Determination of subject-specific hemodynamic response functions

Abstract

In a medical system 100, 300, execution of machine executable instructions 120 includes receiving (200) a time series 122 of R2 star maps for a brain volume 500, receiving (202) a stimulation signal 124 describing the occurrence of a sensory stimulus, receiving (204) a selection of one or more seed voxels 126 identified in the time series of R2 star maps, calculating (206) a denoised time series 128 of the R2 star maps, and calculating (207) a denoised time series 128 of the R2 star maps. Disclosed herein is a medical system 100, 300 that has a computational system 104 compute 208 a correlation map 130 between a denoised time series of taps, determine 210 activation regions 132 of the brain volume using voxels identified in the correlation map, provide 212 a hemodynamic response 134 function for each voxel and each occurrence of a sensory stimulus, and provide 214 a subject-specific hemodynamic response function 136 by averaging the hemodynamic response functions.

Description

[Technical Field]

[0001] The present invention relates to functional magnetic resonance imaging, and in particular to hemodynamic response functions. [Background technology]

[0002] A large static magnetic field is used by magnetic resonance imaging (MRI) scanners to align the nuclear spins of atoms as part of a procedure to generate images of the inside of a subject. This large static magnetic field is called the B0 field or main magnetic field. MRI can be used to spatially measure various quantities or properties of a subject. Various imaging protocols can be implemented by using pulse sequences to control the collection of magnetic resonance data, and such imaging protocols can be used to measure various properties of a subject.

[0003] For example, functional magnetic resonance imaging (fMRI) uses magnetic resonance imaging to measure brain activity. A common type of functional magnetic resonance imaging is blood-oxygen-level dependent (BOLD) contrast. BOLD imaging relies on the properties of oxygenated and deoxygenated hemoglobin. Oxygenated hemoglobin is paramagnetic, and deoxygenated hemoglobin is diamagnetic. Therefore, T2-star weighted pulse sequences can detect changes in blood oxygenation in the brain. The BOLD signal is characterized by a hemodynamic response function.

[0004] U.S. Patent Publication No. 9,116,219 (B1) discloses a system and method for high-speed functional magnetic resonance imaging using multi-slab echo volume imaging (EVI), in particular a combination of multi-slab excitation and single-shot 3D encoding with parallel imaging, to reduce geometric image distortion and blurring and increase blood oxygenation level-dependent (BOLD) sensitivity compared to conventional echo-planar imaging (EPI). Summary of the Invention

[0005] The present invention provides a medical system, a computer program and a method in the independent claims. Embodiments are given in the dependent claims.

[0006] A drawback in performing functional magnetic resonance imaging (fMRI) is that current clinical techniques assume a constant hemodynamic response function. The embodiments disclosed herein provide a means for quickly and accurately determining a subject-specific hemodynamic response function. A subject-specific hemodynamic response function has several benefits. First, the subject-specific hemodynamic response function can be used to improve the accuracy of subsequent fMRI studies on the same subject. Another advantage is that, as shown herein, hemodynamic response functions vary from individual to individual, making them useful as a diagnostic test.

[0007] In one aspect, the present invention provides a medical system including a memory storing machine-executable instructions and a computing system. The computing system can be implemented in different configurations. For example, in one example, the computing system is available over the Internet or other network system, e.g., as a cloud-based system providing processing or image processing services. In another example, the computing system can be a workstation or other computer system used by a radiologist or other medical professional. In yet another example, the computing system can be a control system for a magnetic resonance imaging system, or can be integrated into or be a module of a control system for a magnetic resonance imaging system.

[0008] Execution of the machine-executable instructions causes the computing system to receive a time series of R2 star maps for a brain volume of the subject. The time series of R2 star maps is a mapping of R2 star values ​​within the brain volume of the subject. The maps may be provided as a time series or as a series of these maps presented sequentially in time. Execution of the machine-executable instructions also causes the computing system to receive a stimulus signal describing the occurrence of sensory stimuli repeatedly presented to the subject. The stimulus signal is synchronized with the time series of R2 star maps. A sensory stimulus is a stimulus presented to a subject that activates the subject. Because sensory stimuli are passive stimuli to which a subject mentally responds, this is useful for determining R2 star values. For example, in a functional magnetic resonance imaging study, a subject may be tasked with performing a mental task. However, a sensory stimulus may be presented to a subject without mental effort or thought, such as an auditory or visual signal.

