US-12626366-B2 - Automated image segmentation system and method

Abstract

A method of training a learning network to perform image segmentation includes receiving training data from a trusted source and an alternative source; assigning weights to data in the second corpus; dividing the data into batches with samples from the first and second corpuses; and completing training epochs by, for each batch: generating an output for each sample; determining a loss by comparing the generated output with a ground truth for each sample; defining a discrepancy for each second corpus sample as a difference between the sample loss and the average loss for the first corpus, approximated as the average loss for the first corpus from the preceding epoch modified by a change in the average loss from the preceding to the current epoch; identifying weights for the next batch which minimise the weighted sum of discrepancies and network parameters which minimise the average weighted loss.

Inventors

• Jing Qin
• Youyi SONG

Assignees

• THE HONG KONG POLYTECHNIC UNIVERSITY

Dates

Publication Date

20260512

Application Date

20230712

Priority Date

20230526

Claims (15)

1. 1 . A computer-implemented method of training a learning network to perform image segmentation, the method comprising: receiving a first corpus of training data from a trusted source; receiving a second corpus of training data from at least one alternative source; assigning an initial weight to each item of data in the second corpus; dividing the training data into one or more batches, where each batch includes at least one sample from the first corpus and at least one sample from the second corpus; and completing a predefined number of training epochs, wherein each training epoch comprises, for each batch: generating an output for each sample in the batch using the learning network; determining a loss value for each sample in the batch by comparing the generated output with a ground truth for the sample; defining a discrepancy for each second corpus sample in the batch as a difference between the loss value for the sample and the average loss for all first corpus samples, wherein the average loss for all first corpus samples is approximated as the average loss for all first corpus samples from the preceding training epoch modified by a change in the average loss for first corpus samples in the batch from the preceding training epoch to the current training epoch; identifying a set of updated weights for the next batch which minimise the weighted sum of the discrepancies for each second corpus sample; and identifying updated network parameters for the learning network which minimise the average loss for each first corpus sample in the batch and the average weighted loss for each second corpus sample in the batch.
2. 2 . The computer-implemented method of claim 1 , further comprising training the learning network using the first corpus of training data to obtain a first optimum network, wherein network parameters are identified for each batch subject to a first constraint that the average loss for each first corpus sample in the batch must be less than the average loss for each first corpus sample processed by the first optimum network.
3. 3 . The computer-implemented method of claim 2 , wherein the network parameters are identified by gradient descent using the sum of weighted losses for the batch and a first constraint term representing the first constraint.
4. 4 . The computer-implemented method of claim 3 , wherein the first constraint term is determined by: comparing the loss value for each first corpus sample in the batch with the loss value for the sample when processed by the first optimum network; and for the one or more loss values for a first corpus sample in the batch which are greater than the corresponding loss values for the samples when processed by the first optimum network, summing the differences between the compared loss values to generate the first constraint term.
5. 5 . The computer-implemented method of claim 4 , wherein the first constraint term is modified by a penalty coefficient which is initiated at a small value and updated to increase with each batch.
6. 6 . The computer-implemented method of claim 5 , wherein the penalty coefficient is updated to increase by a value based on the first constraint term for the batch.
7. 7 . The computer-implemented method of claim 5 , wherein the penalty coefficient is weighted to reduce gradually after a predetermined number of epochs have been completed.
8. 8 . The computer-implemented method of claim 1 , wherein network parameters are identified for each batch subject to a second constraint that the sum of the weights is non-zero.
9. 9 . The computer-implemented method of claim 1 , wherein the identified set of updated weights minimises the weighted sum of the discrepancies in combination with a factor based on the scalar length of a vector formed by the plurality of weights.
10. 10 . The computer-implemented method of claim 1 , wherein the discrepancy for each second corpus sample is further multiplied by the difference between the loss value for the sample in the current epoch and the loss value for the sample in the previous epoch.
11. 11 . The computer-implemented method of claim 1 , wherein each sample is a computed tomography, CT, scan.
12. 12 . The computer-implemented method of claim 1 , wherein the average loss for all first corpus samples is set as 1 in the first epoch.
13. 13 . The computer-implemented method of claim 1 , further comprising performing image segmentation by: receiving one or more samples of image data; and generating an output of segmented image data for each received sample.
14. 14 . A non-transitory computer-readable medium comprising instructions which, when executed by a processor, cause the processor to execute the method of claim 1 .
15. 15 . An image segmentation network, trained according to the method of claim 1 .

