Title

Abdominal palpation characterization using computer vision

Date of Publication

2016

Document Type

Thesis

Degree Name

Bachelor of Science in Computer Engineering

College

Gokongwei College of Engineering

Department/Unit

Electronics And Communications Engg

Thesis Adviser

Cabatuan, Melvin K.

Defense Panel Chair

Abad, Alexander C.

Defense Panel Member

Torregoza, Mark Lorenze
Navea, Roy Francis R.

Abstract/Summary

This study intends to develop a tracking algorithm that will segment the hand from the skin using real time multicolor space segmentation. The tracking system uses keypoints and descriptors as its feature extraction. The system will work on different skin colors, namely, fair, brown, and dark. The generalization of these levels used calibrated weights as its basis on how much force the user needs to exert to his hand while palpating. Its measurement is in terms of grams. The result of experimenting with how much force is the light palpation ranges from 0 to 1.25 kilograms. Medium level palpation ranges from 1.25 to 1.75 kilograms while deep level palpation starts from 1.75 kilograms and above.

The system will also differentiate the levels of palpation with the force exerted by the hand. The levels of palpation that were used in this study are light, medium, and deep. Support Vector Machine (SVM) was used in separating the levels of palpation from each other. SVM is an algorithm that sorts out the features that belong to a group and acts as an efficient classification method. In each of the datasets, the person who is palpating uses different combinations of palpation level throughout the video. This tested the system's effectiveness on how it can handle the levels of palpation while also using the system in different skin colors.

Abstract Format

html

Format

Print

Accession Number

TU17414

Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

xi, 128 leaves ; illustrations (some color) ; 28 cm. ; 1 computer disc ; 4 3/4 in.

Keywords

Palpation; Computer vision; Computer vision in medicine

This document is currently not available here.

Share

COinS