[0009] Execution of the machine-executable instructions further causes the computational system to receive a selection of one or more seed voxels identified in the time series of R2 star maps. The one or more seed voxels are, for example, portions of a brain volume identified as being activated by a sensory stimulus. The selection may be performed, for example, by an automated system or by operator selection. Execution of the machine-executable instructions further causes the computational system to calculate a denoised time series of R2 star maps by applying a temporal filter algorithm to the time series of R2 star maps. For example, the R2 star values ​​in a particular voxel can be viewed as a function of time. This time series can be denoised using a temporal filter. A digital filtering algorithm, or something as simple as fitting a spline to the R2 star values ​​for a particular voxel, can be applied.

[0010] Execution of the machine-executable instructions further causes the computing system to compute a correlation map for each voxel of the one or more seed voxels by calculating a pixel-by-pixel calculation between each voxel of the one or more seed voxels and the denoised time series of the R2 star map. That is, for each of the one or more seed voxels, a correlation map is calculated between that voxel and all other voxels in the R2 star map. This is accomplished, for example, by calculating a correlation coefficient. For each particular voxel, there is a time series of R2 star values. A correlation coefficient is calculated for the time signal of the R2 star values ​​for each voxel with each voxel of the one or more seed voxels.

[0011] Execution of the machine-executable instructions further causes the computing system to determine activated regions of the brain volume by combining voxels identified in the correlation map for each of the voxels of one or more seed voxels that exceed a predetermined threshold. For example, for each of one or more seed voxels, there is a correlation map between that seed voxel and other voxels in the brain volume. Each of these correlation maps can be thresholded and then combined. This ensures that areas in the brain activated by sensory stimuli are not ignored.

[0012] Execution of the machine-executable instructions further causes the computational system to provide a hemodynamic response function for each voxel and each occurrence of a sensory stimulus in an activated region of the brain volume by aligning the time series of the R2 star map with the stimulation signal. The activated region is a portion of the brain volume that has been shown to correlate with one or more seed voxels. The voxels in the activated region are then used to calculate the hemodynamic response function. The stimulation signal indicates whenever a sensory stimulus is presented to the subject. This is used to take the time series of the R2 star map for each voxel in the activated region and align these individual response functions together.

[0013] Execution of the machine-executable instructions further causes the computational system to yield a subject-specific hemodynamic response function by averaging the hemodynamic response functions for each voxel at each occurrence of sensory stimulation in the activated region of the brain volume. For each voxel, each occurrence of sensory stimulation generates data describing the hemodynamic response function. The subject-specific hemodynamic response function is calculated by collecting all the hemodynamic response functions for each occurrence of sensory stimulation for each voxel in the activated region and then averaging the hemodynamic response functions.

[0014] This embodiment has the advantage of providing a subject-specific hemodynamic response function that can be used to evaluate the subject. The subject-specific hemodynamic response function can vary from person to person and depending on the person's health status. Having a subject-specific hemodynamic response function is also extremely useful because it allows other functional magnetic resonance imaging studies to be performed more accurately. Generally, during functional magnetic resonance imaging, it is assumed that the hemodynamic response function has a specific value. As will be demonstrated later, this is not the case and there can be a great deal of variability between individuals. This variability can be used to study or provide information that is useful to physicians, as well as to provide more accurate functional magnetic resonance imaging in other studies, as previously mentioned.

[0015] In another embodiment, execution of the machine-executable instructions further causes the computing system to calculate, for each voxel and for each occurrence of the sensory stimulus in the activated region of the brain volume, a time of maximum value of each hemodynamic response function. Execution of the machine-executable instructions further causes the computing system to calculate statistical properties of the time of maximum value of each hemodynamic response function for each voxel and for each occurrence of the sensory stimulus in the activated region of the brain volume.

