• Hardware
• Software
• Applications
• • Healthcare
  • • Objectives of the user tests
    • Participants in the tests
    • Task scenario
    • Experiment setup
    • Testing methodology and procedure
    • Data collection and analysis
    • Results
  • Emergency response
  • Maintenance
  • Production

Technical Lessons from Deployment Test

Despite intensive lab validation before the deployment, technical problems have invariably emerged during the test. Detailed analysis is currently underway. Key preliminary findings are:

• The person independence of the gesture recognition was not satisfactory. While the gesture recognition worked fine for one of the doctors (as well as for test persons in our lab), the other had major problems handling it. 
• The procedures needed for system initialization and user registration were not robust enough. Obviously in many cases switching devices on/off, registering users, and coming into/out of range of Bluetooth nodes has been performed in sequence and at speeds which we did not consider in our lab trials. This often caused components to freeze. At the same time the ward round needed to proceed 
• Similar problems were caused by problems with wireless links. In particular when there were many people in the room (to view a demonstration of the system) Bluetooth links or the proximity detection signals would be blocked. In addition the way doctors were handling the wrist worn RFID reader combined with the new mounting often resulted in occlusions of the Bluetooth communication.

User Feedback Conclusions

• The staff reactions were strongly overshadowed by the technical issues described above. Clearly the problems with gesture training were, like in the first prototype, a major critique point and people were suggesting that speech control might be better. The fact that the system allows hands to remain sterile was nonetheless seen as positive.
• The patients were positive about the system. They stated that seeing the results on the screen and having them explained in real time by the doctor increases patient’s trust in staff. None of the patients interviewed felt such a system would have an effect on their re-ationship with staff, as the system will not prevent them from talking to the doctors and nurses. The patients interviewed stated that they would be happy to see such a system working in the hospitals in the future.

User Feedback Conclusions

As a conclusion of the test we have decided to

• Go from macroscopic, acceleration sensor based gesture to micro gestures (small finger based gestures) based on a capacitive, “touch less” sensor. Such sensor guarantees 100 recognition rates. They also correspond to what users have stated as being acceptable.
• Go from Bluetooth connections to ZigBee which is much more robust. In addition the number of components in the system will be reduced (e.g. the doctor will have only wearable interaction devices, no computing device).