Systems and methods for detecting complex networks in MRI image data

• US 10,034,645 B1
• Filed: 11/21/2017
• Issued: 07/31/2018
• Est. Priority Date: 04/13/2017
• Status: Active Grant

• Current Assignee: Board of Trustees of the Leland Stanford Junior University (Stanford Management Co.)
• Original Assignee: Board of Trustees of the Leland Stanford Junior University (Stanford Management Co.)
• Inventors: Williams, Leanne Maree
• Primary Examiner(s): Seth, Manav
• Application Number:
  US15/820,338
• Time in Patent Office: 252 Days
• CPC Class Codes: A61B 2576/026 for the brain
• CPC Class Codes: A61B 5/055 involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
• CPC Class Codes: A61B 5/165 Evaluating the state of mind, e.g. depression, anxiety
• CPC Class Codes: A61B 5/168 Evaluating attention deficit, hyperactivity
• CPC Class Codes: A61B 5/4064 Evaluating the brain A61B5/031, A61B5/369, A61B5/14553 take precedence
• CPC Class Codes: A61B 5/4088 Diagnosing of monitoring cognitive diseases, e.g. Alzheimer, prion diseases or dementia
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: A61B 6/501 for diagnosis of the head, e.g. neuroimaging or craniography
• CPC Class Codes: G06T 2207/10076 4D tomography; Time-sequential 3D tomography
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/10104 Positron emission tomography [PET]
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/30016 Brain
• CPC Class Codes: G06T 3/18 Image warping, e.g. rearranging pixels individually
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/0014 using an image reference approach
• CPC Class Codes: G06T 7/0016 involving temporal comparison
• CPC Class Codes: G06V 10/26 Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
• CPC Class Codes: G06V 10/34 Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
• CPC Class Codes: G06V 20/64 : Three-dimensional objects
• CPC Class Codes: G06V 20/695 : Preprocessing, e.g. image segmentation
• CPC Class Codes: G06V 2201/031 : of internal organs
• CPC Class Codes: G16H 30/40 : for processing medical images, e.g. editing

Abstract

Systems and methods for detecting complex networks in MRI image data in accordance with embodiments of the invention are illustrated. One embodiment includes an image processing system, including a processor, a display device connected to the processor, an image capture device connected to the processor, and a memory connected to the processor, the memory containing an image processing application, wherein the image processing application directs the processor to obtain a time-series sequence of image data from the image capture device, identify complex networks within the time-series sequence of image data, and provide the identified complex networks using the display device.

Claims

1. A magnetic resonance imaging image processing system, comprising:

at least one processor;

a memory connected to the at least one processor and containing an image processing application;

a display in communication with the at least one processor;

synchronization circuitry;

a magnetic resonance imaging machine in communication with the synchronization circuitry and the at least one processor, wherein the magnetic resonance imaging machine is directed to obtain image data describing at least a patient'"'"'s brain; and

a stimulus device connected to the synchronization circuitry, wherein the stimulus device is configured to provide a stimulus to a patient;

wherein the image processing application directs the processor to;

obtain the image data from the magnetic resonance imaging machine via a network;

generate a time-series sequence of image data of the patient'"'"'s brain, wherein the time-series sequence of image data is time-stamped with the times of stimuli provided by the stimulus device using the synchronization circuitry; and

pre-process the time-series sequence of image data to identify brain regions;

generate at least one neurological model, the at least one neurological model comprising;

a data structure describing at least one network within the time-series sequence of image data, and

metrics specifying neurological activity observable within the time—

series sequence of image data;

assign a biotype to the patient based on the neurological model; and

provide a graphical user interface containing the assigned biotype using the display.

2. The magnetic resonance imaging image processing system of claim 1, wherein preprocessing the time-series sequence of image data further comprises:

realigning the time-series sequence of image data;

unwarping the time-series sequence of image data;

despiking the time-series sequence of image data; and

registering brain structures observable within the time-series sequence of image data and a brain atlas, wherein the brain atlas comprises a data structure stored in the memory and describes physical regions within a reference brain.

3. The magnetic resonance imaging image processing system of claim 2, wherein despiking the time-series sequence of image data comprises measuring displacement between images in the time-series sequence of image data, and identifying time periods in the time-series sequence of image data where the displacement between images is greater than a frame displacement threshold.

4. The magnetic resonance imaging image processing system of claim 2, wherein the atlas is the Automated Anatomical Labeling atlas.

5. The magnetic resonance imaging image processing system of claim 1, wherein at least one of the at least one neurological model is a first level reactivity model, and the metrics specifying the neurological activity specify strength of neuronal activation across brain regions.

