Flood hazard risk assessment model based on random forest

. Flood loss models are one important source of uncertainty in flood risk
assessments. Many countries experience sparseness or absence of comprehensive
high-quality flood loss data, which is often rooted in a lack of protocols and
reference procedures for compiling loss datasets after flood events. Such
data are an important reference for developing and validating flood loss
models. We consider the Secchia River flood event of January 2014, when a
sudden levee breach caused the inundation of nearly 52 km 2 in northern
Italy. After this event local authorities collected a comprehensive flood
loss dataset of affected private households including building footprints and
structures and damages to buildings and contents. The dataset was enriched with
further information compiled by us, including economic building values,
maximum water depths, velocities and flood durations for each building. By
analyzing this dataset we tackle the problem of flood damage estimation in
Emilia-Romagna (Italy) by identifying empirical uni- and multivariable loss
models for residential buildings and contents. The accuracy of the proposed
models is compared with that of several flood damage models reported in the
literature, providing additional insights into the transferability of the
models among different contexts. Our results show that (1) even simple
univariable damage models based on local data are significantly more
accurate than literature models derived for different contexts;
(2) multivariable models that consider several explanatory variables
outperform univariable models, which use only water depth. However,
multivariable models can only be effectively developed and applied if
sufficient and detailed information is available.

/pdf/development-and-assessment-of-uni-and-multivariable-flood-3gw1fpihpo.pdf

Development and assessment of uni- and multivariable flood loss models for Emilia-Romagna (Italy)

Hepatic steatosis can occur because of nonalcoholic fatty liver disease (NAFLD), alcoholism, chemotherapy, and metabolic, toxic, and infectious causes. Pediatric hepatic steatosis is also becoming more frequent and can have distinctive features. The most common pattern is diffuse form; however, it can present in heterogenous, focal, multinodular, perilesional, perivascular, subcapsular, and lobar forms. Focal steatosis and fat sparing can occur because of the presence of veins of Sappey, pancreaticoduodenal vein, and aberrant right and left gastric veins, which drain into the liver as third inflow. Hypersteatosis and multinodular forms can mimic metastasis in patients with cancer. Perilesional fat can be seen in insulinoma. Recent introduction of proton-density fat fraction enabled easy and reproducible quantification of hepatic fat. Follow up of patients with NAFLD can be performed for the assessment of treatment response using proton-density fat fraction as biomarker. Multiecho gradient-echo techniques also simultaneously calculate T2* maps, which is important to rule out coexisting hepatic iron overload. NAFLD can progress to steatohepatitis (nonalcoholic steatohepatitis), which can result in cirrhosis. Magnetic resonance (MR) elastography and functional evaluation with Gd-EOB-DTPA are becoming important for monitoring this process. Hepatocellular carcinoma can develop in patients with NAFLD, which is usually a large tumor with necrotic center. In the future, fatty acid maps obtained by MR imaging may allow more detailed analysis of steatosis. MR imaging is superior to ultrasonography and computed tomography for comprehensive evaluation of steatosis.

Hepatic Steatosis: Etiology, Patterns, and Quantification

Quantification of liver fat

Focal disturbance ratio calculated by using acoustic structure quantification is comparable to hepatic fat fraction calculated by using MR spectroscopy to quantify hepatic steatosis and detect hepatic steatosis greater than 10% in living donor liver patients.

Hepatic Steatosis: Assessment with Acoustic Structure Quantification of US Imaging

In this paper we discuss the problem of computer aided evaluation of the severity of steatosis disease using ultrasound images, the aim of the study being to compare the automatic evaluation of liver steatosis using random forests (RF) and support vector machine (SVM) classifiers. Material and method: One hundred and twenty consecutive patients with steatosis or normal liver, assessed by ultrasound by the same expert, were enrolled. We graded steatosis in four stages and trained two classifiers to rate the severity of disease, based on a large set of labeled images and a large set of features, including several features obtained by robust estimation techniques. We compared RF and SVM classifiers. The classifiers were trained using cross-validation. There was 80% of data randomly selected for training and 20% for testing the classifier. This procedure was performed 20 times. The main measure of performance was the accuracy. Results: From all cases, 10 were rated as normal liver, 70 as hav ing mild,  33 moderate, and 7 severe steatosis. Our best experts’ ratings were used as ground truth data. RF outperformed the SVM classifier and confirmed the ability of this classifier to perform well without feature selection. In contrast, the performance of the SVM classifier was poor without feature selection and improved significantly after feature selection. Conclusion:  The ability and accuracy of RF to classify well the steatosis severity, without feature selection, were superior as compared to SVM.

/pdf/computer-aided-diagnosis-method-for-steatosis-rating-in-1kem8qk72m.pdf

Computer aided diagnosis method for steatosis rating in ultrasound images using random forests

This paper presents a new method for computer based diagnosis in liver steatosis diagnosis. The effectiveness of several image features extracted by robust estimation in conjunction with a random forests classifier for steatosis stage assessment is proved. We found that a successive dichotomy approach using random forests leads to better diagnosis than simultaneous classification with random forests.

Automatic evaluation of steatosis by ultrasound image analysis

Parameters extracted from time activity curves in contrast enhanced ultrasound images play an important role in independent or computer aided diagnosis of hepatic focal lesions. Due to noise and errors induced by movement of ultrasound probe and breathing of patients, reproducible extraction time activity curve parameters is challenging. In this paper we propose a new solution that combines filtering and tracking, in the theoretical framework of robust estimation and the mean shift. To cope with the gradual apparence changes of the tracked lesions, we propose a new dynamic scale selection method.

Dan Mihai Mihailescu

Papers

Computer aided diagnosis method for steatosis rating in ultrasound images using random forests

Automatic evaluation of steatosis by ultrasound image analysis

Simultaneous filtering and tracking of focal liver lesion for time intensity curve analysis in contrast enhanced ultrasound imagery

Comparison of patient internal dosimetry using the reference computational and xcat phantoms with monte carlo method