Semi-automatic segmentation algorithm for pet oncology images

#This invention relates to segmentation of medical images. More particularly, the invention relates to a method and system for segmenting an region of interest in three-dimensional medical images for use in volumetric measurement.

#It is well-known to obtain three-dimensional (3D) arrays of data representing one or more physical properties within an interior of a solid body, for example, anatomical structures. In medical imaging, such data is obtained by a variety of noninvasive methods such as positron emission tomography (PET), computed tomography (CT), magnetic resonance imaging (MRI), ultrasound, x-ray or a combination thereof. Regardless of the image data acquisition method, the 3D array of data typically consists of a plurality of sets of three-dimensional coordinates distributed at regular positions about the body of interest. There are a variety of techniques available to generate a three-dimensional model or structure. Typically, a seed voxel (volume element) is placed within the anatomical structure of interest and adjacent voxels are successively analyzed and identified as belonging to the same structure generally if they are adjacent to a previously identified voxel and they meet a specified attribute, such as intensity or radiological density. In accordance with any of the known techniques, a 3D image is obtained for visualization.

#The three-dimensional (3D) visualization of internal anatomical structures is a technique by medical professionals and research scientists. Three-dimensional models enable the ability to rotate the model or virtual representation of the anatomical structure, as well as adjust a point of perspective and zoom in/out from features of interest. Additionally, volumetric measurements are enabled by a variety of known 3D image processing techniques.

#FIG. 1 is a block diagram view of a medical imaging system for use in a representative embodiment of the present invention;

#FIG. 2 is a diagrammatic view of the positioning of the seed points in relation to the generated spheres in a representative embodiment of the present invention;

#FIGS. 3-5 are diagrammatic views of a representative embodiment of processing used in semi-automatic segmentation of a medical image; and

#Referring to FIG. 1, a block diagram of a system 100 for semi-automatic segmentation of a medical image 135. The system 100 includes an imaging device 110 that can be selected from various medical imaging devices for generating a plurality of medical images 135. In one embodiment, medical imaging techniques, such as, for example, positron emission tomography (PET) and magnetic resonance imaging (MRI) systems, are used to generate the medical images 135.

#In one embodiment, during a PET imaging session, a patient is injected with a radio-labeled ligand that specifically targets metabolically hyperactive regions like tumors. The patient horizontally lies inside the scanner. Photon emissions from the decay of the radio-active ligand are detected by a series of photon detectors. The detectors measure the amount radio-active emissions from inside the patient's body. This information is used to compute the concentration of the radio-labeled ligand for sample points in the body. A gray scale image is then constructed based upon the calculated radio-ligand concentrations. The shades of gray in the image contrast the amount of radio-ligand concentration of every point within the slice. The slices obtained during a PET session can be reconstructed to provide an functionally correct representation of the area of interest within the body.

#In another embodiment, during a MR imaging session, the patient is placed inside a strong magnetic field generated by a large magnet. Magnetized protons within the patient, such as hydrogen atoms, align with the magnetic field produced by the magnet. A particular slice of the patient is exposed to radio waves that create an oscillating magnetic field perpendicular to the main magnetic field. The slices can be taken in any plane chosen by the physician or technician (hereinafter the “operator”) performing the imaging session. The protons in the patient's body first absorb the radio waves and then emit the waves by moving out of alignment with the field. As the protons return to their original state (before excitation), diagnostic images based upon the waves emitted by the patient's body are created. Like PET image slices, MR image slices can be reconstructed to provide an overall picture of the body area of interest. Parts of the body that produce a high signal are displayed as white in an MR image, while those with the lowest signals are displayed as black. Other body parts that have varying signal intensities between high and low are displayed by various shades of gray.