Phone: (709) 737-4368
Fax: (709) 737-2009
Current Research Interest:
Artificial intelligence, computer vision, knowledge-based systems, pattern recognition and robotics.
An Intelligent Vision System for Mobile Robots (J. Shieh)
Intelligent vision systems are required for mobile robots working in the environments that have not been specially structured or engineered. This kind of system is not only required to locate objects, but also is required to recognize the objects in order to manipulate them properly. Thus, the system should be able to analyze multiple views from planned viewpoints, to construct descriptions of objects, and to convert the descriptions into conceptual models in order to classify and recognize them quickly. Current research focuses on developing object representation schemes, developing memory organization and classification methods for objects, and developing schemes for arranging the viewpoints and views for mobile robots.
A Development Environment for Knowledge Acquisition, Management and Maintenance (J. Shieh)
Knowledge-based systems have many important applications in varied domains. In knowledge engineering practice, the task of acquisition of large amount of knowledge is difficult. Furthermore, knowledge acquisition can not be limited to a system's development phase. As domain knowledge grows and changes, there is a corresponding need to update and reconfigure system's knowledge bases. Thus, development of methodologies for knowledge acquisition, management and maintenance is important. Current research mainly focuses on user-oriented interfaces, specification of knowledge-based systems, conversion between knowledge representations, organization and management of knowledge.
An Intelligent Map Understanding System (J. Shieh)
A map understanding system not only is useful for the navigation of autonomous vehicles and robots, but also has a great impact on geographic information systems(GIS) and automatic cartography systems. Automatically converting scanned images of paper maps into vector-based maps will greatly enhance the entrance, storage, analysis, processing and utilization of spatial data. Current research mainly focuses on representations of maps, methodologies for extraction of complex features from images of maps, and the mechanisms for search features on maps.
An Expert System for Selection of Acupuncture Prescriptions based on
Traditional Chinese Medicine (TCM)(J.Shieh)
TCM is growing in popularity and offers an important alternative or complement to conventional medical care in Canada and around the world. TCM encompasses the disciplines of herbology, acupuncture, moxibustion, Chinese nutrition, TCM orthopedics and traumatology, tuina, and qigong. For thousands of years, the Chinese people have accepted acupuncture and moxibustion for their curative effect, wide range of indications, low cost, and safety. The effectiveness of these treatments has been proven by countless cases. In 1979 the World Health Organization Interregional Seminar produced a list of 43 diseases and disorders that lend themselves to acupuncture treatment. Clinically, acupuncture and moxibustion have been used to treat more than 300 disorders. For achieving desired effects, the selection of an acupuncture prescription is most important. A prescription consists of acupoints along with the corresponding needle manipulation methods. A good acupuncture prescription should be decided based on the theories of TCM and the accumulated experience of acupuncture practice, also on the correct judgment of a patientís condition. One of my research projects is to develop a computer expert system that will be able to select an acupuncture prescription in aiding acupuncture practitioners to provide a quality service and to focus on the treatment.
A Knowledge-Based Consulting System for Chinese Herbal Treatments of
Cancers (J. Shieh)
Cancers are diseases characterized by excessive, uncontrolled growth of abnormal cells that invade and destroy other tissues. Cancer is the leading cause of death in Canada and second only to heart disease in the United States. Each year more than 1.2 million Americans and 130,000 Canadians are diagnosed with cancer, and more than 1,700 people die from cancer each day in the United States and Canada. The major conventional treatments of cancers currently available are surgery, radiation therapy, and chemotherapy. In addition to the mainstream treatments, Chinese herbal medicine is an important alternative method for cancer treatment. Herbs contain a large number of naturally occurring chemicals that have biological activity. There are more than 6000 herbs in the traditional Chinese pharmacopoeia. Scientific and clinical evidence has shown that more than 100 Chinese herbs can be used to treat cancer patients effectively. Chinese herbs can stimulate the immune system, improve the body condition, reduce cancer symptoms, control and restrict the rapid development of cancer cells, and protect healthy organs. However, the application of Chinese herbal medicine should be under the theories of TCM and in accordance with the actual condition of diseases, the characters and functions of herbs, and therapeutic requirements. The formation of a prescription is neither simply to pile up herbs with similar functions and effects, nor to treat the symptoms only. It is on the basis of the differentiation of syndromes and the establishment of the therapeutic methods, to pick the proper herbs and/or formulas and organically constitute them in light of the principles of forming a prescription. The formation of a prescription is further accommodated in light of concrete conditions such as the state of illness, sex, age, the patient's constitution, the occurring season and the circumstances. This ensures the suitability of the herbs to the stage of the illness and to achieve the therapeutic effects. One of my researches is to develop a knowledge-based computer consulting system to: (1) explain the functions and side effects of an individual herb, (2) analyze and report the functions and side effects of a prescription, (3) recommend a prescription for a cancer patient according to the patientís symptoms and relevant information, (4) show similar cases that have been effectively treated, and (5) help a Doctor of TCM to quickly form a patient's prescription.