Universiti Teknologi Malaysia Institutional Repository

New development of classifier for prediction of cancellous bone failure

Haron, Habibollah (2008) New development of classifier for prediction of cancellous bone failure. In: Proceedings of the 9th Asia Pacific Industrial Engineering & Management Systems Conference (APIEMS 2008), 2008, Nusa Dua Beach Hotel & Spa Bali,.

Full text not available from this repository.


The study and monitoring of cancellous bone is particularly important for medical image analyses which are generally related to orthopedics field. In order to understand more about the cancellous bone, a three-dimensional (3D) imaging method is mainly being used for visualizing the morphological data of the bone. This study furthermore presents a new development of classifier to predict the failure of trabecular bone structure from the analyses of the morphological parameters. More particularly, the present study relates to the methods and systems for 3D object classification using Bayes decision rules. Classification method in this study is considered as an important approach or tool for recognizing the cancellous bone failure such Osteoporosis’s conditions in the difference bone parts of the body. Moreover, structural information of the cancellous bone failure is also attained from the classification which is varies depending to the bone parts. This algorithm are tested and validated by using the morphological datasets of the trabecular bone structure taken from tibia bone specimen. The results of this algorithm are compared to the original morphological datasets for observing the accuracy of the results with the original datasets. These studies in advance have provided some add up value as part of the function in the commercial high visualization and analysis software system for medical diagnosis.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:kiv record
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:Computer Science and Information System (Formerly known)
ID Code:19712
Deposited By: Mrs Liza Porijo
Deposited On:07 Feb 2017 07:25
Last Modified:11 Oct 2017 01:59

Repository Staff Only: item control page