Evaluating Quality of Segmentation of an Image into Different Types of Tissue for Planning Treatment Using Tumor Treating Fields (TTFields)
To plan tumor treating fields (TTFields) therapy, a model of a patient's head is often used to determine where to position the transducer arrays during treatment, and the accuracy of this model depends in large part on an accurate segmentation of MRI images. The quality of a segmentation can be improved by presenting the segmentation to a previously-trained machine learning system. The machine learning system generates a quality score for the segmentation. Revisions to the segmentation are accepted, and the machine learning system scores the revised segmentation. The quality scores are used to determine which segmentation provides better results, optionally by running simulations for models that correspond to each segmentation for a plurality of different transducer array layouts.
1 . A method of estimating a quality of a segmentation of an image, the method comprising:
presenting a new image and a segmentation of the new image to a machine learning system, wherein the machine learning system has been trained to estimate a quality of a segmentation of an image based on a plurality of reference images and at least one quality score that has been assigned to each of the reference images, wherein the machine learning system has been trained to estimate the quality of segmentation of an image based on (a) a quality of affine registration and (b) a quality of deformable registration;
receiving, from the machine learning system, at least one first quality score for the segmentation of the new image; and
outputting the at least one first quality score for the segmentation of the new image.
2 . The method of claim 1 , wherein the quality of deformable registration is determined based on a deformation's field bias, directional variability, and mean per-axis variability.
3 . The method of claim 1 , further comprising training the machine learning system to estimate the quality of a segmentation of an image based on the plurality of reference images and at least one quality score that has been assigned to each of the reference images,
wherein the training occurs prior to the presenting of the new image.
4 . A method of estimating a quality of a segmentation of an image, the method comprising:
presenting a new image and a segmentation of the new image to a machine learning system, wherein the machine learning system has been trained to estimate a quality of a segmentation of an image based on a plurality of reference images and at least one quality score that has been assigned to each of the reference images, and wherein the machine learning system has been trained to estimate the quality of segmentation of an image based at least one global quality feature, at least one local quality feature, and a shortest axis length;
receiving, from the machine learning system, at least one first quality score for the segmentation of the new image; and
outputting the at least one first quality score for the segmentation of the new image.
5 . The method of claim 4 , wherein the machine learning system has been trained to estimate the quality of segmentation of an image based at least one global quality feature, at least one local quality feature, and a shortest axis length for intra-cranial tissues.
6 . The method of claim 5 , wherein the machine learning system has been trained to estimate the quality of segmentation of the new image based on image quality and tissue's shape properties for extra-cranial tissues.
7 . The method of claim 4 , further comprising training the machine learning system to estimate the quality of a segmentation of an image based on the plurality of reference images and at least one quality score that has been assigned to each of the reference images,
wherein the training occurs prior to the presenting of the new image.
8 . A method of estimating a quality of a segmentation of an image, the method comprising:
presenting a new image and a segmentation of the new image to a machine learning system, wherein the machine learning system has been trained to estimate a quality of a segmentation of an image based on a plurality of reference images and at least one quality score that has been assigned to each of the reference images;
receiving, from the machine learning system, at least one first quality score for the segmentation of the new image;
outputting the at least one first quality score for the segmentation of the new image;
making an automatic adjustment to the segmentation;
presenting the adjusted segmentation to the machine learning system;
receiving, from the machine learning system, at least one second quality score for the adjusted segmentation; and
outputting an indication when the at least one second quality score indicates an improved quality with respect to the at least one first quality score.
9 . The method of claim 8 , further comprising training the machine learning system to estimate the quality of a segmentation of an image based on the plurality of reference images and at least one quality score that has been assigned to each of the reference images,
wherein the training occurs prior to the presenting of the new image.
10 . The method of claim 8 , wherein the machine learning system has been trained to estimate the quality of segmentation of the new image based on image quality and tissue's shape properties for extra-cranial tissues.