A method of, and apparatus for, improved estimation of fetal characteristics

Abstract

Computer-implemented method of estimating fetal weight using a machine learning model (MLM) 100. Ultrasound images 106 comprise image data representative of the anatomical structure and appearance of

Claims

1. A computer-implemented method of estimating fetal weight, the method comprising the steps of:a) providing a plurality of ultrasound images comprising image data representative of the anatomical structure and appearance of a head, an abdomen and a femur of a fetus in utero;b) utilising the image data in a first machine learning model comprising one or more neural networks to generate one or more latent space representations of the anatomical structure and appearance of the fetus and one or more predicted measurements of the anatomical structure of the fetus; andc) utilising the one or more predicted measurements and the one or more latent space representations in a second machine learning model to generate an estimated weight of the fetus.

2. A computer-implemented method according to claim 1, wherein the generated one or more predicted measurements corresponding to the anatomical structure of the fetus comprise one or more of: head circumference (HC); biparietal diameter (BPD); abdominal circumference (AC); and femur length (FL).

3. A computer-implemented method according to claim 1 or 2, wherein the plurality of ultrasound images of the fetus comprise:a head image of the fetus;an abdomen image of the fetus; and a femur image.

4. A computer-implemented method according to claim 3, wherein the head image, the abdomen image and femur image are taken in or close to respective standard planes.

5. A computer-implemented method according to claim 3 or 4, wherein the first machine learning model comprises a plurality of sub-networks, a discrete sub-network being allocated for each of the head, abdomen and femur images.

6. A computer-implemented method according to claim 5, wherein each sub-network comprises a first neural network for generating a respective latent space representation of anatomical structure and appearance and a second neural network for generating, from therespective latent space representation, one or more predicted measurements of the anatomical structure of the fetus.

7. A computer-implemented method according to claim 6, wherein the first neural network comprises a convolutional neural network (CNN).

8. A computer-implemented method according to claim 6 or 7, wherein the second neural network comprises a dense neural network (DNN).

9. A computer-implemented method according to any one of claims 5 to 8, wherein the sub-network allocated to the head image is operable to generate a latent space representation corresponding to the anatomical structure and appearance of the head of the fetus and a predicted measurement of the head circumference (HC) and / or the biparietal diameter (BPD) of the fetus; and / orwherein the sub-network allocated to the abdominal image is operable to generate a latent space representation corresponding to the anatomical structure and appearance of the abdomen of the fetus and a predicted measurement of the abdominal circumference (AC) of the fetus; and / orwherein the sub-network allocated to the femur image is operable to generate a latent space representation corresponding to the anatomical structure and appearance of the femur of the fetus and a predicted measurement of the femur length (FL) of the fetus.

10. A computer-implemented method according to any one of claims 3 to 8, wherein the second machine learning model comprises a dense neural network (DNN).

11. A computer-implemented method according to any one of the preceding claims, further comprising:d) generating an uncertainty estimate for the estimated fetal weight.

12. A computer-implemented method according to claim 11, wherein step d) further comprises:e) applying one or more transforms to the plurality of ultrasound images to generate a set of augmented images;f) performing steps b) and c) using the set of augmented images to generate an estimate of fetal weight based on the set of augmented images.

13. A computer-implemented method according to claim 12, wherein the steps e) and f) are repeated N times to generate N estimates of fetal weight.

14. A computer-implemented method according to claim 12 or 13, wherein the one or more transforms are randomly selected from one or more of: rotation; shear; translation; brightness; contrast; and horizontal flip.

15. A computer-implemented method according to any one of the preceding claims, wherein step a) comprises:g) generating a plurality of ultrasound images of the fetus in utero using an ultrasound imaging apparatus.

16. A computer-implemented method according to any one of the preceding claims, wherein the or each latent space representation of the anatomical structure and appearance of the fetus comprises an embedding vector.

17. An ensemble machine learning model for estimating fetal weight, the ensemble machine learning model comprising:a first machine learning model comprising one or more neural networks, the first machine learning model being configured to utilise a plurality of ultrasound images comprising image data representative of the anatomical structure of a head, an abdomen and a femur of a fetus in utero to generate one or more latent space representations of the anatomical structure and appearance of the fetus and one or more predicted measurements of the anatomical structure of the fetus; anda second machine learning model configured to utilise the one or more predicted measurements and the one or more latent space representations of the anatomical structure to generate an estimated weight of the fetus.

18. An ensemble machine learning model according to claim 17, wherein the plurality of ultrasound images of the fetus comprise: a head image of the fetus; an abdomen image of the fetus; and a femur image, andwherein the first machine learning model comprises a plurality of sub-networks, a discrete sub-network being allocated for each of the head, abdomen and femur images.

