Post operative prognostic prediction of esophageal cancer cases using bayesian networks and support vector machines

Download

index.pdf

Date

2014

Author

Bagherzadi, Negin

Metadata

Show full item record

Item Usage Stats

150
views

46
downloads

The objective of this thesis is to develop and analyze the performances of a number of classifiers in prognosis classification based on a medical history data set. Generally, data mining uses algorithms originating in different disciplines such as artificial intelligence, statistics, optimization, database theory etc., to clarify available data. In this study, Support Vector Machine and Bayesian Network methods have been used. The data analyzed are clinical pathology records of patients with esophageal cancer who received an esophagectomy operation between 2003 and 2011. A large number of prognostic factors have been considered to classify the patients on prognosis. These factors found to be predictive were age, sex, dysphagia, odynophagia, lost weight, vomit, nausea, pathological N, pathological T, FEV1, tumor grade and tumor length. Classification trials for Support Vector Machine have shown %72.38 training accuracy with a generalization accuracy of %70.58, which was established by cross-validation. Support Vector Machine was used as the first method for data and SVM was found to achieve good accuracy on a data set of 119 patients when used in conjunction with PCA; SVM can be helpful for black-box analysis when data are irregularly distributed and produce accurate classifiers. Bayesian Network is the second method that is used in this study to solve missing data problem in the previous method. Bayesian Network can classify several different types of variables simultaneously. The quality of the results of the network that has been created for this study depends on the quality of the model. Performance of the model with Bayesian Network that is used in this study is 73.10% for 119 patients.

Subject Keywords

Esophagus, Esophagus, Cancer, Cancer, Prognosis, Bayesian networks., Support vector machines.

URI

http://etd.lib.metu.edu.tr/upload/12617197/index.pdf
https://hdl.handle.net/11511/23534

Collections

Graduate School of Informatics, Thesis

Suggestions

OpenMETU
Core

Mapping and clustering analysis on neuroscience literature in Turkey: a bibliometric analysis from 2000 to 2017 Kocak, Murat; Garcia-Zorita, Carlos; Marugan-Lazaro, Sergio; Çakır, Murat Perit; Sanz-Casado, Elias (2019-12-01) This study is aim to determine significant changes and trends in neuroscience literature concerning Medical Subject Headings (MeSH) from 2000 to 2017. This article encourage researchers to help identify perhaps the most influential studies and combine detailed evidence into study-considered higher education. To our knowledge and understanding, almost no study found alterations in neuroscientific literature during the last decade in Turkey. In this study, the main aim is to present a science map of "Neurosci...
Examination of the coping processes of caregivers of children with cancer Demirtepe Saygılı, Dilek; Bozo Özen, Özlem; Department of Psychology (2013) The general aim of the present dissertation was to examine the coping processes of the caregivers of children with cancer. It was composed of three studies, whose aims were (1) to reveal the predictors of caregiver distress by using the Transactional Theory of Coping among the caregivers of children with cancer; (2) to identify the factors associated with caregiving process from the caregiver’s perspective qualitatively; (3) to develop and implement an intervention program. Study 1 was conducted with 105 pa...
CAD for detection of microcalcification and classification in mammograms AKBAY, Cansu; Gençer, Nevzat Güneri; GENÇER, Gülay (2014-10-17) In this study, computer aided diagnosis (CAD) is developed to detect microcalficication cluster which is one of the important radiological findings of breast cancer diagnosis and classificiation. For this purpose, image processing and pattern recognition algorithms are applied on mamographic images. To make microcalcifications more visible wavelet transform and nonsubsampled contourlet transform (NSCT) methods are used for image enhancement. Their performances are compared. 52 features are extracted from th...
A systematic review of the reliability of objective structured clinical examination scores Brannick, Michael T.; Erol-Korkmaz, H. Tugba; Prewett, Matthew (2011-12-01) CONTEXT The objective structured clinical examination (OSCE) is comprised of a series of simulations used to assess the skill of medical practitioners in the diagnosis and treatment of patients. It is often used in high-stakes examinations and therefore it is important to assess its reliability and validity.
Analysis and testing of a contraction-and-expansion micromixer for micromilled microfluidics Yıldırım, Ender (2017-10-01) In this paper, numerical analysis and experimental investigation of a micromixer, which was specifically designed for microfluidic devices fabricated by micromilling, is presented. The mixer is composed of series of contractions and expansions in zigzag arrangement along a mixing channel. Mixers, fabricated by micromilling on polymethylmethacrylate (PMMA), were tested with %0.1 Ponceau 4R red food dye solution and distilled water. According to experiment results, over 70% mixing efficiency could be obtained...

Citation Formats

N. Bagherzadi, “Post operative prognostic prediction of esophageal cancer cases using bayesian networks and support vector machines,” M.S. - Master of Science, Middle East Technical University, 2014.