[0016] Execution of the machine-executable instructions further causes the computing system to exclude any hemodynamic response function from the calculation of the subject-specific hemodynamic response function if that hemodynamic response function fails to meet a predetermined criterion determined using the statistical properties. This is used, for example, to exclude abnormal data. For example, a person may move, or part of the data may be unavailable or irrelevant. The statistical properties may be, for example, a mean value, or a certain number of standard deviations, or a window from a histogram. This provides a convenient means for improving the quality of the estimate of the subject-specific hemodynamic response function.

[0017] In another embodiment, the time of maximum value of each hemodynamic response function for each voxel at each occurrence of sensory stimulation in activated regions of the brain volume is calculated using a smoothing function. For example, a digital filter or curve is fitted and used as the smoothing function. For example, a spline is fitted to each individual hemodynamic response function to provide a better estimate of the maximum value.

[0018] In another embodiment, the machine-executable instructions further cause the computing system to receive multiple acquisitions of EPI multi-echo T2 star-weighted k-space data describing a brain volume of the subject, the multiple acquisitions being synchronized to the stimulation signal. Execution of the machine-executable instructions further causes the computing system to receive T1-weighted k-space data describing the brain volume of the subject. Execution of the machine-executable instructions further causes the computing system to reconstruct a T1-weighted image of the brain volume from the T1-weighted k-space data. Execution of the machine-executable instructions further causes the computing system to reconstruct a T2 star-weighted image for each echo of the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data.

[0019] Execution of the machine-executable instructions further causes the computing system to calculate an aligned T2 star weighted image for each echo of the multiple acquisitions of EPI multi-echo T2 star weighted k-space data by performing pre-processing to align the T2 star weighted image for each echo of the multiple acquisitions of EPI multi-echo T2 star weighted k-space data with the T1 weighted image of the brain volume. For example, a subject may move during the course of acquiring the T2 star weighted k-space data. The T1 weighted k-space data of the brain volume and one of the resulting T1 weighted image or the T2 star weighted image are used as a reference to correct for this movement.

[0020] Execution of the machine-executable instructions further causes the computing system to calculate a time series of R2 star maps for the subject's brain volume for each voxel by fitting a curve to the aligned T2 star weighted images for each echo of the multiple acquisitions of EPI multi-echo T2 star weighted k-space data. The data for the multiple echoes is used to calculate a decay rate for each voxel, thereby calculating an R2 star value. This is advantageous because it provides a highly accurate estimate of the R2 star value for a particular voxel.

[0021] In another embodiment, calculating the aligned T2 star weighted image for each echo of the multiple acquisitions of EPI multi-echo T2 star weighted k-space data is performed by pre-processing to align the T2 star weighted image for each echo of the multiple acquisitions of EPI multi-echo T2 star weighted k-space data with a T1 weighted image of the brain volume, the pre-processing comprising: a first step of co-registering a selected image corresponding to a first echo of a selected acquisition of the multiple acquisitions of EPI multi-echo T2 star weighted k-space data with the T1 weighted image; and a second step of segmenting the T1 weighted image to generate gray matter segmentation, white matter segmentation, and cerebrospinal fluid segmentation.

[0022] In the next step, using co-registration between the selected image and the T1-weighted image, the T1-weighted image, as well as the gray matter segmentation, white matter segmentation, and cerebrospinal fluid segmentation, are re-sliced ​​to match the selected image. In this way, all data sets are spatially aligned. In the next step, a brain mask is constructed using the gray matter segmentation, white matter segmentation, and cerebrospinal fluid segmentation. In the final step, the T2 star-weighted images for each echo of the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data are re-aligned with the corresponding images of the selected image. The brain mask is used to determine regions to be ignored in the analysis.

[0023] In another embodiment, the EPI multi-echo T2 star weighted k-space data describing the subject's brain volume includes k-space data for three echoes. The use of three echoes is beneficial because it results in an accurate determination of the T2 star or R2 star value. Using only two echoes may result in an inaccurate measurement of the T2 star value. Using four echoes may not significantly improve the estimate but may take significantly longer.