19. An ensemble machine learning model according to claim 18, wherein each subnetwork comprises a first neural network for generating a respective latent space representation of anatomical structure and appearance and a second neural network for generating, from the respective latent space representation, one or more predicted measurements of the anatomical structure of the fetus.

20. An ensemble machine learning model according to claim 19, wherein the first neural network comprises a convolutional neural network (CNN) and / or the second neural network comprises a dense neural network (DNN) and / or the second machine learning model comprises a dense neural network (DNN).

21. An ensemble machine learning model according to claim 19 or 20, wherein the subnetwork allocated to the head image is operable to generate a latent space representation corresponding to the anatomical structure and appearance of the head of the fetus and a predicted measurement of the head circumference (HC) and / or the biparietal diameter (BPD) of the fetus; and / orwherein the sub-network allocated to the abdominal image is operable to generate a latent space representation corresponding to the anatomical structure and appearance of the abdomen of the fetus and a predicted measurement of the abdominal circumference (AC) of the fetus; and / orwherein the sub-network allocated to the femur image is operable to generate a latent space representation corresponding to the anatomical structure and appearance of the femur of the fetus and a predicted measurement of the femur length (FL) of the fetus.

22. An ensemble machine learning model according to any one of claims 17 to 21, wherein the or each latent space representation of the anatomical structure and appearance of the fetus comprises an embedding vector.

23. A computing system for estimating fetal weight, the computing system comprising:at least one hardware processor; andan analyser, the analyser comprising:a first machine learning model comprising one or more neural networks, the first machine learning model being configured to utilise a plurality of ultrasound images comprisingimage data representative of the anatomical structure of a head, an abdomen and a femur of a fetus in utero to generate one or one or more latent space representations of the anatomical structure and appearance of the fetus and one or more predicted measurements of the anatomical structure of the fetus; anda second machine learning model configured to utilise the one or more predicted measurements and the one or more embedding vectors to generate an estimated weight of the fetus.

24. A computing system according to claim 23, wherein the generated one or more predicted measurements corresponding to the anatomical structure of the fetus comprise one or more of: head circumference (HC); biparietal diameter (BPD); abdominal circumference (AC); and femur length (FL).

25. A computing system according to claim 23 or 24, wherein the plurality of ultrasound images of the fetus comprise:a head image of the fetus;an abdomen image of the fetus; and a femur image.

26. A computing system according to claim 25, wherein the first machine learning model comprises a plurality of sub-networks, a discrete sub-network being allocated for each of the head, abdomen and femur images.

27. A computing system according to claim 26, wherein each sub-network comprises a first neural network for generating a respective latent space representation of anatomical structure and appearance and a second neural network for generating, from the respective latent space representation, one or more predicted measurements of the anatomical structure of the fetus.

28. A computing system according to claim 27, wherein the first neural network comprises a convolutional neural network (CNN) and / or the second neural network comprises a dense neural network (DNN) and / or the second machine learning model comprises a dense neural network (DNN).

29. A computing system according to claim 26, 27 or 28, wherein the sub-network allocated to the head image is operable to generate a latent space representationcorresponding to the anatomical structure and appearance of the head of the fetus and a predicted measurement of the head circumference (HC) and / or the biparietal diameter (BPD) of the fetus; and / orwherein the sub-network allocated to the abdominal image is operable to generate a latent space representation corresponding to the anatomical structure and appearance of the abdomen of the fetus and a predicted measurement of the abdominal circumference (AC) of the fetus; and / orwherein the sub-network allocated to the femur image is operable to generate a latent space representation corresponding to the anatomical structure and appearance of the femur of the fetus and a predicted measurement of the femur length (FL) of the fetus.

30. A computing system according to any one of claims 23 to 29, wherein the analyser is further configured to generate an uncertainty estimate for the estimated fetal weight.

31. A computing system according to claim 30, wherein the analyser is configured to generate an uncertainty estimate for the estimated fetal weight by:applying one or more transforms to the plurality of ultrasound images to generate a set of augmented images;utilising the set of augmented images to generate an estimate of fetal weight based on the set of augmented images; andrepeating the applying and utilising N times to generate N estimates of fetal weight.

32. A computing system according to claim 31, wherein the one or more transforms are randomly selected from one or more of: rotation; shear; translation; brightness; contrast; and horizontal flip.

33. A computing system according to any one of claims 23 to 32, wherein the or each latent space representation of the anatomical structure and appearance of the fetus comprises an embedding vector.

34. An ultrasound imaging apparatus comprising an ultrasound scanner configured to generate a plurality of ultrasound images of the fetus in utero and the computing system according to any one of claims 23 to 33.

Images