Yuki Ishida, Presenter: Nothing to Disclose

Toshihiro Ogura PhD, Abstract Co-Author: Nothing to Disclose

Norio Hayashi PhD, Abstract Co-Author: Nothing to Disclose

Mitsuru Sato, Abstract Co-Author: Nothing to Disclose

Mika Okajima, Abstract Co-Author: Nothing to Disclose

Kunio Doi, Abstract Co-Author: Nothing to Disclose

During angiographic examinations, interventional radiologists need various image manipulation such as paging, roaming, enlarging and fusion in order to assess vessels and sources of bleeding. However, radiologist could not touch screen by hand, because of his/her gloves to be kept clean. Usually, radiologists provide verbal instructions to technologists, who may operate to display appropriate images with various image processing. However, if technologist may not be at the console due to the preparation of the catheter or contrast agent, quick operation may not be performed. Therefore, we developed a new image operation system using a motion sensor for angiography. With this system, radiologists could operate various image processing only with movements in hands.

We used the Kinect technique which is a gesture recognition technology to read the movement of radiologists. The Kinect sensor consisted of an infrared laser emitter, an infrared camera and an RGB camera for gesture recognition. Measurements of depth were made by triangulation using the infrared camera. For initialization, radiologist’s hands were recognized by computer for right palm forward protrusion.

The recognition time of the palms was 2.4±0.2 seconds for initialization of the system. The right hand moved the cursor and the left hand controlled the left mouse button. When the left hand was raised, the mouse button was held down. When the left hand was lowered, the mouse button was released. For roaming images, he/she may grasp an image by holding the palm on the image, and then move it with the right hand. The distance for recognition of hands ranged from one meter to 2.6m, when a radiologist entered the recognition area far from the sensor. The lateral recognition range was 2.9m at 2.6m distance from the sensor.

With the image operation system using a motion sensor for angiography, various image processing can be performed with our hand movements, which would be a new technology for angiography in operating rooms.

This system can be implemented as a useful tool to radiologists for control of image viewing without touching the workstation in existing angiography system in operating rooms.

Ishida, Y, Ogura, T, Hayashi, N, Sato, M, Okajima, M, Doi, K, Development of a New Image Operation System with Hand Movements Using a Kinect Sensor for Angiography.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13019136.html