Biomedical subjects

Takeo Kanade

Publications and source records attributed to Takeo Kanade.

5 recordsLinked to original sources

PubMed2006

Meticulously detailed eye region model and its application to analysis of facial images.

We propose a system that is capable of detailed analysis of eye region images in terms of the position of the iris, degree of eyelid opening, and the shape, complexity, and texture of the eyelids. The system uses a generative eye region model that parameterizes the fine structure and motion of an eye. The structure parameters represent structural individuality of the eye, including the size and color of the iris, the width, boldness, and complexity of the eyelids, the width of the bulge below the eye, and the width of the illumination reflection on the bulge. The motion parameters represent movement of the eye, including the up-down position of the upper and lower eyelids and the 2D position of the iris. The system first registers the eye model to the input in a particular frame and individualizes it by adjusting the structure parameters. The system then tracks motion of the eye by estimating the motion parameters across the entire image sequence. Combined with image stabilization to compensate for appearance changes due to head motion, the system achieves accurate registration and motion recovery of eyes.

Algorithms↗

PubMed2005

Sensor guided ablation procedure of left atrial endocardium.

In this paper, we present a sensor guided ablation procedure of highly motile left atrium. It uses a system which automatically registers the 4D heart model with the position sensor on the catheter, and visualizes the heart model and the position of the catheter together in real time. With this system clinicians can easily map the motile left atrium shape and see where the catheter is inside it, therefore greatly improve the efficiency of the ablation operation.

Catheter Ablation↗

PubMed2003

Ultrasound registration of the bone surface for surgical navigation.

OBJECTIVE: To allow non-invasive registration of the bone surface for computer-assisted surgery (CAS), this investigation reports the development and evaluation of intraoperative registration using 2D ultrasound (US) images. This approach employs automatic segmentation of the bone surface reflection from US images tagged with the 3D position to enable the application of CAS to minimally invasive procedures. METHODS: The US-based registration method was evaluated in comparison to point-based registration, which is the predominant method in current clinical use. The absolute accuracy of the US-based registration was determined using a phantom pelvis, with fiducial registration providing the ground truth. The relative accuracy was determined by an intraoperative study comparing the US registration to the point-based registration obtained as part of the HipNav experimental protocol. RESULTS: The phantom pelvis study demonstrated equivalent accuracy between point- and US-based registration under in vitro conditions. In the intraoperative study, the US-based registration was sufficiently consistent with the point-based registration to warrant larger-scale clinical trials of this non-invasive registration method. CONCLUSION: Ultrasound-based registration eliminates the need for physical contact with the bone surface as in point-based registration. As a result, non-invasive registration could fully unlock the potential of computer-assisted surgery, enabling development of the next generation of minimally invasive surgical procedures.

Humans↗

PubMed2003

Automatic recognition of eye blinking in spontaneously occurring behavior.

Previous research in automatic facial expression recognition has been limited to recognition of gross expression categories (e.g., joy or anger) in posed facial behavior under well-controlled conditions (e.g., frontal pose and minimal out-of-plane head motion). We have developed a system that detects a discrete and important facial action (e.g., eye blinking) in spontaneously occurring facial behavior that has been measured with a nonfrontal pose, moderate out-of-plane head motion, and occlusion. The system recovers three-dimensional motion parameters, stabilizes facial regions, extracts motion and appearance information, and recognizes discrete facial actions in spontaneous facial behavior. We tested the system in video data from a two-person interview. The 10 subjects were ethnically diverse, action units occurred during speech, and out-of-plane motion and occlusion from head motion and glasses were common. The video data were originally collected to answer substantive questions in psychology and represent a substantial challenge to automated action unit recognition. In analysis of blinks, the system achieved 98% accuracy.

Adult↗

PubMed1998

Temporal photoreception for adaptive dynamic range image sensing and encoding.

We have implemented two analog VLSI computational sensors for sensing and encoding high dynamic range images by exploiting temporal dimension of photoreception. The first sensor is a multi-integration time photoreceptor that automatically adapts to use different integration periods depending on light intensity. It exhibits a dynamic range 128 times larger than that of a single integration period photoreceptor, approximately 1:128000. The second sensor is an intensity-to-time processing paradigm that is based on the notion that stronger stimuli elicit responses before weaker ones. The paradigm sorts pixels of sensed images by their intensities, thus achieving information-theoretic optimal encoding of images. It handles dynamic range of approximately 1:1000000. Both implementations can operate at standard video rate of 30framess(-1).

Journal Article↗