Abstract: A difference image representing intensity differences between a first medical image and a second medical image is generated. A mixture model is fitted to an intensity distribution of the difference image to identify a plurality of probability distributions which collectively model the intensity distribution. A plurality of intensity ranges is determined as a function of the plurality of probability distributions. Image data of the difference image is labeled by determining into which of the plurality of intensity ranges said labeled image data falls. This technique more accurately details changes in medical images than known systems and methods.

Abstract: The invention provides for a medical instrument (100) comprising a processor (134) and a memory (138) containing machine executable instructions (140). Execution of the machine executable instructions causes the processor to: receive (200) a first magnetic resonance image data set (146) descriptive of a first region of interest (122) of a subject (118) and receive (202) at least one second magnetic resonance image data set (152, 152?) descriptive of a second region of interest (124) of the subject. The first region of interest at least partially comprises the second region of interest. Execution of the machine executable instructions further cause the processor to receive (204) an analysis region (126) within both the first region of interest and within the second region of interest.

Abstract: The invention provides for a medical instrument (100) comprising a processor (134) and a memory (138) containing machine executable instructions (140). Execution of the machine executable instructions causes the processor to: receive (200) a first magnetic resonance image data set (146) descriptive of a first region of interest (122) of a subject (118) and receive (202) at least one second magnetic resonance image data set (152, 152?) descriptive of a second region of interest (124) of the subject. The first region of interest at least partially comprises the second region of interest. Execution of the machine executable instructions further cause the processor to receive (204) an analysis region (126) within both the first region of interest and within the second region of interest.

Abstract: A method comprises the step of determining a minimum length of a lesion based on an imaging modality used to capture an image of the lesion and a slice thickness of the image, generating an extent cursor corresponding to the minimum size of the lesion, the extent cursor having a circular shape with a diameter corresponding to the minimum length and displaying the image of the lesion with the extent cursor positioned thereover.

Abstract: A system and method are provided for change detection in medical images. A difference image representing intensity differences between a first medical image and a second medical image is generated. A mixture model is fitted to an intensity distribution of the difference image to identify a plurality of probability distributions which collectively model the intensity distribution. A plurality of intensity ranges is determined as a function of the plurality of probability distributions. Image data of the difference image is labeled by determining into which of the plurality of intensity ranges said labeled image data falls. Accordingly, more accurate change detection is obtained than known systems and method.

Abstract: A radiological imaging apparatus (10) acquires a radiological brain image of a subject after administration of a radio tracer binding to a target substance indicative of a clinical pathology. In one embodiment, the clinical pathology is amyloid deposits in the brain at a level correlative with Alzheimer's disease and the target substance is amyloid deposits. A processor (C) tests for the clinical pathology by: performing non-rigid registration of the brain image with a positive template (32P) indicative of having the clinical pathology and with a negative template (32N) indicative of not having the clinical pathology; generating positive and negative result metrics (36P, 36N) quantifying closeness of the registration with the positive and negative template respectively; and generating a test result (54) based on the positive result metric and the negative result metric. An independent test result is generated by quantifying a second mode of an intensity histogram for the brain image.

Abstract: An imaging work station (20) includes one or more processors programmed to receive (170) an image depicting a distribution of a radiotracer in a brain or other region of interest. The radiotracer includes at least one of [18F]-Flutemetamol, [18F]-Florbetaben, and [18F]-Florbetapir which highlights amyloid deposits. The image and a template or an MRI image of the region of interest which includes a segmented anatomical feature, such as gray matter, are registered (180) to a common space. A volume representation of the image which depicts the distribution of the radiotracer in the segmented gray matter and suppresses the radiotracer outside of the segmented anatomical feature in white matter is extracted (210).

Abstract: A method comprises the step of determining a minimum length of a lesion based on an imaging modality used to capture an image of the lesion and a slice thickness of the image, generating an extent cursor corresponding to the minimum size of the lesion, the extent cursor having a circular shape with a diameter corresponding to the minimum length and displaying the image of the lesion with the extent cursor positioned thereover.

