Endoscope report integrated with automatic image labeling system | |
| Author | Peeranat Sangkatumvong |
| Call Number | AIT Thesis no.CS-16-01 |
| Subject(s) | Computer vision in medicine Endoscopes Machine learning |
| Note | A research study submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Computer Science, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. CS-16-01 |
| Abstract | In the field of medical information systems, most modern medical equipment is integrated with software. The techniques of computer vision have been introduced in the medical field to improve quality of service and convenient admission of patients. In this study, I explore the possibility of automatically providing labels of parts of the human digestive system, as another step in technology to enhance diagnosis and treatments of patients. I prefer an empirical study of several possible techniques to allow a computer to categorize specific parts of the upper digestive system, from the esophagus to the duodenum. My method is to detect feature points in the image then compute SIFT or SURF descriptions integrated with the bag of visual words technique. I also explore HOG and LBP method. For classification, I select SVMs and decision tree to categorize image descriptors. The results of my experiments are presented in the form of confusion matrices and the overall accuracy each approach. I find that the best method for digestive system endoscope image classification is SVMs with SURF integrated with bag of visual words, with an overall accuracy of 0.57478 over a classifier 1,677 images, I conclude that although the accuracy is not yet high enough to be deployed as a diagnostic to use in commercial software, the research are premising enough for future research on other research to analysis anatomical of human |
| Year | 2015 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis : no. CS-16-01 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Dailey, Matthew N.; |
| Examination Committee(s) | Manukid Parnichkun;Mongkol Ekpanyapong; |
| Scholarship Donor(s) | Asian Institute of Technology Fellowship; |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2016 |