[0024] In another embodiment, the memory further stores EPI multi-echo pulse sequence commands and T1-weighted pulse sequence commands. The medical system further comprises a magnetic resonance imaging system. The medical system further comprises a stimulation system for providing sensory stimulation to the subject. This may be, for example, a display or projector that provides visual stimulation to the subject. In another example, the stimulation system is provided by an auditory or speaker or headphone system. In yet another example, the sensory stimulation is a tactile stimulation.

[0025] Execution of the machine-executable instructions further causes the computing system to acquire T1-weighted k-space data by controlling the magnetic resonance imaging system with T1-weighted pulse sequence commands. Execution of the machine-executable instructions further causes the computing system to acquire multiple acquisitions of EPI multi-echo T2 star-weighted k-space data by controlling the magnetic resonance imaging system with EPI multi-echo pulse sequence commands. Finally, execution of the machine-executable instructions further causes the computing system to control the stimulation system with a stimulation signal during acquisition of the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data. In some examples, there is metadata recorded in the k-space data or elsewhere, or in clock information used to synchronize the stimulation signal with the acquisition of the T2 star-weighted k-space data.

[0026] In another embodiment, the stimulation system is a visual stimulation system. The brain volume includes the visual cortex. The stimulation signal provided by the stimulation system may vary from example to example. In some cases, the timing of events in the stimulation signal is randomly varied to prevent a person from counting or predicting when the stimulation will occur.

[0027] In another embodiment, the EPI multi-echo pulse sequence command is a single-shot EPI pulse sequence command. The EPI multi-echo pulse sequence command is a multi-band pulse sequence command, which has the advantage that T2 star-weighted k-space data is acquired more quickly.

[0028] In another embodiment, execution of the machine-executable instructions further causes the computing system to calculate any one of the following parameters from the subject-specific hemodynamic response function: maximum amplitude, time to maximum amplitude, FWHM or full width at half maximum of the subject-specific hemodynamic response function, skewness of the subject-specific hemodynamic response function, integral of the subject-specific hemodynamic response function, initial maximum upslope, maximum downslope, and combinations thereof, all of which are potentially useful in providing diagnostic information describing the subject.

[0029] In another embodiment, execution of the machine-executable instructions further causes the computing system to receive functional magnetic resonance imaging k-space data describing a region of the subject's brain. Execution of the machine-executable instructions further causes the computing system to calculate a functional magnetic resonance image using the functional magnetic resonance imaging k-space data and a subject-specific hemodynamic response function. In this embodiment, the subject-specific hemodynamic response function was used to analyze or reconstruct a conventional functional magnetic resonance image. This is advantageous because conventionally, a specific value for the hemodynamic response function is used. However, this is not the case because this function can vary from person to person. This embodiment provides higher quality and / or more accurate functional magnetic resonance images.

[0030] In another embodiment, execution of the machine-executable instructions further causes the computing system to construct a percentage change mapping from the R2 star map using the stimulus signal. Execution of the machine-executable instructions further causes the computing system to derive one or more seed voxels by searching for a predetermined threshold of voxels in the percentage change mapping. In this embodiment, the seed voxels are searched for using event-specific data.

[0031] In another embodiment, the time series of R2 star maps includes a block-relationship R2 star map. In block-relationship functional magnetic resonance imaging, a subject is presented with stimuli over time periods separated by rest periods. This is useful for identifying specific brain regions activated by specific stimuli. In this embodiment, the block-relationship R2 star map is a map spanning such a time period. Execution of the machine-executable instructions further causes the computing system to identify one or more seed voxels by constructing a percentage change mapping from the block-relationship R2 star map by calculating the change between rest and stimulation blocks. Execution of the machine-executable instructions further causes the computing system to identify one or more seed voxels by searching for voxels above a predetermined threshold in the percentage change mapping, resulting in one or more seed voxels. This also corresponds to searching for a certain number of voxels with the highest values.

