Device and method for automatic pneumothorax detection

1. An ultrasound system for scanning a lung of a subject comprising:

a probe configured to obtain a sequence of temporal frames of ultrasound data relating to a scanning region including at least part of a pleural interface of the lung, each temporal frame of ultrasound data at least extending along a range direction of the probe; and

a data analyzer, comprising a processing circuitry, and configured to:

derive one or more than one correlation maps, each correlation maps being derived from a pair of temporal frames of the sequence of temporal frames;

filter the one or more than one correlation maps to obtain one or more than one filtered correlation maps;

derive one or more than one ultrasound data maps, each ultrasound map being derived from one temporal frame or a compounded version of more than one temporal frames of the sequence of temporal frames;

filter the one or more than one ultrasound data maps to obtain one or more than one filtered ultrasound data maps;

identify a first boundary line comprising a point from each line of a plurality of scanning lines, at which point a value change along the scanning line exceeds a first predetermined threshold in the one filtered correlation map or a compounded version of the more than one filtered correlation maps;

identify a starting point as a point at which there is a value increase along the first boundary line in the one filtered ultrasound data map or the compounded version of the more than one filtered ultrasound data maps, which value increase exceeds a second predetermined threshold;

identify an end point as a point at which there is a value decrease along the first boundary line in the one filtered ultrasound data map or the compounded version of the more than one filtered ultrasound data maps, which value decrease exceeds a third predetermined threshold.

2. The system of claim 1 , wherein deriving one or more than one correlation maps comprises deriving one or more than one cross correlation maps.

3. The system of claim 1 , wherein the data analyzer is further configured to:

determine motion of the identified first boundary, based on a normalized correlation function of the sequence of the temporal frames.

4. The system of claim 1 , wherein the data analyzer is further configured to:

determine motion of the identified first boundary, based on a normalized cross correlation function of the sequence of the temporal frames.

5. The system of claim 4 , wherein to determine motion of the identified first boundary, the data analyzer is further configured to:

determine a first region covering the first boundary; and

determine motion of the first region as motion of the identified first boundary.

6. The system of claim 5 , wherein, to determine motion of the identified first boundary, the data analyzer is further configured to:

determine a second region containing only background soft tissue which is closer to the probe than the first boundary; and

determine the relative motion between the first and the second region as the motion of the identified first boundary.

7. The system of claim 1 , wherein the data analyzer is further configured to:

identify, in the one filtered correlation map or a compounded version of the more than one filtered correlation maps, a second boundary at which a value change in coefficients of the corresponding map along a direction orthogonal to the range direction exceeds a fourth predetermined threshold.

8. The system of claim 7 , wherein the data analyzer is further configured to determine a lung point, based on the second boundary if the second boundary is identified.

9. The system of claim 1 , further comprising a display, configured to present the ultrasound data and processed data.

10. A method of scanning a lung of a subject using ultrasound, comprising:

obtaining a sequence of temporal frames of ultrasound data relating to a scanning region including at least part of a pleural interface of the lung, each temporal frame of ultrasound data at least extending along a range direction of the probe;

deriving one or more than one correlation maps, each from a pair of temporal frames of the sequence of temporal frames;

filtering the one or more than one correlation maps to obtain one or more than one filtered correlation maps;

deriving one or more than one ultrasound data maps, each ultrasound map being derived from one temporal frame or a compounded version of more than one temporal frames of the sequence of temporal frames;

filtering the one or more than one ultrasound data maps to obtain one or more than one filtered ultrasound data maps;

identifying a first boundary line comprising a point from each line of a plurality of scanning lines, at which point a value change along the scanning line exceeds a first predetermined threshold in the one filtered correlation map or a compounded version of the more than one filtered correlation maps;

identify a starting point as a point at which there is a value increase along the first boundary line in the one filtered ultrasound data map or the compounded version of the more than one filtered ultrasound data maps, which value increase exceeds a second predetermined threshold; and

identifying an end point as a point at which there is a value decrease along the first boundary line in the one filtered ultrasound data map or the compounded version of the more than one filtered ultrasound data maps, which value decrease exceeds a third predetermined threshold.

11. The method of claim 10 , further comprising:

determining motion of the identified first boundary, based on a normalized correlation function derived from the sequence of the temporal frames.

12. The method of claim 11 , wherein determining motion of the identified first boundary comprises:

determining a first region covering the first boundary; and

determining motion of the first region as the motion of the first boundary.

13. The method of claim 12 , wherein determining motion of the identified first boundary comprises:

determining a second region only containing background soft tissue which is closer to the probe than the first boundary;

determining the relative motion between the first and the second region as the motion of the first boundary.

14. The method of claim 13 , further comprising:

detecting, in the one filtered correlation map or a compounded version of the more than one filtered correlation maps, a second boundary at which a value change along a direction orthogonal to the range direction exceeds a fourth predetermined threshold.

15. An ultrasound system for scanning a lung of a subject, comprising:

a probe configured to obtain a sequence of temporal frames of ultrasound data relating to a region including at least part of a pleural interface of the lung, each temporal frame of ultrasound data at least extending along a range direction of the probe; and

a data analyzer, comprising:

a map deriver configured to derive one or more than one correlation maps, each correlation map being derived from a pair of temporal frames of the sequence of temporal frames, and to derive one or more than one ultrasound data maps, each ultrasound map being derived from one temporal frame or a compounded version of more than one temporal frames of the sequence of temporal frames;

a map filter configured to filter the one or more than one correlation maps to obtain one or more than one filtered correlation maps, and to filter the one or more than one ultrasound data maps to obtain one or more than one filtered ultrasound data maps;

a boundary identifier configured to identify a first boundary line comprising a point from each line of a plurality of scanning lines, at which point a value change along the scanning line exceeds a first predetermined threshold in the one filtered correlation map or a compounded version of the more than one filtered correlation maps, to identify a starting point as a point at which there is a value increase along the first boundary line in the one filtered ultrasound data map or the compounded version of the more than one filtered ultrasound data maps, which value increase exceeds a second predetermined threshold, and to identify an end point as a point at which there is a value decrease along the first boundary line in the one filtered ultrasound data map or the compounded version of the more than one filtered ultrasound data maps, which value decrease exceeds a third predetermined threshold.