6. The magnetic resonance imaging image processing system of claim 5, wherein the image processing application further directs the processor to generate a first level reactivity model, wherein generating the first level reactivity model comprises:

extracting image data describing brain structures from image data describing non-brain structures in the time-series sequence of image data;

removing global signals from the image data using a white noise mask;

spatially smoothing the image data using a blurring kernel; and

determining a degree of neuronal activation in response to stimuli delivered corresponding to the time-stamps in the time-series sequence of image data.

7. The magnetic resonance imaging image processing system of claim 6, wherein the blurring kernel is an 8 millimeter blurring kernel.

8. The magnetic resonance imaging image processing system of claim 1, wherein the at least one of the at least one neurological model is a psychophysiological interaction model, and the metrics specifying the neurological activity specify degree of connectivity between brain regions.

9. The magnetic resonance imaging image processing system of claim 8, wherein the image processing application further directs the processor to generate a psychophysiological interaction model from the pre-processed time-series sequence of image data, wherein generating the psychophysiological interaction model comprises:

performing slice time correction on the pre-processed time-series sequence of image data using a descending interleaved acquisition sequence;

normalizing the pre-processed time-series sequence of image data to a coordinate system;

spatially smoothing the pre-processed time-series sequence of image data using a blurring kernel;

defining at least one volume of interest in the pre-processed time-series sequence of image data using a mask;

extracting an eigenvariate from the at least one volume of interest;

generating a volume of interest data structure comprising;

a deconvolved time course comprising the physiological component of the psychophysiological interaction model;

a psychological variable describing a parametric contrast of task onsets from a contrast of interest; and

data describing the interaction between the physiological and psychological variables.

10. The magnetic resonance imaging image processing system of claim 1, wherein the at least one neurological model comprises a resting state model.

11. The magnetic resonance imaging image processing system of claim 10, wherein generating the resting state model comprises:

concatenating preprocessed image data across stimuli;

segmenting the concatenated image data by tissue types;

generating at least one regressor matrix;

generating residual images from the segmented preprocessed image data; and

bandpass filtering the residual images.

12. The magnetic resonance imaging image processing system of claim 11, wherein the tissue types are at least white matter, grey matter, and cerebrospinal fluid.

13. The magnetic resonance imaging image processing system of claim 11, wherein the at least one resting state model is a region of interest resting state model.

14. The magnetic resonance imaging image processing system of claim 11, wherein the at least one resting state model is a voxel wise resting state model.

15. The magnetic resonance imaging image processing system of claim 14, wherein generating the voxel wise resting state model further comprises:

extracting a time-series sequence of image data corresponding to a particular set of stimuli from the segmented time-series sequence of image data; and

applying regression against the extracted time-series sequence of image data for a particular set of stimuli against the segmented time-series sequence of image data for all voxels, indicating the overall resting state for each voxel.

16. The magnetic resonance imaging image processing system of claim 15, wherein assigning a biotype comprises:

obtaining a database of biotype classifications comprising image data annotated with reactivity and connectivity metrics associated with specific biotypes;

matching metrics from the resting state model to the best fitting biotype classification.

17. The magnetic resonance imaging image processing system of claim 16, wherein the database of biotype classifications is generated using machine learning on a training data set.

18. The magnetic resonance imaging image processing system of claim 1, wherein assigning a biotype comprises generating an indicator of fit describing how closely the biotype matches the patient.

19. The magnetic resonance imaging system of claim 1, wherein the assigned biotype is associated with at least one treatment, and the image processing application further directs the processor to recommend the at least one treatment to the user.

20. The magnetic resonance imaging system of claim 19, wherein the at least one treatment is an at least one drug, and a medical professional treats the patient by prescribing the at least one drug to the patient.

21. The magnetic resonance imaging image processing system of claim 20, wherein the image processing application further directs the processor to generate at least one efficacy metric based on the assigned biotype, wherein the at least one efficacy metric indicates the likelihood of success of treating the patient using the at least one drug.

22. The magnetic resonance imaging image processing system of claim 1, wherein the magnetic resonance imaging machine is connected to a first communications port capable of transmitting data over a network, the processor is connected to a second communications port capable of receiving data over the network;

and the magnetic resonance imaging machine and the processor are connected via the network using the first communications port and the second communications port.

23. The magnetic resonance imaging image processing system of claim 1, wherein the stimulus device provides a Go No-Go test battery and an emotional regulation test battery to the patient.

24. The magnetic resonance imaging image processing system of claim 1, wherein the stimulus device is a visual display device.