Imaging segmentation using multi-scale machine learning approach
AssigneeInternational Business Machines Corporation
InventorsMani ABEDINI, Rajib CHAKRAVORTY, Rahil GARNAVI, Munawar HAYAT
Life of this patent
Estimating the term…
Abstract
A robust segmentation technique based on multi-layer classification technique to identify the lesion boundary is described. The inventors have discovered a technique based on training several classifiers such that to classify each pixel as lesion versus normal Each classifier is trained on a specific range of image resolutions. Then, for a new test image, the trained classifiers are applied on the image. Then by fusing the prediction results in pixel level a probability map is generated. In the next step, a thresholding method is applied to convert the probability map to a binary mask, which determines a mole border.
Claims20
A computer-based method for segmenting an image to identify a border using supervised machine learning, the method comprising:
- a segmentation phase including
- a) accessing an image set with at least one image;
- b) selecting a discrete number of image sub-sections of the image within a given set of numbers that have been used for generating, in a training phase, a discrete number of training sub-sections of a training image;
- c) applying a trained classifier to each image sub-section for the discrete number of image sub-sections, the trained classifier representing a probability that the image sub-section is a boundary type;
- d) outputting of classifiers and generating a probability map then overlaying the probability map on other probability maps generated from other classifiers containing the image sub-section;
- e) in response to the overlaid trained classifier being above a settable threshold, identifying the image sub-section as a border segment;
- and
- repeating a through e until each discrete number of sub-sections in the given set of numbers has been generated for each image in the image set.
A system for segmenting an image to identify a border using supervised machine learning, the system comprising:
- a memory;
- a processor communicatively coupled to the memory, where the processor is configured to perform
- a segmentation phase including
- a) accessing an image set with at least one image;
- b) selecting a discrete number of image sub-sections of the image within a given set of numbers that have been used for generating, in a training phase, a discrete number of training sub-sections of a training image;
- c) applying a trained classifier to each image sub-section for the discrete number of image sub-sections, the trained classifier representing a probability that the image sub-section is a boundary type;
- d) outputting of classifiers and generating a probability map then overlaying the probability map on other probability maps generated from other classifiers containing the image sub-section;
- e) in response to the overlaid trained classifier being above a settable threshold, identifying the image sub-section as a border segment;
- and
- repeating a through e until each discrete number of sub-sections in the given set of numbers has been generated for each image in the image set.
A non-transitory computer program product for segmenting an image to identify a border using supervised machine learning comprising a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code configured to perform:
- a segmentation phase including
- a) accessing an image set with at least one image;
- b) selecting a discrete number of image sub-sections of the image within a given set of numbers that have been used for generating, in a training phase, a discrete number of training sub-sections of a training image;
- c) applying a trained classifier to each image sub-section for the discrete number of image sub-sections, the trained classifier representing a probability that the image sub-section is a boundary type;
- d) outputting of classifiers and generating a probability map then overlaying the probability map on other probability maps generated from other classifiers containing the image sub-section;
- e) in response to the overlaid trained classifier being above a settable threshold, identifying the image sub-section as a border segment;
- and
- repeating a through e until each discrete number of sub-sections in the given set of numbers has been generated for each image in the image set.
Dependencies are inferred from the claim text.