# US 8,787,635 B2 · Optimization of multiple candidates in medical device or feature tracking > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,787,635 B2 - Document ID: US8787635B2 - Status: Granted - Kind: B2 - Application number: 13/097,497 - Filed: 2011-04-29 - Published: 2012-01-05 - Granted: 2014-07-22 - Earliest claimed: 2010-05-03 - Assignees: Siemens Aktiengesellschaft - Applicants: Renbin Peng; Peng Wang; Terrence Chen; Dorin Comaniciu - Inventors: Renbin Peng; Peng Wang; Terrence Chen; Dorin Comaniciu - Web page: https://patentgrep.com/patent/US8787635B2 ## CPC - A61B 8/12: in body cavities or body tracts, e.g. by using catheters - A61B 6/12: Arrangements for detecting or locating foreign bodies - A61B 6/487: {involving fluoroscopy} - A61B 6/504: {for diagnosis of blood vessels, e.g. by angiography} - A61B 6/5217: {extracting a diagnostic or physiological parameter from medical diagnostic data} - G16H 50/30: for calculating health indices; for individual health risk assessment - A61B 2034/2063: {Acoustic tracking systems, e.g. using ultrasound} - A61B 2034/2065: {Tracking using image or pattern recognition} - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA ## Abstract Multiple candidates are optimized in medical device or feature tracking. Possible locations of medical devices or features for each of a plurality of different times are received. The possible locations of devices are modeled using a probability function. An iterative solution to obtain the maximum of the probability function determines the possible locations to be used as the locations of the medical devices or features for each time. Where two or more medical devices or features are provided with a geometric relationship, such as being connected by a detected guide wire, the probability function may account for the geometric relationship, such as a geodesic distance between the possible locations for the two medical devices. ## Claims ### Claim 1 A method for determining a first location of a first medical device in imaging, the method comprising: obtaining a plurality of first candidate locations of a first medical device for each frame of a plurality of frames of medical data representing a patient, wherein at least one first candidate location of the plurality of first candidate locations does not represent the first medical device; obtaining a plurality of second candidate locations of a second medical device for each frame of the plurality of the frames of the medical data; selecting, with a processor, the first location of the first medical device in each frame of the plurality of frames from the plurality of first candidate locations for the respective frame, the selecting being iterative as a function of a probability function, the probability function being a function of the plurality of first candidate locations and the plurality of second candidate locations and distances between the plurality of first candidate location and the plurality of second candidate locations; and outputting an indication of the first location for at least one frame of the plurality of frames. ### Claim 2 (depends on 1) The method of claim 1 wherein selecting comprises identifying with the distances comprising geodesic distance along a curve. ### Claim 3 (depends on 1) The method of claim 1 further comprising obtaining the plurality of frames as angiography data with the first medical device comprising an intravascular ultrasound catheter, and the second medical device comprising a guide catheter. ### Claim 4 (depends on 3) The method of claim 3 wherein the plurality of first candidate locations are for a center of a transducer of the intravascular ultrasound catheter, and the plurality of second candidate locations are for a center of an end of the guide catheter. ### Claim 5 (depends on 1) The method of claim 1 wherein selecting comprises iteratively maximizing posterior probabilities of a Markov random chain, the Markov random chain being of the first locations in a sequence of the frames. ### Claim 6 (depends on 1) The method of claim 1 wherein selecting comprises optimizing the probability function, the probability function including differences between an inferred first location and a first candidate location of the plurality of first candidate locations nearest to the inferred first location, including differences between an inferred second location of the second medical device and a second candidate location of the plurality of second candidate locations nearest to the inferred second location, and including the distances between the plurality of first candidate locations and the plurality of second candidate locations. ### Claim 7 (depends on 6) The method of claim 6 wherein the differences between the inferred first location and the first candidate location and the differences between the inferred second location and the second candidate location comprise Euclidean distances. ### Claim 8 (depends on 1) The method of claim 1 wherein selecting comprises identifying with the probability function including probabilities for a first frame of the plurality of frames and probabilities for other frames of the plurality of frames, the probabilities for the other frames including predicted locations of the first location in the other frames, the probabilities being based on a Gaussian distribution. ### Claim 9 (depends on 1) The method of claim 1 further comprising: checking validity of the plurality of first candidate locations, for each frame of the plurality of frames, as a function of location relative to a reference; and