[0032] In another aspect, the present invention provides a method of medical imaging. The method includes receiving a time series of R2 star maps for a brain volume of a subject. The method further includes receiving a stimulation signal describing occurrences of sensory stimuli repeatedly presented to the subject. The stimulation signal is synchronized to the time series of R2 star maps. The method further includes receiving a selection of one or more seed voxels identified in the time series of R2 star maps. The method further includes calculating a denoised time series of R2 star maps by applying a temporal filter algorithm to the time series of R2 star maps. The method further includes calculating a correlation map for each voxel of the one or more seed voxels by calculating a pixel-by-pixel correlation value between each voxel of the one or more seed voxels and the denoised time series of R2 star maps. The method further includes determining activated regions of the brain volume by combining voxels identified in the correlation map for each voxel of the one or more seed voxels that exceed a predetermined threshold.

[0033] The method further includes aligning the time series of R2 star maps with the stimulation signal to generate a hemodynamic response function for each voxel at each occurrence of the sensory stimulus in the activated region of the brain volume. The method further includes averaging the hemodynamic response function for each voxel and the hemodynamic response function for each occurrence of the sensory stimulus in the activated region of the brain volume to generate a subject-specific hemodynamic response function. The advantages of this embodiment have been described above.

[0034] In another aspect, the present invention provides a computer program comprising machine-executable instructions for execution by a computing system, the computer program being stored in a memory or storage device, for example a non-transitory storage medium.

[0035] Execution of the machine-executable instructions causes the computing system to receive a time series of R2 star maps for a brain volume of the subject. Execution of the machine-executable instructions further causes the computing system to receive a stimulus signal describing the occurrence of a sensory stimulus repeatedly presented to the subject, the stimulus signal being synchronized to the time series of R2 star maps. Execution of the machine-executable instructions further causes the computing system to receive a selection of one or more seed voxels identified in the time series of R2 star maps. Execution of the machine-executable instructions further causes the computing system to calculate a denoised time series of R2 star maps by applying a temporal filter algorithm to the time series of R2 star maps.

[0036] Execution of the machine-executable instructions further causes the computing system to calculate a correlation map for each voxel of the one or more seed voxels by calculating a pixel-by-pixel correlation value between each voxel of the one or more seed voxels and the denoised time series of the R2 star map. Execution of the machine-executable instructions further causes the computing system to determine regions of activation of the brain volume by combining voxels identified in the correlation map for each voxel of the one or more seed voxels that exceed a predetermined threshold.

[0037] Execution of the machine-executable instructions further causes the computing system to align the time series of the R2 star map with the stimulation signal to yield a hemodynamic response function for each voxel at each occurrence of the sensory stimulus in the activated region of the brain volume. Execution of the machine-executable instructions further causes the computing system to averaging the hemodynamic response functions for each voxel at each occurrence of the sensory stimulus in the activated region of the brain volume to yield a subject-specific hemodynamic response function. The advantages of this embodiment have been described above.

[0038] It should be understood that one or more of the above-described embodiments of the present invention may be combined, as long as the combined embodiments are not mutually exclusive.

[0039] As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as an apparatus, a method, or a computer program product. Accordingly, aspects of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, microcode, etc.), or an embodiment combining software and hardware aspects, all generally referred to herein as a "circuit," "module," or "system." Furthermore, aspects of the present invention may take the form of a computer program product embodied in one or more computer-readable medium(s) having computer-executable code embodied thereon.

[0040] Any combination of one or more computer-readable media may be utilized. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. As used herein, "computer-readable storage medium" encompasses any tangible storage medium that stores instructions executable by a processor or computing system of a computing device. The computer-readable storage medium may also be referred to as a computer-readable non-transitory storage medium. The computer-readable storage medium may also be referred to as a tangible computer-readable medium. In some embodiments, the computer-readable storage medium may also store data that can be accessed by the computing system of a computing device. Examples of computer-readable storage media include, but are not limited to, floppy disks, magnetic hard disk drives, solid-state hard disks, flash memory, USB thumb drives, random access memory (RAM), read-only memory (ROM), optical disks, magneto-optical disks, and computing system register files. Examples of optical disks include compact disks (CDs) and digital versatile disks (DVDs), such as CD-ROM disks, CD-RW disks, CD-R disks, DVD-ROM disks, DVD-RW disks, or DVD-R disks. The term computer-readable storage medium also refers to various types of storage media that can be accessed by a computing device over a network or communications link. For example, data may be retrieved via a modem, over the Internet, or over a local area network. Computer-executable code embodied on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, fiber optic cable, RF, etc., or any suitable combination of the above.

