Granted Patent B2

US 12,490,953 · App. 18/397,826 · Granted Dec 9, 2025

Inventors: Davinder S. Dhatt (Bothell, WA); Christopher Aleksandr White (Bothell, WA); Adam Benjamin Pely (Bothell, WA); Thomas Michael Duffy (Bothell, WA); Paul Tomotaro Danset (Bothell, WA); Diku Pranav Mandavia (Bothell, WA)

Assignee: FUJIFILM SONOSITE, INC.

A61B8/085A61B8/0891A61B8/12A61B8/445A61B8/461A61B8/467A61B8/5207A61B8/5223G06N3/02

Abstract

Claims (82)

1 . A computing device to implement an ultrasound system for guiding insertion of an instrument, the computing device comprising:

a memory to maintain one or more ultrasound images;

a neural network implemented at least partially in hardware of the computing device to identify blood vessels in the one or more ultrasound images and determine diameters of the blood vessels including a first diameter of a first blood vessel; and

a processor system to:

determine, for one or more of the blood vessels, an instrument size based on the diameters of the one or more of the blood vessels;

select a color based on the instrument size;

indicate, in the one or more ultrasound images, the one or more of the blood vessels with the color;

determine a threshold insertion length of the instrument to be inserted into the first blood vessel based on the first diameter, the threshold insertion length indicating an amount of the instrument inside the first blood vessel;

determine a first length of the instrument based on the threshold insertion length;

display in the one or more ultrasound images an indication of the first length of the instrument to be inserted into the first blood vessel;

display an outline of the first blood vessel with the color;

assign one of a vein classification and an artery classification to the first blood vessel;

determine a confidence level for the one of the vein classification and the artery classification assigned to the first blood vessel; and

cause display in the one or more ultrasound images of the outline of the first blood vessel with an opacity based on the confidence level for the one of the vein classification and the artery classification.

2 . The computing device as described in claim 1 , wherein the processor system is implemented to

display an outline of the first blood vessel with the color that matches a color coding of the largest instrument size for the first blood vessel.

3 . The computing device as described in claim 1 , wherein the processor system is implemented to:

receive a user selection of at least the first blood vessel; and

cause display in the one or more ultrasound images of the first diameter of the selected the at least the first blood vessel.

4 . The computing device as described in claim 1 , wherein:

the neural network is implemented to determine depths of the blood vessels including a first depth of the first blood vessel in the one or more ultrasound images; and

the processor system is implemented to:

receive a user selection of at least the first blood vessel; and

cause display in the one or more ultrasound images of the first depth of the at least the first blood vessel determined by the neural network.

5 . The computing device as described in claim 1 , wherein:

the neural network is implemented to determine depths of the blood vessels, including a first depth of the first blood vessel in the one or more ultrasound images; and

the processor system is implemented to:

determine that the first depth is a shallowest depth of the depths; and

cause display in the one or more ultrasound images, automatically and without user intervention, of the first depth and the first diameter of the first blood vessel determined by the neural network.

6 . A computer implemented method for an ultrasound system to guide insertion of an instrument, the method comprising:

identifying, using a neural network, blood vessels in one or more ultrasound images stored in a memory;

determining diameters of the blood vessels including a first diameter of a first blood vessel;

determining, for one or more of the blood vessels, an instrument size based on the diameters of the one or more of the blood vessels;

selecting a color based on the instrument size;

indicating, in the one or more ultrasound images, the one or more of the blood vessels with the color;

determining a threshold insertion length of the instrument to be inserted into the first blood vessel based on the first diameter, the threshold insertion length indicating an amount of the instrument inside the first blood vessel;

determining a first length of the instrument based on the threshold insertion length;

displaying in the one or more ultrasound images an indication of the first length of the instrument to be inserted into the first blood vessel;

displaying an outline of the first blood vessel with the color;

assigning one of a vein classification and an artery classification to the first blood vessel;

determining a confidence level for the one of the vein classification and the artery classification assigned to the first blood vessel; and

causing display in the one or more ultrasound images of the outline of the first blood vessel with an opacity based on the confidence level for the one of the vein classification and the artery classification.

7 . The method as described in claim 6 , further comprising:

displaying an outline of the first blood vessel with the color that matches a color coding of the largest instrument size for the first blood vessel.

8 . The method as described in claim 6 , further comprising:

receiving a user selection of at least the first blood vessel; and

causing display in the one or more ultrasound images of the first diameter of the selected the at least the first blood vessel.

9 . The method as described in claim 6 , further comprising:

determining, using the neural network, depths of the blood vessels including a first depth of the first blood vessel in the one or more ultrasound images;

receiving a user selection of at least the first blood vessel; and

causing display in the one or more ultrasound images of the first depth of the at least the first blood vessel determined by the neural network.

10 . The method as described in claim 6 , further comprising:

determining, using the neural network, depths of the blood vessels including a first depth of the first blood vessel in the one or more ultrasound images;

determine that the first depth is a shallowest depth of the depths; and

cause display in the one or more ultrasound images, automatically and without user intervention, of the first depth and the first diameter of the first blood vessel determined by the neural network.

11 . A non-transitory machine-readable medium storing executable instructions to cause a computing device to perform a method for an ultrasound system to guide insertion of an instrument comprising:

identifying, using a neural network, blood vessels in one or more ultrasound images stored in a memory;

determining diameters of the blood vessels including a first diameter of a first blood vessel;

determining, for one or more of the blood vessels, an instrument size based on the

diameters of the one or more of the blood vessels;

selecting a color based on the instrument size;

indicating, in the one or more ultrasound images, the one or more of the blood vessels with the color;

determining a threshold insertion length of the instrument to be inserted into the first blood vessel based on the first diameter, the threshold insertion length indicating an amount of the instrument inside the first blood vessel;

determining a first length of the instrument based on the threshold insertion length;

displaying in the one or more ultrasound images an indication of the first length of the instrument to be inserted into the first blood vessel;

displaying an outline of the first blood vessel with the color;

assigning one of a vein classification and an artery classification to the first blood vessel;

determining a confidence level for the one of the vein classification and the artery classification assigned to the first blood vessel; and

causing display in the one or more ultrasound images of the outline of the first blood vessel with an opacity based on the confidence level for the one of the vein classification and the artery classification.

12 . The non-transitory machine-readable medium, as described in claim 11 , wherein the method further comprises:

displaying an outline of the first blood vessel with the color that matches a color coding of the largest instrument size for the first blood vessel.

13 . The non-transitory machine-readable medium, as described in claim 11 , wherein the method further comprises:

receiving a user selection of at least the first blood vessel; and

causing display in the one or more ultrasound images of the first diameter of the at least the first blood vessel.

14 . The non-transitory machine-readable medium, as described in claim 11 , wherein the method further comprises:

determining, using the neural network, depths of the blood vessels including a first depth of the first blood vessel in the one or more ultrasound images;

receiving a user selection of at least the first blood vessel; and

causing display in the one or more ultrasound images of the first depth of the at least the first blood vessel determined by the neural network.

15 . The non-transitory machine-readable medium, as described in claim 11 , wherein the method further comprises:

determining, using the neural network, depths of the blood vessels including a first depth of the first blood vessel in the one or more ultrasound images;

determine that the first depth is a shallowest depth of the depths; and

cause display in the one or more ultrasound images, automatically and without user intervention, of the first depth and the first diameter of the first blood vessel determined by the neural network.

Continuity (2)

References Cited (79)