IP Library Granted Patent US 12,572,868
Granted Patent B2
US 12,572,868 · App. 17/325,247 · Granted Mar 10, 2026

Systems and methods for using geometry sensitivity information for guiding workflow

Inventors: Sethuraman Sankaran (Palo Alto, CA); Leo Grady (Darien, CT); Charles A. Taylor (Atherton, CA)
Assignee: Heartflow, Inc.
G06Q10/0633G06F30/33G06Q10/063G16B15/00G16H50/50
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,572,868
App. No.
17/325,247
Granted
Mar 10, 2026
Kind
B2
Abstract

Systems and methods are disclosed for using geometry sensitivity information for guiding workflows in order to produce reliable models and quantities of interest. One method includes determining a geometric model associated with a target object; determining one or more quantities of interest; determining sensitivity information associated with one or more subdivisions of the geometric model and the one or more quantities of interest; and generating, using a processor, a workflow based on the sensitivity information.

Claims (70)

1 . A computer-implemented method, comprising:

generating a three-dimensional estimation model of a geometry of patient-specific vascular anatomy based on received object patient-specific cardiac imaging data, the three-dimensional estimation model defining a set of geometries that are possible for the patient-specific vascular anatomy based on the received patient-specific cardiac imaging data, wherein the set of geometries that are possible for the patient-specific vascular anatomy is defined based on one or more of:

an accuracy associated with the received patient-specific cardiac imaging data;

a noisiness of the received patient-specific cardiac imaging data;

an artifact in the received patient-specific cardiac imaging data; or

a reconstruction algorithm used to generate the three-dimensional estimation model;

for at least one portion of the three-dimensional estimation model, performing a sensitivity evaluation that includes:

using a corresponding portion of each of the set of possible geometries to determine a blood flow characteristic for the patient-specific vascular anatomy, the blood flow characteristic having a value that is dependent on the geometry of the patient-specific vascular anatomy; and

defining a sensitivity for the at least one portion based on a variance in the blood flow characteristic over the set of possible geometries; and

modifying the at least one portion of the three-dimensional estimation model based on the defined sensitivity.

2 . The method of claim 1 , wherein, upon the sensitivity defined for a portion of the three-dimensional estimation model being above a predetermined threshold:

the portion is segmented into at least two further portions; and

the sensitivity evaluation is performed on the at least two further portions.

3 . The method of claim 1 , wherein the sensitivity is defined based on a standard deviation of the blood flow characteristic over the set of possible geometries.

4 . The method of claim 1 , wherein:

the three-dimensional estimation model is segmented into a set of portions; and

the sensitivity evaluation is performed on each of the set of portions.

5 . The method of claim 1 , further comprising:

using the modified three-dimensional estimation model to generate a sensitivity map for the geometry of the patient-specific vascular anatomy; and

outputting the sensitivity map via a display.

6 . The method of claim 5 , wherein outputting the sensitivity map includes:

generating a sequence of views of the modified three-dimensional estimation model, each view corresponding to a respective portion of the modified three-dimensional estimation model, the sequence of views being ordered based on sensitivity magnitude; and

outputting the sequence of views of the modified three-dimensional estimation model based on the generated sequence.

7 . The method of claim 5 , further comprising:

outputting a prompt for user approval of the at least one portion;

receiving user input including one or more of approval of or modification to the at least one portion; and

generating an approved three-dimensional model of the patient-specific vascular anatomy based on the modified three-dimensional estimation model and the received user input.

8 . The method of claim 1 , wherein:

the patient-specific vascular anatomy includes at least a portion of blood vessels of a person; and

the blood flow characteristic is fractional flow reserve for blood flow through the at least a portion of blood vessels.

9 . The method of claim 8 , wherein the patient-specific vascular anatomy includes one or more of a coronary vessel or a coronary tree.

10 . A system, comprising:

at least one memory storing instructions;

a display; and

at least one processor operatively connected to the memory, and configured to execute the instructions to perform operations, including:

generating a three-dimensional estimation model of a geometry of patient-specific vascular anatomy based on received patient-specific cardiac imaging data, the three-dimensional estimation model defining a set of geometries that are possible for the patient-specific vascular anatomy based on the received patient-specific cardiac imaging data;

for at least one portion of the three-dimensional estimation model, performing a sensitivity evaluation that includes:

using a corresponding portion of each of the set of possible geometries to determine a blood flow characteristic for the patient-specific vascular anatomy, the blood flow characteristic having a value that is dependent on the geometry of the patient-specific vascular anatomy; and

defining a sensitivity for the at least one portion based on a variance in the blood flow characteristic over the set of possible geometries;

modifying the at least one portion of the three-dimensional estimation model based on the defined sensitivity;

using the modified three-dimensional estimation model to generate a sensitivity map for the geometry of the patient-specific vascular anatomy; and

outputting the sensitivity map via a display.

11 . The system of claim 10 , wherein, upon the sensitivity defined for a portion of the three-dimensional estimation model being above a predetermined threshold, the at least one processor is further configured to:

segment the portion into at least two further portions; and

perform the sensitivity evaluation on the at least two further portions.

12 . The system of claim 10 , wherein the sensitivity is defined based on a standard deviation of the blood flow characteristic over the set of possible geometries.

13 . The system of claim 10 , wherein the at least one processor is further configured to:

segment the three-dimensional estimation model into a set of portions; and

perform the sensitivity evaluation on each of the set of portions.

14 . The system of claim 10 , wherein outputting the sensitivity map includes:

generating a sequence of views of the modified three-dimensional estimation model, each view corresponding to a respective portion of the modified three-dimensional estimation model, the sequence of views being ordered based on sensitivity magnitude; and

outputting the sequence of views of the modified three-dimensional estimation model based on the generated sequence.

