Semi-supervised learning framework for oil and gas pipeline failure detection.

Quantifying failure events of oil and gas pipelines in real- or near-real-time facilitates a faster and more appropriate response plan. Developing a data-driven pipeline failure assessment model, however, faces a major challenge; failure history, in the form of incident reports, suffers from limited and missing information, making it difficult to incorporate a persistent input configuration to a supervised machine learning model. The literature falls short on the development of appropriate solutions to utilize incomplete databases and incident reports in the pipeline failure problem. This work proposes a semi-supervised machine learning framework which mines existing oil and gas pipeline failure databases. The proposed cluster-impute-classify (CIC) approach maps a relevant subset of the failure databases through which missing information in the incident report is reconstructed. A classifier is then trained on the fly to learn the functional relationship between the descriptors from a diverse feature set. The proposed approach, presented within an ensemble learning architecture, is easily scalable to various pipeline failure databases. The results show up to 91% detection accuracy and stable generalization ability against increased rate of missing information.

Predicting seismic-induced liquefaction through ensemble learning frameworks.

The regional nature of liquefaction records and limited information available for a certain set of explanatories motivate the development of complex prediction techniques. Indirect methods are commonly applied to incidentally derive a hyperplane to this binary classification problem. Machine learning approaches offer evolutionary prediction models which can be used as direct prediction methods to liquefaction occurrence. Ensemble learning is a recent advancement in this field. According to a predefined ensemble architecture, a number of learners are trained and their inferences are integrated to produce stable and improved generalization ability. However, there is a need to consider several aspects of the ensemble learning frameworks when exploiting them for a particular application; a comprehensive evaluation of an ensemble learner's generalization ability is required but usually overlooked. Also, the literature falls short on work utilizing ensemble learning in liquefaction prediction. To this extent, this work examines useful ensemble learning approaches for seismic-induced liquefaction prediction. A comprehensive analysis of fifteen ensemble models is performed. The results show improved prediction performance and diminishing uncertainty of ensembles, compared with single machine learning models.

Benign focal liver lesions: discrimination from malignant mimickers.

Focal lesions of the liver often have various imaging characteristics which may be interpreted as either benign or malignant. Understanding the underlying pathophysiology of these liver lesions may lead to characteristic imaging manifestations, which direct the radiologist to the diagnosis. Benign lesions include congenital hepatic cyst, autosomal dominant polycystic disease, hemangioma, focal nodular hyperplasia (FNH), hepatic adenoma, inflammatory pseudotumor, peliosis hepatis, focal fatty infiltration, hamartoma, and infectious processes such as hepatic abscess, echinococcal cyst, and candidiasis. Characteristic imaging features, clinical symptoms, and treatment/prognosis will be discussed. Emphasis will be placed on key reliable features of each disease to develop a method of discriminating these lesions from other benign and malignant disorders.

Malignant cystic and necrotic liver lesions: a pattern approach to discrimination.

Malignant primary and metastatic lesions of the liver have a myriad of imaging appearances. Discriminating between the various lesions requires an understanding of the underlying pathophysiology and imaging characteristics that lead to their malignant appearances. A pattern approach of recognition by imaging with understanding of why particular lesions behave the way they do on sonography, CT, MRI, and nuclear scintigraphy can be developed. A broad selection of malignant cystic liver lesions will be displayed in a multimodality pictorial style including hepatocellular carcinoma, cholangiocarcinoma, cystic and necrotic hepatic metastases, biliary cystadenoma/cystadenocarcinoma, epitheloid hemangioendothelioma, hepatoblastoma, and cystic sarcomas. Characteristic imaging features, clinical symptoms, and prognosis will be discussed. Emphasis will be placed on consistent features of these malignant lesions that help to discriminate them from other hepatic disorders.

Current trends in imaging evaluation of acute cholecystitis.

This study was designed to retrospectively determine recent clinical trends of initial radiological evaluation in patients pathologically proven to have acute cholecystitis (AC) and to assess the methodology that led to its diagnosis. Over a 28-month period, the medical records and imaging studies of 117 consecutive patients who had pathologically confirmed AC were retrospectively analyzed. The sensitivities of ultrasound (US) and hepatobiliary 99mTc-iminodiacetic acid (HIDA) were computed. The false-negative scans were retrospectively reviewed by a blinded radiologist to determine the limitations and advantages of each modality. The 117 patients were grouped into six categories based on the type of imaging examination they underwent prior to cholecystectomy: initial US evaluation only (n=80, 68.4%), initial US followed by HIDA (n=17, 14.5%), initial HIDA only (n=2, 1.7%), initial HIDA followed by US (n=3, 2.6%), initial CT (n=5, 4.3%), and no imaging evaluation (n=10, 8.6%). HIDA scan had a calculated sensitivity of 90.9% (20 true-positive, 2 false-negative) while US had a sensitivity of 62% (62 true-positive, 38 false-negative). Current practice in the initial radiological evaluation of acute cholecystitis remains outdated. The vast majority of patients in our study group were initially worked up using US, although HIDA scan has been shown to have greater sensitivity for the diagnosis of acute cholecystitis.

Value of bone window settings on CT for revealing appendicoliths in patients with appendicitis.

OBJECTIVE: Our retrospective study was designed to determine whether the use of bone window settings increases sensitivity of CT for diagnosing appendicitis and for detecting an appendicolith in patients with pathologically confirmed appendicitis. CONCLUSION: The use of bone window settings is helpful for detecting appendicoliths when evaluating patients for acute appendicitis, particularly patients in whom evidence of appendicitis is equivocal. In this era of PACS (picture archiving and communication systems), bone window settings should be used routinely.