparison, another 75 patients with chronic hepatitis B and C infections and histologically proven liver fibrosis were similarly analyzed. Each patient underwent spin-echo echo-planar-imaging MRE at 60 Hz with 3D wave postprocessing. Logistic regression and ordinary regression models were used to compare the 3D-MRE model with liver shear stiffness. RESULTS: Liver shear stiffness, loss modulus, and damping ratio were incorporated into a 3D-MRE model, which significantly outperformed shear stiffness in predicting OJ severity (accuracy: 0.801 vs 0.672; p < 0.001). Both the 3D-MRE model and liver shear stiffness performed equally well in predicting the outcome of biliary drainage procedure (C-statistics: 0.852 vs 0.847; p = 0.48). The 3D-MRE model also demonstrated significantly better C-statistics than that of liver shear stiffness in discriminating mild OJ from F1-F2 liver fibrosis (0.765 vs 0.641; p = 0.005) and severe OJ from F3-F4 liver fibrosis (0.750 vs 0.635; p = 0.031). CONCLUSIONS: 3D-MRE is an innovative imaging method for gauging OJ severity, predicting the outcome of biliary drainage procedure, and discriminating OJ from liver fibrosis. KEY POINTS: • 3D-MR elastography achieved promising results for predicting the severity of obstructive jaundice. • Advanced parameters of 3D-MR elastography demonstrated significantly better performance than that of shear stiffness of 2D-MR elastography in discriminating obstructive jaundice from liver fibrosis caused by chronic hepatitis B/C. • Both 3D-MR elastography and 2D-MR elastography were equivalent in predicting the outcome of biliary drainage procedure.

DeepPrognosis: Preoperative prediction of pancreatic cancer survival and surgical margin via comprehensive understanding of dynamic contrast-enhanced CT imaging and tumor-vascular contact parsing.

Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal cancers and carries a dismal prognosis of ∼10% in five year survival rate. Surgery remains the best option of a potential cure for patients who are evaluated to be eligible for initial resection of PDAC. However, outcomes vary significantly even among the resected patients who were the same cancer stage and received similar treatments. Accurate quantitative preoperative prediction of primary resectable PDACs for personalized cancer treatment is thus highly desired. Nevertheless, there are a very few automated methods yet to fully exploit the contrast-enhanced computed tomography (CE-CT) imaging for PDAC prognosis assessment. CE-CT plays a critical role in PDAC staging and resectability evaluation. In this work, we propose a novel deep neural network model for the survival prediction of primary resectable PDAC patients, named as 3D Contrast-Enhanced Convolutional Long Short-Term Memory network (CE-ConvLSTM), which can derive the tumor attenuation signatures or patterns from patient CE-CT imaging studies. Tumor-vascular relationships, which might indicate the resection margin status, have also been proven to hold strong relationships with the overall survival of PDAC patients. To capture such relationships, we propose a self-learning approach for automated pancreas and peripancreatic anatomy segmentation without requiring any annotations on our PDAC datasets. We then employ a multi-task convolutional neural network (CNN) to accomplish both tasks of survival outcome and margin prediction where the network benefits from learning the resection margin related image features to improve the survival prediction. Our presented framework can improve overall survival prediction performances compared with existing state-of-the-art survival analysis approaches. The new staging biomarker integrating both the proposed risk signature and margin prediction has evidently added values to be combined with the current clinical staging system.