3.3.1 Technological issues 
System Latency 
An essential component of see-trough HMDs is the capacity to 
properly register a users surrounding and the synthetic space. A 
geometric calibration between the tracking devices and the 
HMD must be performed. The major impediment to achieving 
registration is the gap in time, referred as lag, between the 
moment when the HMD position is measured and the moment 
when the synthetic image for that position is fully rendered and 
presented to the user. 
Real Scene Resolution and Distortion 
The best real-scene resolution that a see-trough device can 
provide is that perceived with the unarmed eye under unit 
magnification of the real scene. Optical see-trough HMDs take 
what might called a *minimal obtrusive approach; that is, they 
leave the view of the real world nearly intact and attempt to 
augment it by merging a reflected image of the computer- 
generated scene into the view of the real world. Video see- 
trough HMDs are typically more obtrusive in the sense that they 
block out the real-world view in exchange for the ability to 
merge the two views more convincingly. 
Overlay and Peripheral Field of View 
The term overlay FOV is defined as the region of the FOV 
where graphical information and real information are 
superimposed. The peripheral FOV is the real-world FOV 
beyond the overlay FOV. Large FOV is especially important for 
tasks that require grabbing and moving objects. Most current 
high-resolution HMDs achieve higher resolution at the expense 
of a reduced FOV. In surgery the resolution is more important 
than a large FOV. 
Viewpoint Matching 
In video see-trough HMDs, the camera viewpoint (the entrance 
pupil) must be matched to the viewpoint of the observer (the 
entrance pupil of the eye) 
Engineering and Cost Factors 
HMD designs often suffer from fairly low resolution, limited 
FOV, poor ergonomic designs and excessive weight. A good 
ergonomic design requires an HMD whose weight is similar to 
a pair of eyeglasses. To our knowledge, at present, no large- 
FOV stereo see-trough HMDs of any type are comparable in 
weight to a pair of eyeglasses. 
3.3.3 Human Factor / Perceptual issues 
The following issues could be discussed from both a 
technological and human-factors standpoint: 
User Acceptance and Safety 
Perceived Depth 
- Occlusion 
- Rendered Locations of Objects in Depth 
- FOV and Frame-Buffer Overscan 
- Specification of Eyepoint Location 
- Residual Optical Distortions 
- Perceived Location of Objects in depth 
Adaption 
Peripheral FOV 
Depth of field 
Qualitative Aspects 
3.3.3 Design issues 
The following design issues are important aspects of augmented 
reality systems and wearable computers: 
Display Technology 
Input / Output Devices 
Power Supplies 
Image Registration Techniques 
Required Accuracy 
3.3.4 Clinical and technical requirements 
Our extension is to use augmented reality not only for the 
preoperative surgical planning. For an intraoperative solution 
we require a system with very good real time quality. In this 
context we must achieve high accuracy in tracking and 
registration. So we estimate the required accuracy of 
registration under 1 cm. In optical case of HMD, the virtual 
image is projected at some distance away from the user. This 
distance should be adjustable, although it is often fixed. 
Therefore, the virtual objects are all projected to the same 
distance while the real objects are at varying distances from the 
user. If the virtual and real distances are not matched for the 
particular objects that the physician is looking at, it may not be 
possible to clearly view both simultaneously (Azuma, 2001). In 
our case, the virtual objects should be projected in the distance 
of the working hand of the surgeon. 
The surgeons expect an improved orientation during the 
intervention by the three-dimensional visualisation of the 
complex structure and context of the organ's anatomy. A typical 
task of an IGSS is the virtual depiction of the surgical 
instruments in spatial relation to the individual anatomy of the 
patient. Therefore, preoperative CT- and MRI data are post 
processed and enhanced with data from interventional planning. 
These data are then registered with the current situs during an 
intervention and readapted to the current deformation of the 
organ. In many cases the mutual depiction of pre and intra 
interventional data is required, which permits the specific 
selection of the kind of data that is currently of interest during 
the intervention. The depiction of the surgical instrument in 
relation to the anatomy and the chosen data from planning is 
important for the ability of the surgeon to orientate by means of 
the virtual visualisations. Therefore, the visual depth perception 
and a stereoscopic AR system are important for intraoperative 
orientation and navigation. The display devices used in our 
application may have less stringent requirements than for 
example accuracy or stereoscopic view. Monochrome displays 
may be adequate for our application. Furthermore, the 
resolution of the monitor in an optical see-trough HMD might 
be high enough for our three-dimensional virtual data because 
the low resolution of a see-trough HMD does not reduce the 
resolution of the real environment. The whole AR systems 
should be designed in such a quality, that it is well accepted by 
the surgeons. In order to achieve this requirement, the system 
should be not only fast and accurate the components should be 
easy, robust and relatively inexpensive. Within the operation 
theatre more than one surgeon might want enhanced reality and 
observe the operation with AR techniques. All hardware must 
fulfil the sterile conditions in the operation theatre. 
3.4 Concept for augmented reality in liver surgery 
The prerequisite for augmented reality in liver surgery is given 
with Module 1 to 4 from our ARIONTM System described in 
chapter 2. The future extension should visualise the data with a 
—278—