[0041] A computer-readable signal medium includes a propagated data signal having computer-executable code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including, but not limited to, an electromagnetic signal, an optical signal, or any suitable combination thereof. A computer-readable signal medium may be any computer-readable medium, other than a computer-readable storage medium, that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0042] "Computer memory" or "memory" is an example of a computer-readable storage medium. Computer memory is any memory that is directly accessible to a computing system. "Computer storage" or "storage" is a further example of a computer-readable storage medium. Computer storage is any non-volatile computer-readable storage medium. In some embodiments, computer storage can also be computer memory, and computer memory can also be computer storage.

[0043] As used herein, a "computing system" encompasses an electronic component capable of executing a program or machine-executable instructions or computer-executable code. References to a computing system, including examples of a "computing system," should be interpreted as including more than one computing system or processing core, as the case may be. A computing system is, for example, a multi-core processor. A computing system may also refer to a collection of computing systems, either within a single computer system or distributed among multiple computer systems. The term computing system should also be interpreted as referring to a collection or network of computing devices, each of which may include a processor or computing system. Machine-executable code or instructions may be executed by multiple computing systems or processors, either within the same computing device or even distributed across multiple computing devices.

[0044] Machine-executable instructions or computer-executable code include instructions or programs that cause a processor or other computing system to perform aspects of the present invention. Computer-executable code for performing operations for aspects of the present invention may be written in any combination of one or more programming languages ​​and compiled into machine-executable instructions, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. In some cases, the computer-executable code is in the form of a high-level language or pre-compiled and used with an interpreter that generates the machine-executable instructions on the fly. In other cases, the machine-executable instructions or computer-executable code is in the form of programming for a programmable logic gate array.

[0045] The computer executable code may execute entirely on the user's computer, as a stand-alone software package, partially on the user's computer, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be to an external computer (e.g., via the Internet using an Internet Service Provider).

[0046] Aspects of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that, where applicable, each block or portion of a block in the flowcharts, illustrations, and / or block diagrams can be implemented by computer program instructions in the form of computer-executable code. It should also be understood that combinations of blocks in different flowcharts, illustrations, and / or block diagrams can be combined when not mutually exclusive. These computer program instructions are provided to a general-purpose computer, a special-purpose computer, or other programmable data processing device computing system to produce a machine, such that the instructions, when executed by the computer or other programmable data processing device computing system, create means for performing the functions / acts specified in the flowcharts and / or one or more blocks of the block diagrams.

[0047] These machine-executable instructions or computer program instructions may also be stored on a computer-readable medium that can direct a computer, other programmable data processing apparatus, or other device to function in a particular manner, such that the instructions stored on the computer-readable medium produce an article of manufacture including instructions that implement the functions / acts specified in the flowcharts and / or one or more blocks of the block diagrams.

[0048] The machine-executable instructions or computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause the computer, other programmable apparatus, or other device to perform a series of operational steps to generate a computer-implemented process, such that the instructions, executing on the computer or other programmable apparatus, result in a process for performing the functions / acts specified in the flowcharts and / or one or more blocks of the block diagrams.

[0049] As used herein, a "user interface" is an interface that allows a user or operator to interact with a computer or computer system. A "user interface" is sometimes referred to as a "human interface device." A user interface provides information or data to an operator and / or receives information or data from an operator. A user interface allows a computer to receive input from an operator and provides output from the computer to a user. In other words, a user interface allows an operator to control or manipulate a computer, and the interface allows a computer to show the effects of the operator's control or manipulation. The display of data or information on a display or graphical user interface is an example of providing information to an operator. Receiving data via a keyboard, mouse, trackball, touchpad, pointing stick, graphics tablet, joystick, gamepad, webcam, headset, pedals, wired gloves, remote control, and accelerometer are all examples of user interface components that allow for the reception of information or data from an operator.

