Masters Thesis Defense "Computer Graphics and Visualization based Analysis and Record System for Hand Surgery and Therapy Practice” by Venkatamanikanta Subrahmanyakartheek Gokavarapu

Committee:  Drs. Yong Pei, Advisor, Mateen Rizki, Co-Advisor, and TK Prasad

ABSTRACT:

In this thesis, we have designed and developed a computer graphics and visualization based analysis and record system for hand surgery and therapy practice. In particular, we have designed and developed three novel technologies: (i) model-based data compression for hand motion records (ii) model-based surface area estimation of a human hand and (iii) an emulated study of hand wound area estimation.

First, we have presented a new data compression technique to better address the needs of electronic health record systems, such as file storage and privacy. In our proposed approach, we will extract the patient’s hand motion information and store the motion data in a binary format and then we use a loss less data compression to further reduce the file size. To illustrate this idea, we have built a prototype, which demonstrate our entire work flow. Our experiment results have shown the effective compression performance as well as added benefit of 3-D review, enhanced privacy and review capability at different playback speeds and view angles.

Next, we developed a new approach, to estimate the surface area of a patient’s hand accurately and quickly with a low-cost imaging sensor and a graphics based hand model. Through the image capturing device, we capture infrared images of iii the patient’s hand. Once we get the input images, we run them through the image analysis engine to extract the required information in order to obtain a customized graphics hand object. Once customized we use the graphics hand object to calculate the surface area of the patient’s hand. To illustrate this idea, we have built a prototype system, which demonstrate the entire work flow. Our experiment results have shown considerable reproducibility and consistency.

Finally, we came up with a novel system, capable of identifying simulated wounds on a human hand and estimating their area. Through the imaging device, wound analysis engines, we are able to identify simulated wounds on a patient’s hand and measure their area using optical equations. To illustrate this idea, we have built a prototype system, which demonstrate our entire work flow. Our experiments show considerable reproducibility, accuracy and consistency. In summary, we have developed prototypes for each of the approaches to demonstrate its capabilities. Experiments and analysis are carried out to study their performance and complexity. We believe these new approaches can significantly improve the current practice in hand surgery analysis and therapy practice.