Design and development of the intelligent operating architecture for audio-visual breast self-examination multimedia training system

• Main Author: Billones, Robert Kerwin C.
• Format: text
• Language: English
• Published: Animo Repository 2015
• Online Access: https://animorepository.dlsu.edu.ph/etd_masteral/4952
• Institution: De La Salle University

Breast cancer poses a major health care challenge, especially in developing countries. As the incidence of breast cancer continues to rise steadily in the developing world, the lack of awareness of this disease and the absence of breast cancer screening programs make it almost certain that the majority of breast cancers are diagnosed at an advanced stage. Early detection of breast cancer together with proper treatment increases the rate of survival. Studies suggest that the majority of breast cancers could be first detected during breast self-examination (BSE). The use of computer-guided BSE multimedia training system (BSE-MTS) can increase the probabilities of detecting breast abnormalities compared with non-guided regular BSE. This study is in conjunction with the research initiative of the Commission on Higher Education Philippine Higher Education Research Network (CHED-PHERNet) on Addressing the High Breast Cancer Incidence in Bacolod City. De La Salle University (DLSU) Manila is in collaboration with University of St. La Salle (USLS) Bacolod in undertaking this research. The research has several activities which include the development of a tool for breast cancer monitoring and education. The development of an interactive breast self-examination multimedia training system that can be easily used by the local female population in Western Visayas region of the Philippines, particularly in Bacolod city, can help awareness and prevention of this dreaded disease by early detection of any breast malignancies while it is still in its early stage. The interactive BSE multimedia training system aims to improve the efficiency and accuracy in performing BSE and to promote the ease of use of the system. The major goal of this research is to design and develop the intelligent operating architecture (IOA) for audio-visual BSE multimedia training system that will include computer vision (CV), speech recognition (SR), speech synthesis (SS), and audio-visual (AV) feedback systems. The artificial intelligence (AI) named BEA which is an acronym for Breast Examination Assistant, a virtual health care assistant, will run using the IOA to assist the user in performing BSE. This aims to improve the user experience and confidence by deploying BEA in the real time evaluation of BSE. The research has four key areas of study which are computer vision, speech recognition, speech synthesis, and audio-visual feedback systems. The computer vision system will be used for breast region identification, nipple detection and tracking, hand motion detection and tracking, and palpation level detection. The fundamental algorithms for computer vision has already been developed by the Intelligent Systems Laboratory (ISL) of De La Salle University Manila. Most of these algorithms will be used in this system. The speech recognition system will be used to identify speech commands issued by the user for the speech controlled human-computer interface (HCI) of the system. This is an important feature because in performing BSE it is beneficial if the user can indicate the possible tumor locations just by dictating it to the system. This will allow less body movement during BSE performance and thus contributing less error in the computer vision algorithms. Speech recognition for English and Hiligaynon languages will be developed for this system allowing flexibility in user preferences. Speech synthesis system will be used for audio feedback response of the system to the user. It will be developed to provide a synthesized female voice for BEA. The output of speech synthesis will be fed to the AV feedback system. Speech database will be collected from different female users for speech recognition and synthesis training. The audio-visual feedback system will be used for user guidance in performing BSE for better user experience. It will also have the capability of providing information to user queries. The intelligent operating architecture will serve as the integration system that will supervise the over-all operation of the BSE multimedia training system. BSE performance is a simple yet sensitive healthcare practice. This research also aims to increase the technology acceptance of the users of the system by creating BEA to appear like a personal female healthcare assistant. This technology can bridge the gap in the comfort levels of the female users to the overall benefits of the system.