Description

FIELD OF THE DISCLOSURE The present disclosure relates to an image segmentation system and method, especially suitable for analysing medical images. BACKGROUND OF THE DISCLOSURE Image segmentation is a method of dividing a digital image into segments which can be further processed or analysed. Computer assisted volumetric image segmentation of medical images is a particularly challenging type of image segmentation. Typically, computer assisted image segmentation supports the daily work of radiologists who visually analyse anatomical structures in medical images and discern subtle variations in size shape and structure which may be indicative of a disease state of a subject. A typical approach in training machine learning based (deep neural network) image segmentation of medical images lies in the use of a large internal dataset of images (e.g. Computer Tomography or CT images) which have voxel level annotation of target structures performed by domain experts such as radiologists. It would be appreciated that preparation of this internal training data is laborious and time consuming; and may even require multiple rounds of consultation between experts to reach a consensus. Supplementation of such internal data with external data comprising CT images sourced from websites or other institutions is possible. However, this is complicated by variation in image quality and relevance, and difficulty in discerning between useful and harmful data from such images. One solution would be to drop external data of such images with a large loss value during training but this would distort machine learning by favouring easy to learn patterns; ignoring informative data with useful hard patterns that make deep models more accurate and robust. At the same time, it would be appreciated that if external data with a large loss value is not dropped then associated learning may be inappropriately affected by hard patterns from outliers. Various approaches have been utilised to learn external data weights including by using gradient descent detection of mini-batch data; however such approaches are computationally expensive; as they require computation of second order gradients of the network. It is an object of the present disclosure to address or at least partially ameliorate some of the above problems of the current approaches. SUMMARY OF THE DISCLOSURE Features and advantages of the disclosure will be set forth in the description which follows, and in part will be obvious from the description, or can be learned by practice of the herein disclosed principles. The features and advantages of the disclosure can be realized and obtained by means of the instruments and combinations particularly pointed out in the appended claims. In accordance with a first aspect of the present disclosure, there is provided a computer-implemented method of training a learning network to perform image segmentation, the method comprising: receiving a first corpus of training data from a trusted source; receiving a second corpus of training data from at least one alternative source; assigning an initial weight to each item of data in the second corpus; dividing the training data into one or more batches, where each batch includes at least one sample from the first corpus and at least one sample from the second corpus; and completing a predefined number of training epochs, wherein each training epoch comprises, for each batch: generating an output for each sample in the batch using the learning network; determining a loss value for each sample in the batch by comparing the generated output with a ground truth for the sample; defining a discrepancy for each second corpus sample in the batch as a difference between the loss value for the sample and the average loss for all first corpus samples, wherein the average loss for all first corpus samples is approximated as the average loss for all first corpus samples from the preceding training epoch modified by a change in the average loss for first corpus samples in the batch from the preceding training epoch to the current training epoch; identifying a set of updated weights for the next batch which minimise the weighted sum of the discrepancies for each second corpus sample; identifying updated network parameters for the learning network which minimise the average loss for each first corpus sample in the batch and the average weighted loss for each second corpus sample in the batch. The method may include training the learning network using the first corpus of training data to obtain a first optimum network, wherein network parameters are identified for each batch subject to a first constraint that the average loss for each first corpus sample in the batch must be less than the average loss for each first corpus sample processed by the first optimum network. The network parameters may be identified by gradient descent using the sum of weighted losses for the batch and a first constraint term representing the