removing, from the selecting, any frames of the plurality of frames where the plurality of first candidate locations are not valid. ### Claim 10 (depends on 1) The method of claim 1 further comprising: determining differences between the selected first locations of the plurality of frames; and interpolating the first location for any frame of the plurality of frames where the difference is greater than a threshold, the interpolation being from the first locations of other frames of the plurality of frames. ### Claim 11 In a non-transitory computer readable storage medium having stored therein data representing instructions executable by a programmed processor for optimizing multiple candidates in medical device or feature tracking, the non-transitory computer readable storage medium comprising instructions for: receiving a plurality of detected positions of a medical device or feature for each image frame of a plurality of image frames representing a patient, wherein at least one detected position of the plurality of detected positions does not represent the medical device or feature; and selecting one detected position of the plurality of detected positions in each image frame of the plurality of image frames, the selecting iteratively solving, for each image frame of the plurality of image frames, a probability function based on the plurality of detected positions in the image frame and a neighborhood of other image frames of the plurality of image frames. ### Claim 12 (depends on 11) The non-transitory computer readable storage medium of claim 11 wherein iteratively solving comprises solving as a function of a geodesic distance between the plurality of detected positions of the medical device and detected positions of another medical device, the geodesic distance being alone a curve. ### Claim 13 (depends on 11) The non-transitory computer readable storage medium of claim 11 wherein iteratively solving comprises solving the probability function, the probability function comprising probabilities for a first image frame of the plurality of image frames and probabilities for other image frames of the plurality of image frames, the probabilities for the other image frames including predicted positions of the selected one detected position in the other image frames, the probabilities being based on a Gaussian distribution. ### Claim 14 (depends on 11) The non-transitory computer readable storage medium of claim 11 further comprising: checking validity of the plurality of detected positions, for each image frame of the plurality of image frames, as a function of position relative to a reference; and removing, from the selecting, any of image frames of the plurality of image frames where the plurality of detected positions are not valid. ### Claim 15 (depends on 11) The non-transitory computer readable storage medium of claim 11 further comprising: determining differences between the selected ones of the plurality of detected positions of the plurality of image frames; and interpolating a device position for any image frames of the plurality of image frames where the difference is greater than a threshold, the interpolation being from the selected ones of the plurality of detected locations of other frames of the plurality of frames. ### Claim 16 (depends on 11) The non-transitory computer readable storage medium of claim 11 wherein receiving comprises receiving the plurality of image frames as fluoroscopy data with the medical device comprising an intravascular ultrasound catheter, and the plurality of detected positions are candidate locations of a center of the intravascular ultrasound catheter. ### Claim 17 (depends on 11) The non-transitory computer readable storage medium of claim 11 wherein iteratively solving comprises iteratively maximizing posterior probabilities of a Markov random chain. ### Claim 18 (depends on 11) The non-transitory computer readable storage medium of claim 11 wherein iteratively solving comprises optimizing the probability function, the probability function including differences between an inferred position and a detected position of the plurality of detected positions nearest to the inferred position, including differences between another inferred position of another medical device and another detected position of the other medical device nearest to the other inferred position, and including distances between the plurality of detected positions and the other detected positions, the differences comprising Euclidean distances and the distances comprising geodesic distances. ### Claim 19 A system for optimizing multiple candidates in medical device tracking, the system comprising: a memory configured to store data representing a region, at different times, of a patient and a catheter in the region; a processor configured to calculate a location of the catheter in the region as represented by the data, the location calculated by inferring from a plurality of possible locations in the data at the different times in a Markov random chain, wherein at least one possible location of the plurality of possible locations does not represent the catheter, the inferring using iterated conditional modes to maximize posterior probabilities of the Markov random chain; and a display configured to display the location as calculated for each of the different times. ### Claim 20 (depends on 19) The system of claim 19 wherein the posterior probabilities are a function of a geodesic distance between the plurality of possible locations of the catheter and possible locations of a catheter guide.