Method and system for selecting a region of interest in an image

1. A computer-implemented method for selecting one or more regions of interest (ROIs) in an image, the method comprising:

identifying one or more objects of interest that have been segmented from the image;

identifying predefined landmarks of the objects;

determining reference morphometries pertaining to the objects by performing morphometrics on the objects by reference to the landmarks;

selecting one or more ROIs from the objects according to the reference morphometries, comprising identifying respective locations of the ROIs relative to the reference morphometries; and

outputting the selected one or more ROIs;

wherein determining the reference morphometries comprises determining a shape, a size, or a shape and a size, of the one or more objects.

2. A method as claimed in claim 1 , wherein performing the reference morphometrics comprises

(a) measuring basic morphometries by performing basic morphometrics on the objects by reference to the detected landmarks and determining the reference morphometries based on the measured basic morphometries, or

(b) measuring basic morphometries by performing basic morphometrics on the objects by reference to the detected landmarks and determining the reference morphometries based on the measured basic morphometries by employing one or more trained deep learning reference morphometric models.

3. A method as claimed in claim 1 , wherein the selecting of the one or more ROIs from the objects according to the reference morphometries further comprises determining the shape and size of the ROIs.

4. A method as claimed in claim 1 , further comprising segmenting the objects from the image.

5. A method as claimed in claim 4 , wherein segmenting the objects

(a) comprises generating masks of the respective segmented objects; and/or

(b) employs one or more trained machine learning segmentation models.

6. A method as claimed in claim 1 , further comprising preparing the image for segmentation by pre-processing the image, wherein pre-processing comprises applying a Gaussian blur function to the image to reduce noise, and/or applying histogram equalization to enhance contrast of the image.

7. A method as claimed in claim 1 , further comprising normalizing the objects from an image coordinate system into an anatomical coordinate system, by coordinate transformation or otherwise, before performing the reference morphometrics on the objects.

8. A method as claimed in claim 1 , further comprising determining density of one or more of the selected ROIs using attenuation of a referent material.

9. A computer program stored on a non-transitory computer-readable medium, the computer program comprising program code configured to, when executed by one or more processors, implement the method for selecting a region of interest (ROI) in an image of claim 1 .

10. A non-transitory computer-readable medium, comprising a computer program that comprises program code configured to, when executed by one or more processors, implement the method for selecting a region of interest (ROI) in an image of claim 1 .

11. A system for selecting one or more regions of interest (ROIs) in an image, the system comprising:

an object and landmark identifier configured to identify one or more objects of interest that have been segmented from the image, and to identify predefined landmarks of the objects;

a morphometer configured to determine reference morphometries pertaining to the objects by performing morphometrics on the objects by reference to the landmarks;

a region selector configured to select one or more ROIs from the objects according to the reference morphometries, comprising identifying respective locations of the ROIs and determining respective shapes and sizes of the ROIs relative to the reference morphometries; and

a results output configured to output the selected ROIs;

wherein determining the reference morphometries comprises determining a shape, a size, or a shape and a size, of the one or more objects.

12. A system as claimed in claim 11 , wherein the morphometer is configured to perform the reference morphometrics by

(a) measuring basic morphometries by performing basic morphometrics on the objects by reference to the detected landmarks and determining the reference morphometries based on the measured basic morphometries; or

(b) measuring basic morphometries by performing basic morphometrics on the objects by reference to the detected landmarks and determining the reference morphometries based on the measured basic morphometries by employing one or more trained deep learning reference morphometric models.

13. A system as claimed in claim 11 , wherein the region selector is further configured to determine the shape and size of the ROIs relative to the reference morphometries.

14. A system as claimed in claim 11 , further comprising a segmenter configured to segment the objects from the image.

15. A system as claimed in claim 14 , wherein the segmenter is further configured

(a) to generate masks of the respective segmented objects; and/or

(b) to employ one or more trained machine learning segmentation models to segment the objects.

16. A system as claimed in claim 11 , further comprising an image pre-processor configured to prepare the image for segmentation

(a) by pre-processing the image to reduce noise by applying a Gaussian blur function to the image; and/or

(b) by pre-processing the image to enhance contrast of the image by applying histogram equalization to the image.

17. A system as claimed in claim 11 , further comprising a normalizer configured to normalize the objects from an image coordinate system into an anatomical coordinate system, by coordinate transformation or otherwise, before reference morphometrics are performed on the objects.

18. A system as claimed in claim 11 , further comprising a density determiner configured to determine density of one or more of the selected ROIs using attenuation of a referent material.