Session Chair: Theodoros Arvanitis

Virtual Reality in Biomedicine: revolutionising the provision of patient care
Theodoros N. Arvanitis
Biomedical Informatics, Signals and Systems Research Lab
Department of Electronic, Electrical & Computer Engineering, School of Engineering
The University of Birmingham, UK

Continual achievements in technology development have dramatically changed the ways in which we conduct our daily activities and life. The practice of biomedicine and the provision of healthcare is not an exception. Technological advances during the past several decades have had an enormous impact on the way in which we diagnose and treat disease. Today, advances in science and engineering involving microelectronics, biochips, genomics, and biomaterials raise the potential for medical technology to expand the frontier of healthcare to unimaginable accomplishments. Virtual Reality technologies have played an important role in revolutionising the practical provision of patient care. In recent years, VR technologies and their application in biomedicine are not any more a research curiosity; in several areas of biomedical and clinical disciplines such technologies have matured enough to be adopted in routine practice, providing powerful tools in diagnostics, therapeutic planning and interventions.

User immersion in a synthetic world is one of the main characteristics of Virtual Reality, as a form of human-machine interaction. In biomedicine, the concept of VR is broader than its traditional interpretation of immersion through technology. Virtual and real objects are fused in a unified space, in order to provide appropriate visualisation of and behavioural interaction with human anatomy and physiology. Thus, virtual reality, biomedical imaging and medical informatics related technologies are used in synergy for the creation of a broad range of applications that facilitate routine practice in the biomedical field. Such applications include a varied range of tools for; simulation, planning and performance of surgical procedures; virtual diagnostic techniques appropriate for screening (e.g. virtual endoscopic procedures, such as colonoscopy); visualisation of imaging data in the form of virtual anatomical atlases for medical education purposes; neuro-physiological assessment and rehabilitation of patients; biologic-specific cellular modelling in medicinal biochemistry applications, etc.
This paper will explore the technological quest of virtual reality in biomedicine and discuss its impact in the provision of healthcare services to the point of need from the perspective of research motivation, socio-economic effect, and medico-ethical concerns on the use and application of such technologies. Furthermore, we will conjecture on future technological directions and progress in this field.

Intervention Records by Video Image for Emergency Medical Care
Hirotsugu ASAI, Satoru HAYAMIZU, Youhei ISHII,
Makoto KANAGAWA, Hideki TANAHASHI
Gifu Prefectural Research Institute of Manufacturing Information Technology, Japan
Department of information science faculty of Engineering, Gifu University, Japan
Human Ecology Research Center, Sanyo Electric Co., Japan

The attempt to record intervention by video image is increasing especially in emergency medical care and surgery. However, it is not considered enough how to record and how to use the records. So we proposed the framework to record video image of intervention associated with treatments easily by using ACLS (Advanced Cardiovascular Life Support) which is a standard treatment program and intuitive camera control interface and confirmed its effectiveness.

Framework for Annotation of Haptic Data in Simulated Surgical Procedures
*Mikko RISSANEN , *Naoto KUME , **Yoshihiro KURODA ,
***Megumi NAKAO , ****Tomohiro KURODA , ****Hiroyuki YOSHIHARA
*Graduate School of Informatics, Kyoto University, Japan
**Graduate School of Medicine, Kyoto University, Japan
***Graduate School of Information Science, NAIST, Japan
**** Dept. of Medical Informatics, Kyoto University Hospital, Japan

Surgical skills that are increasing in complexity require extensive practice and virtual reality is expected to solve problems in surgical training. This paper proposes a composite recording and playback model for surgical simulation, thus enabling development of annotation framework that would allow simulation to be enhanced with textbook characteristics. Usually, manipulation data of simulation from haptic interface is recorded and playback of the simulation is produced through re-simulation, which results in difficult control of the playback. Alternatively, states of graphic objects can be stored, but this strategy loses the original manipulation data and is not scalable. The composite model combined the two strategies by using offline re-simulation and improved scalability of graphics recording without losing high quality manipulation data.

EDOG – An Application for Early Detection of Glaucoma
*J. Vernon Odom, **Prasad Ghude, ** Frances L. VanScoy, ***James Tong
*WVU Eye Institute, Morgantown, USA;
** Lane Dept. of CSEE, West Virginia University, Morgantown, USA
*** Computer Science Dept., Carnegie Mellon University, Pittsburgh, USA

Glaucoma is a major debilitating eye disease for humans. It steals the sight without warning and often without symptoms. Glaucoma results in reduced visual function and loss of visual field, secondary to the loss of visual cells in the retina. The vision loss is caused by damage to the optic nerve. Increased intraocular pressure is a sensitive but not very specific indicator of glaucomatous damage. Therefore, additional tests to determine the absence or presence of visual loss would aide the early detection of damage due to glaucoma.
The purpose of our efforts is to develop several candidate methods of early detection of damage due to glaucoma. An application has been developed using Microsoft visual basic to test visual performance in four types of visual discrimination: Vernier acuity, Dot counting, Line length discrimination and Orientation discrimination. In each test, patients can be tested with and without noise in a two-alternative, spatial forced-choice tracking procedure. The meaning of noise varies with test, however, in all cases; the observer specifies the standard deviation of a Gaussian distribution from which the noise is drawn. The ultimate goal is to obtain estimates of "detection efficiency" and "internal noise". By obtaining detection scores at several noise levels, it is possible to estimate observer efficiency and internal noise.