[0050] As used herein, a "hardware interface" encompasses an interface that allows a computing system of a computer system to interact with and / or control external computing devices and / or devices. A hardware interface allows a computing system to send control signals or commands to external computing devices and / or devices. A hardware interface also allows a computing system to exchange data with external computing devices and / or devices. Examples of hardware interfaces include, but are not limited to, a universal serial bus, an IEEE 1394 port, a parallel port, an IEEE 1284 port, a serial port, an RS-232 port, an IEEE-488 port, a Bluetooth connection, a wireless local area network connection, a TCP / IP connection, an Ethernet connection, a control voltage interface, a MIDI interface, an analog input interface, and a digital input interface.

[0051] As used herein, "display" or "display device" encompasses an output device or user interface adapted to display images or data. A display outputs visual, auditory, and / or tactile data. Examples of displays include, but are not limited to, computer monitors, television screens, touch screens, tactile electronic displays, Braille screens, cathode ray tubes (CRTs), storage tubes, bi-stable displays, electronic paper, vector displays, flat panel displays, vacuum fluorescent displays (VFs), light-emitting diode (LED) displays, electroluminescent displays (ELDs), plasma display panels (PDPs), liquid crystal displays (LCDs), organic light-emitting diode displays (OLEDs), projectors, and head-mounted displays.

[0052] K-space data is defined herein as data that are recorded measurements of radio frequency signals emitted by atomic spins using the antenna of a magnetic resonance machine during a magnetic resonance imaging scan.

[0053] A magnetic resonance imaging (MRI) image or MR image is defined herein as an image that is a reconstructed two-dimensional or three-dimensional visualization of the anatomical data contained within the k-space data, which visualization can be performed using a computer.

[0054] Preferred embodiments of the invention will now be described, by way of example only, with reference to the drawings in which: [Brief explanation of the drawings]

[0055] [Figure 1] FIG. 1 illustrates an example of a medical system. [Figure 2] FIG. 2 shows a flowchart illustrating a method of using the medical system of FIG. [Figure 3] FIG. 1 illustrates a further example of a medical system. [Figure 4] FIG. 4 shows a flowchart illustrating a method of using the medical system of FIG. 3. [Figure 5] FIG. 10 is a diagram showing an example of a T1 weighted image. [Figure 6] FIG. 1 shows three views of a 3D R2 star image showing percent change in activation for task blocks. [Figure 7] FIG. 7 shows event-related R2 star data for the seed voxels selected in FIG. 6. [Figure 8] FIG. 1 shows the Pearson correlation coefficient for one slice of the brain volume. [Figure 9] FIG. 1 is a diagram showing an example of an activation region. [Figure 10] FIG. 10 shows the sum of the R2 star values ​​for all activation regions shown in FIG. 9. [Figure 11] FIG. 1 illustrates hemodynamic response functions. [Figure 12] FIG. 12 shows a histogram 1200 of the maximum values ​​of the hemodynamic response function. [Figure 13] FIG. 1 shows the mean hemodynamic response function. [Figure 14] FIG. 1 shows subject-specific hemodynamic response functions for multiple individuals. [Figure 15] FIG. 1 illustrates the advantage of using a three-point measurement to determine the R2 star value. DETAILED DESCRIPTION OF THE INVENTION

[0056] Like numbered elements in these figures are equivalent elements or perform the same function. An element described earlier is not necessarily described in a later figure if the function is equivalent.

[0057] FIG. 1 illustrates an example of a medical system 100. In this example, the medical system 100 includes a computer 102 having a computing system 104. The computer 102 may be, for example, a distributed computing system, as well as a computer located remotely or provided via a cloud web service. The computer 102 may also be a workstation or computer used by a radiologist or other medical professional. The computer 102 may also be a control system for a magnetic resonance imaging system. The computer 102 is shown as including the computing system 104, which is representative of one or more computing systems that may be located in one or more locations. The computing system 104 is connected to an optional hardware interface 106 and an optional user interface 108. If present, the hardware interface 106 allows the computing system 104 to control other components of the medical system 100, such as the magnetic resonance imaging system. The user interface 108 allows an operator to control, operate, and interact with the medical system 100.

