IP Library Granted Patent US 12711620
Granted Patent B2
US 12711620 · App. 18/473,101 · Granted Aug 18, 2026

Cancer mapping using machine learning

Inventors: Shyam Natarajan (Los Angeles, CA); Alan Martin Priester (Mission Viejo, CA); Joshua Dylan Shubert (Los Angeles, CA); Jeremy Tet Kong Bong (Santa Monica, CA); Brittany Nan Pusey (Seattle, WA)
Assignee: AVENDA HEALTH, INC.
G06T7/0012A61B6/469G16H30/20G16H30/40G16H50/20G16H50/30G06T7/11G06T2207/10088G06T2207/20081G06T2207/20084G06T2207/30081G06T2207/30096
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 12711620
App. No.
18/473,101
Granted
Aug 18, 2026
Kind
B2
Abstract

A method for mapping cancer includes inputting data elements from medical images, biopsy, and biopsy pathology labels into a machine learning model estimating the likelihood of clinically significant cancer in a patient and outputting, via the machine learning model, an estimate of the clinically significant cancer likelihood at each voxel of a three-dimensional image.

Claims (36)

1 . A device for mapping cancer, comprising:

a processor electrically coupled to a memory component storing electronic instructions that, when executed by the processor, cause the device to execute a machine learning algorithm configured to receive inputs and produce an output based on the inputs, wherein:

the inputs comprise data elements from a medical image of a patient; and

the output comprises:

an estimate of clinically significant cancer likelihood in the patient; at each voxel of a three-dimensional image;

a lesion contour representing a lesion size of a cancer lesion in the patient; and an encapsulation confidence score representing an estimated likelihood that the lesion contour encompasses all clinically significant prostate cancer in the patient.

2 . The device of claim 1 , wherein:

the inputs further comprise at least one of prostate specific antigen (PSA) or fusion based biopsy data.

3 . The device of claim 1 , wherein the output further comprises at least one of a cancer estimation map (CEM), an estimate of cancer stage, or an estimated likelihood of extracapsular extension (ECE).

4 . The device of claim 3 , wherein the CEM illustrates a color-coded heat map representing a likelihood of cancer at each voxel of the three-dimensional image.

5 . The device of claim 1 , wherein the medical image is an MRI image of a patient's anatomy.

6 . The device of claim 5 , wherein the anatomy includes a prostate.

7 . The device of claim 1 , wherein the output further comprises a visual curve representing the encapsulation confidence score versus the lesion size.

8 . The device of claim 7 , wherein the visual curve includes a point representing a certain lesion size and a certain encapsulation confidence score.

9 . The device of claim 8 , wherein the point is configured to be visually manipulated along the visual curve to change the certain lesion size and the certain encapsulation confidence score represented by the point.

10 . The device of claim 9 , wherein manipulating the point alters the lesion contour.

11 . The device of claim 1 , wherein the medical image includes an MRI image.

12 . A method for mapping cancer, comprising:

inputting data elements from medical images of a patient into a machine learning model estimating a likelihood of clinically significant cancer in a patient; and

generating an output, via the machine learning model, including an estimate of the clinically significant cancer likelihood at each voxel of a three-dimensional image, the output further comprising:

an adjustable lesion contour representing at least one of a size of a cancer lesion or a margin determination for the cancer lesion in the patient; and

an encapsulation confidence score representing a probability of all clinically significant prostate cancer in the patient being within a specified lesion contour.

13 . The method of claim 12 , further comprising inputting data elements from at least one of a biopsy, biopsy pathology labels, or fusion based biopsy data into the machine learning model.

14 . The method of claim 12 , wherein the machine learning model is trained on a population data set including the data elements.

15 . The method of claim 12 , wherein the output includes a visual representation of the three-dimensional image with a color-coded heat map representing the clinically significant cancer likelihood at each voxel.

16 . A method for mapping cancer, comprising:

inputting data elements from medical images of a patient into a machine learning model estimating a likelihood of clinically significant cancer in the patient; and

displaying a visual representation of the likelihood at each voxel of a three-dimensional image, the visual representation comprising:

a cancer estimation map (CEM) illustrating a color-coded heat map representing the likelihood of clinically significant cancer overlying the image, the CEM including a lesion contour representing a size of a cancer lesion; and

a curve representing an encapsulation confidence score versus the lesion size, the curve including a point representing the lesion size and the encapsulation confidence score;

wherein:

the point is configured to be visually manipulated along the curve to change the lesion size and the encapsulation confidence score represented by the point; and

manipulating the point alters the lesion contour.

17 . The method of claim 16 , further including displaying an interventional instrument in a position relative to the image.

18 . The method of claim 17 , wherein a position of the interventional instrument is configured to be altered relative to the image.

19 . The method of claim 17 , further comprising displaying a location of a biopsy core overlying the image.