15 . The system of claim 10 , further comprising:

an input device, wherein the operations further include:

outputting, via the display, a prompt for user approval of the at least one portion;

receiving, via the input device, user input including one or more of approval of or modification to the at least one portion; and

generating an approved three-dimensional model of the patient-specific vascular anatomy based on the modified three-dimensional estimation model and the received user input.

16 . A non-transitory computer-readable medium comprising instructions that are executable by at least one processor to perform operations, including:

generating a three-dimensional estimation model of a geometry of patient-specific vascular anatomy based on received patient-specific cardiac imaging data, the three-dimensional estimation model defining a set of geometries that are possible for the patient-specific vascular anatomy based on the received patient-specific cardiac imaging data;

for at least one portion of the three-dimensional estimation model, performing a sensitivity evaluation that includes:

using a corresponding portion of each of the set of possible geometries to determine a blood flow characteristic for the patient-specific vascular anatomy, the blood flow characteristic having a value that is dependent on the geometry of the patient-specific vascular anatomy; and

defining a sensitivity for the at least one portion based on a variance in the blood flow characteristic interest over the set of possible geometries;

modifying the at least one portion of the three-dimensional estimation model based on the defined sensitivity

using the modified three-dimensional estimation model to generate a sensitivity map for the geometry of the patient-specific vascular anatomy; and

outputting the sensitivity map via a display, by:

generating a sequence of views of the modified three-dimensional estimation model, each view corresponding to a respective portion of the modified three-dimensional estimation model, the sequence of views being ordered based on sensitivity magnitude; and

outputting the sequence of views of the modified three-dimensional estimation model based on the generated sequence.

17 . The non-transitory computer-readable medium of claim 16 , wherein, upon the sensitivity defined for a portion of the three-dimensional estimation model being above a predetermined threshold:

the portion is segmented into at least two further portions; and

the sensitivity evaluation is performed on the at least two further portions.

Assignments (3)
RELEASE OF SECURITY INTEREST Recorded Sep 11, 2025
From: HAYFIN SERVICES LLP
To: HEARTFLOW, INC.
Reel/Frame 072876/0775 →
SECURITY INTEREST Recorded Jun 18, 2024
From: HEARTFLOW, INC.
To: HAYFIN SERVICES LLP
Reel/Frame 067775/0966 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 20, 2021
From: SANKARAN, SETHURAMAN; GRADY, LEO; TAYLOR, CHARLES A.
To: HEARTFLOW, INC.
Reel/Frame 056301/0867 →
Continuity (3)
Continuation 15664384 · Jul 31, 2017
Continuation 14231837 · Apr 1, 2014
Related Publication 20210272030A1 · Sep 2, 2021
References Cited (30)
US 6904161B1 · Becker et al. · 2005 [cited by applicant]
US 7458936B2 · Zhou et al. · 2008 [cited by applicant]
US 8315812B2 · Taylor · 2012 [cited by applicant]
US 8812355B2 · Angell et al. · 2014 [cited by applicant]
US 8824752B1 · Fonte et al. · 2014 [cited by applicant]
US 20040047497A1 · Daw et al. · 2004 [cited by applicant]
US 20040193036A1 · Zhou et al. · 2004 [cited by applicant]
US 20050075905A1 · Bennett et al. · 2005 [cited by applicant]
US 20080015910A1 · Reisz et al. · 2008 [cited by applicant]
US 20090005693A1 · Brauner et al. · 2009 [cited by applicant]
US 20090089107A1 · Angell et al. · 2009 [cited by applicant]
US 20090125840A1 · Squilla et al. · 2009 [cited by applicant]
US 20100036254A1 · Owsley et al. · 2010 [cited by applicant]
US 20120041301A1 · Redel · 2012 [cited by applicant]
US 20120041323A1 · Taylor et al. · 2012 [cited by applicant]
US 20120072190A1 · Sharma et al. · 2012 [cited by applicant]
US 20120203530A1 · Sharma et al. · 2012 [cited by applicant]
US 20130129165A1 · Dekel et al. · 2013 [cited by applicant]
US 20130132054A1 · Sharma et al. · 2013 [cited by applicant]
US 20130251219A1 · Mehta · 2013 [cited by applicant]
US 20140249784A1 · Sankaran et al. · 2014 [cited by applicant]
US 20140275947A1 · Fonte et al. · 2014 [cited by applicant]
US 20150254418A1 · Sankaran et al. · 2015 [cited by applicant]
US 20170039340A1 · Sankaran et al. · 2017 [cited by applicant]
WO WO0239899A2 · 2002 [cited by applicant]
Sankaran et al. in Journal of Biomechanical Engineering (2011) vol. 133:12 pages. [cited by examiner]
Marino et al. in Journal of Theorethical Biology (2008) vol. 254:178-196). [cited by examiner]
International Search Report and Written Opinion for corresponding Application No. PCT/US2015/022945 dated Jul. 15, 2015 (9 pages). [cited by applicant]
Dibyendu Sengupta. 2013. Risk Assessment Using Image-Based Hemodynamic Modeling of Patients with Coronary Artery Aneurysms Caused by Kawasaki Disease. Ph.D. Dissertation. University of California at San Diego, La Jolla,… [cited by applicant]
Sankaran S., Esmaily Moghadam M, Kahn AM, Tseng EE, Guccione JM, Marsden AL.. Patient-specific multiscale modeling of blood flow for coronary artery bypass graft surgery. Ann Biomed Eng. Oct. 2012; 40(10):2228-42. Doi: … [cited by applicant]