Title :
The research on CPA diagnosis application basing on some Bayesian classifiers
Author :
Bei Hui ; Lin Ji
Author_Institution :
Sch. of Comput. Sci. & Eng., Univ. of Electron. Sci. & Technol. of China, Chengdu, China
Abstract :
Cerebellopontine angle(CPA) masses comprise about 8-10% of all intracranial neoplasm. Preoperative diagnosis of a CPA region mass is mainly based on imaging. MRI is the best method for diagnosing the CPAM. But MRI can´t diagnose the different masses accurately. Many computer aided diagnose (CAD) technologies were developed to help radiologists to improve the diagnostic performance. The medical cases in the experiment were from West China Hospital. The experiment validates efficiency and effective of the some kinds Bayesian classifiers by 0-1 Error and RMSE. The result of experiment shows that Bayesian Classification model can classify CPAM effectively.
Keywords :
belief networks; diseases; mean square error methods; medical diagnostic computing; pattern classification; 0-1 error; Bayesian classification model; CAD technologies; CPA diagnosis application; CPA region mass; CPAM classification; MRI; RMSE; West China Hospital; cerebellopontine angle; computer aided diagnosis; intracranial neoplasm; magnetic resonance imaging; root mean square error; Bayes methods; Error analysis; Hospitals; Magnetic resonance imaging; Niobium; Shape; Tumors; Cerebellopontine angle mass; MRI; semi-naive Bayes;
Conference_Titel :
Wavelet Active Media Technology and Information Processing (ICCWAMTIP), 2013 10th International Computer Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4799-2445-5
DOI :
10.1109/ICCWAMTIP.2013.6716602
