Methods and systems for detecting pleural irregularities in medical images

Patent metadata

Field	Value
Publication number	US-11627941-B2
Application number	US-202017005153-A
Country	US
Kind code	B2
Filing date	Aug 27, 2020
Priority date	Aug 27, 2020
Publication date	Apr 18, 2023
Grant date	Apr 18, 2023

Abstract

Various methods and systems are provided for a medical imaging system. In one embodiment, a method includes acquiring a series of medical images of a lung, identifying a pleural line in each medical image of the series, evaluating the pleural line for irregularities in each medical image of the series, and outputting an annotated version of each medical image of the series, the annotated version including visual markers for healthy pleura and irregular pleura. In this way, an operator of the medical imaging system may be alerted to pleural irregularities during a scan.

First claim

The invention claimed is:

1. A method, comprising:
acquiring a series of medical images of a lung according to a lung imaging protocol;
identifying a pleural line in each medical image of the series in real-time, while, for at least some medical images of the series, one or more additional medical images of the series are still being acquired, via a pleural position detection algorithm of an image processing system;
detecting irregular pleura of the pleural line in each medical image of the series in real-time, while, for at least some medical images of the series, one or more additional medical images of the series are still being acquired, via at least one pixel scoring algorithm of the image processing system that scores each pixel of the pleural line based on one or more of a vertical gap of the pleural line at a horizontal location of each pixel and a brightness of each pixel relative to neighboring pixels;
outputting an annotated version of each medical image of the series to a display in real-time, while, for at least some medical images of the series, one or more additional medical images of the series are still being acquired, the annotated version including visual markers for healthy pleura and irregular pleura positioned via algorithms of the image processing system; and
scoring each medical image of the series based on a percentage of the irregular pleura in the pleural line via scoring algorithms of the image processing system in real-time, while, for at least some medical images of the series, one or more additional medical images of the series are still being acquired.

2. The method of claim 1 , wherein the pleural line is a substantially horizontal segment of brighter pixels, and identifying the pleural line in each medical image of the series comprises:
evaluating consecutive images of the series of medical images via the pleural position detection algorithm of the image processing system in real-time, as the consecutive images are acquired, to determine an area having a highest amount of local change between the consecutive images; and
identifying an upper border and a lower border of the pleural line within the determined area based on a brightness change between pixels between the consecutive images via the pleural position detection algorithm of the image processing system in real-time, as the consecutive images are acquired.

3. The method of claim 2 , wherein the pleural position detection algorithm of the image processing system comprises an edge detection algorithm or a gradient change algorithm, and wherein identifying the upper border and the lower border of the pleural line is via the edge detection algorithm or the gradient change algorithm.

4. The method of claim 1 , wherein identifying the pleural line in each medical image of the series comprises identifying, via the pleural position detection algorithm of the image processing system, an upper border and a lower border of the pleural line based on a brightness change between pixels of consecutive images of the series of medical images, and wherein detecting irregular pleura of the pleural line via the at least one pixel scoring algorithm of the image processing system that scores each pixel of the pleural line based on one or more of the vertical gap of the pleural line at the horizontal location of each pixel and the brightness of each pixel relative to neighboring pixels comprises:
determining, via the at least one pixel scoring algorithm of the image processing system, an irregularity score for each pixel between the upper border and the lower border of the pleural line;
characterizing, via the at least one pixel scoring algorithm of the image processing system, a given pixel as healthy responsive to the irregularity score being less than a threshold score; and
characterizing, via the at least one pixel scoring algorithm of the image processing system, the given pixel as irregular responsive to the irregularity score being greater than the threshold score wherein the irregularity score is based on one or more of the vertical gap of the pleural line at the horizontal location of each pixel and the brightness of each pixel relative to neighboring pixels.

5. The method of claim 4 , wherein the irregularity score is a product of a first score and a second score for the given pixel.

6. The method of claim 5 , wherein the first score is determined based on the vertical gap of the pleural line at the horizontal location of the given pixel, the first score increasing as the vertical gap increases, the vertical gap representing a number of pixels vertically between the lower border of the pleural line at a location of the given pixel relative to a neighboring pixel.

7. The method of claim 5 , wherein the second score is determined based on the brightness of the given pixel relative to neighboring pixels, the second score increasing as the brightness of the given pixel decreases relative to the neighboring pixels.

8. The method of claim 4 , wherein the visual markers for the healthy pleura and the irregular pleura include a first visual marker positioned by the image processing system at each location of the pleural line having pixels characterized as healthy and a second visual marker positioned by the image processing system at each location of the pleural line having pixels characterized as irregular.

9. The method of claim 8 , wherein the first visual marker includes one or more of a different shape, color, and size than the second visual marker, and wherein one or both of the first visual marker and the second visual marker is positioned by the image processing system along the upper border and/or the lower border of the pleural line.

10. The method of claim 1 , further comprising:
outputting a highest scoring medical image of the series to the display in response to completion of the lung imaging protocol.

11. The method of claim 1 , wherein acquiring the series of medical images of the lung according to the lung imaging protocol includes acquiring ultrasound imaging data, magnetic resonance imaging data, computed tomography data, x-ray data, and positron emission tomography data according to the lung imaging protocol.

12. A method, comprising:
generating a plurality of lung ultrasound images from ultrasonic signals acquired according to a protocol;
identifying irregular pleura in each of at least a subset of the plurality of lung ultrasound images in real-time while remaining ultrasonic signals are still being acquired according to the protocol by executing, via a processor of an image processing system, a first set of instructions stored in non-transitory memory of the image processing system to identify, via the processor, borders of a pleural line in each of the plurality of lung ultrasound images based on at least pixel brightness, determine, via the processor, an irregularity score for each location of the pleural line, and identify, via the processor, locations of the pleural line having irregularity scores greater than or equal to a threshold as the irregular pleura;
visually indicating the irregular pleura in each of at least the subset of the plurality of lung ultrasound images on a display in real-time by executing, via the processor of the image processing system, a second set of instructions stored in the non-transitory memory of the image processing system in response to the identifying the irregular pleura; and
upon completion of the protocol, selecting one image having a greatest relative amount of the irregular pleura from the plurality of lung ultrasound images and outputting only the one image to the display by executing, via the processor of the image processing system, a third

Assignees

• Ge Prec Healthcare Llc

Inventors

• Halmann Menachem
• Pinkovich Dani
• Shiran Carmit
• Anderson Robert John
• Fermoso Antonio Fabian
• Owen Cynthia

Classifications

• G06T2207/10132

  Ultrasound image · CPC title

• G06T2207/30061

  Lung · CPC title

• A61B6/5235

  combining images from the same or different ionising radiation imaging techniques, e.g. PET and CT · CPC title

• A61B6/032

  Transmission computed tomography [CT] · CPC title

• A61B8/5261

  combining images from different diagnostic modalities, e.g. ultrasound and X-ray · CPC title