METHOD FOR SEGMENTING BRAIN TUMOURS

Abstract

ABSTRACT A method (100) for segmenting brain tumours, the method (100) comprising the steps of receiving a set of three dimensional (3D) MRI (Magnetic Resonance Imaging) images of brain scans with tumour annotations, pre-processing the 3D MRI images to obtain normalised 3D MRI images, applying a convolutional neural network (CNN) model over the normalised 3D MRI images for feature extraction and optimization of the normalised 3D MRI images, generating segmentation masks that delineate tumour regions in the normalised 3D MRI images based on predictions of the CNN model and determining a segmentation accuracy of the delineated tumour regions based on the segmentation masks and one or more metrics. <>

Specification

Description:FIELD OF THE DISCLOSURE
[0001] This invention generally relates to a field of brain tuomors and in particular relates to a method for segmenting brain tumours.
BACKGROUND

[0002] The subject matter discussed in the background section should not be assumed to be prior art merely as a result of its mention in the background section. Similarly, a problem mentioned in the background section or associated with the subject matter of the background section should not be assumed to have been previously recognized in the prior art. The subject matter in the background section merely represents different approaches, which in and of themselves may also correspond to implementations of the claimed technology.
[0003] Brain tumour detection is a critical area in medical diagnostics, as early and accurate identification of abnormal tissue growth within the brain can significantly impact treatment outcomes and patient survival rates. Traditionally, brain tumour detection relies on manual inspection of MRI (Magnetic Resona , Claims:1. A method (100) for segmenting brain tumours, the method (100) comprising the steps of:
receiving a set of three dimensional (3D) MRI (Magnetic Resonance Imaging) images of brain scans with tumour annotations;
pre-processing the 3D MRI images to obtain normalised 3D MRI images;
applying a convolutional neural network (CNN) model over the normalised 3D MRI images for feature extraction and optimization of the normalised 3D MRI images;
generating segmentation masks that delineate tumour regions in the normalised 3D MRI images based on predictions of the CNN model; and
determining a segmentation accuracy of the delineated tumour regions based on the segmentation masks and one or more metrics.

2. The method (100) as claimed in claim 1, further comprising removing noise, normalising intensity values and resizing the 3D MRI images during the pre-processing.

3. The method (100) as claimed in claim 1, wherein the one or more metrics include Dice Similarity Coefficient, sensitivity, and specificity of the de

Documents