Abstract: An apparatus detects asymmetry in an object, such as a brain. The apparatus includes a processor programmed to fit a three-dimensional image of the object to a preselected shape, such as a standard brain atlas. The processor projects the three-dimensional image of the object to a two-dimensional surface image. The processor compares corresponding mirror image symmetric voxel pairs on the left and right sides of the surface image. The processor generates at least one of an asymmetry map and an asymmetry index based on the deviations in the pixel pairs. The processor can also mask, before the comparison, pixels of the surface image which are asymmetric in a normal brain.

Abstract: Generating a patient image collective (34) includes receiving a plurality of candidate images (20) and associated data. At least one inclusion/exclusion rule (44) is applied to the plurality of candidate images and associated data which results in subsets of candidate images (32). The candidate images are tested which result in at least one quality measure (40). The at least one quality measure (40) and associated candidate images are reviewed. The at least one inclusion/exclusion rule (44) is refined based on the reviewed at least one quality measure (40) by at least one of: adding a rule, modifying a rule, deleting a rule, removing a candidate image; and adding a candidate image. The steps of applying the at least one inclusion/exclusion rule through refining the at least one inclusion/exclusion rule are repeated until an optimized collective of images is generated based on a collective size and the at least one quality measure (40). The generated collective is outputted to a data store (34).

Abstract: The present invention relates to a method of differentially diagnosing different types of dementia. In particular, the method relates to the use of specific SPECT tracers for differentially diagnosing Alzheimer's disease, Lewy-Body Dementia, and Frontotemporal Dementia.

Abstract: Generating a patient image collective (34) includes receiving a plurality of candidate images (20) and associated data. At least one inclusion/exclusion rule (44) is applied to the plurality of candidate images and associated data which results in subsets of candidate images (32). The candidate images are tested which result in at least one quality measure (40). The at least one quality measure (40) and associated candidate images are reviewed. The at least one inclusion/exclusion rule (44) is refined based on the reviewed at least one quality measure (40) by at least one of: adding a rule, modifying a rule, deleting a rule, removing a candidate image; and adding a candidate image. The steps of applying the at least one inclusion/exclusion rule through refining the at least one inclusion/exclusion rule are repeated until an optimized collective of images is generated based on a collective size and the at least one quality measure (40). The generated collective is outputted to a data store (34).

Abstract: An imaging work station (20) includes one or more processors programmed to receive (170) an image depicting a distribution of a radiotracer in a brain or other region of interest. The radiotracer includes at least one of [18F]-Flutemetamol, [18F]-Florbetaben, and [18F]-Florbetapir which highlights amyloid deposits. The image and a template or an MRI image of the region of interest which includes a segmented anatomical feature, such as gray matter, are registered (180) to a common space. A volume representation of the image which depicts the distribution of the radiotracer in the segmented gray matter and suppresses the radiotracer outside of the segmented anatomical feature in white matter is extracted (210).

Abstract: A radiological imaging apparatus (10) acquires a radiological brain image of a subject after administration of a radio tracer binding to a target substance indicative of a clinical pathology. In one embodiment, the clinical pathology is amyloid deposits in the brain at a level correlative with Alzheimer's disease and the target substance is amyloid deposits. A processor (C) tests for the clinical pathology by: performing non-rigid registration of the brain image with a positive template (32P) indicative of having the clinical pathology and with a negative template (32N) indicative of not having the clinical pathology; generating positive and negative result metrics (36P, 36N) quantifying closeness of the registration with the positive and negative template respectively; and generating a test result (54) based on the positive result metric and the negative result metric. An independent test result is generated by quantifying a second mode of an intensity histogram for the brain image.

Abstract: An apparatus detects asymmetry in an object, such as a brain. The apparatus includes a processor programmed to fit a three-dimensional image of the object to a preselected shape, such as a standard brain atlas. The processor projects the three-dimensional image of the object to a two-dimensional surface image. The processor compares corresponding mirror image symmetric voxel pairs on the left and right sides of the surface image. The processor generates at least one of an asymmetry map and an asymmetry index based on the deviations in the pixel pairs. The processor can also mask, before the comparison, pixels of the surface image which are asymmetric in a normal brain.

Abstract: The present invention relates to a method of differentially diagnosing different types of dementia. In particular, the method relates to the use of specific SPECT tracers for differentially diagnosing Alzheimer's disease, Lewy-Body Dementia, and Frontotemporal Dementia.