[0058] Computing system 104 is further shown to be in communication with memory 110, which is representative of various types of memory that may be in communication with computing system 104. Memory 110 is shown to contain machine-executable instructions 120, which enable computing system 104 to perform various tasks, such as controlling other components of medical system 100 and performing numerical and image processing tasks. Memory 110 is further shown to contain a time series 122 of R2 star maps for brain volume.

[0059] The memory 110 is further shown as containing a stimulation signal 124 synchronized to the time series 122 of the R2 star map. The memory 110 is further shown as containing a selection of one or more seed voxels 126 in the brain volume. The memory 110 is further shown as containing a denoised time series 128 of the R2 star map. The memory 110 is further shown as containing a correlation map 130 between one or more seed voxels 126 and the remaining voxels in the denoised time series of the R2 star map. The memory 110 is further shown as containing identified activation regions 132 in the brain volume found by thresholding the correlation maps 130 and then combining them with each other.

[0060] Memory 110 is further shown as containing a hemodynamic response function 134 for each voxel in activation region 132 and for each time the stimulus is presented, as shown in stimulus signal 124. For example, for one particular voxel, there may be several hemodynamic response functions 134 for each time the stimulus is presented. Memory 110 is further shown as containing a subject-specific hemodynamic response function 136 found by averaging the hemodynamic response functions 134.

[0061] FIG. 2 shows a flowchart illustrating a method of operating the medical system 100 of FIG. 1. First, in step 200, a time series 122 of R2 star maps is received. Next, in step 202, a stimulus signal 124 is received. As previously described, the stimulus signal 124 is synchronized with the time series 122 of R2 star maps. The stimulus signal 124 represents when a sensory stimulus is given or presented to a subject using a stimulus generator, such as an auditory or visual display. Next, in step 204, a selection of one or more seed voxels 126 is received. This may be done manually by an operator, for example, or via an automated algorithm. Next, in step 206, a denoised time series 128 of R2 star maps is calculated from the time series 122 of R2 star maps.

[0062] There is a series of R2 star maps. Taking a particular voxel from each map and then taking its value yields a time-based R2 star signal for each voxel. This signal for each voxel is then denoised to yield a denoised time series of R2 star maps. This can be achieved, for example, using a digital filter or by fitting a curve, such as a spline, to the data. Next, in step 208, a correlation map is calculated for each of a selection of one or more seed voxels 126. For example, a correlation coefficient is calculated between the time signal for one or more seed voxels and each of the remaining voxels in the denoised time series R2 star maps. In step 210, activation regions 132 of the brain volume are then calculated. This is calculated by taking the correlation maps 130 and thresholding them.

[0063] By thresholding the single correlation map, regions that are temporally correlated with a particular seed voxel are identified. The activation region is then the combination of the thresholded correlation maps. Then, in step 212, the time series of R2 star maps is aligned with the stimulation signal to yield a hemodynamic response function 134 for each voxel in the activation region of the brain volume and each occurrence of the sensory stimulus. Finally, in step 214, a subject-specific hemodynamic response function 136 is then calculated by averaging the hemodynamic response functions 134. Additional operations, such as smoothing or curve fitting, may also be performed, but this is not required.

[0064] 3 shows a diagram illustrating a further example of a medical system 300. The medical system 300 is similar to the medical system 100 shown in FIG. 1, except that the medical system 300 further includes a magnetic resonance imaging system 302.

[0065] The magnetic resonance imaging system 302 includes a magnet 304. The magnet 304 is a superconducting cylindrical magnet with a bore 306 extending through the magnet. Different types of magnets can be used, including both split cylindrical magnets and so-called open magnets. Split cylindrical magnets are similar to standard cylindrical magnets except that the cryostat is split into two sections to allow access to the magnet's isoplane; such magnets are used, for example, with charged particle beam therapy. Open magnets have two magnet sections, one above the other, with a space between them large enough to accommodate a subject. The arrangement of the two sections is similar to that of a Helmholtz coil. Open magnets are popular because they are less confined to the subject. Inside the cryostat of a cylindrical magnet is a collection of superconducting coils.

[0066] Within the bore 306 of the cylindrical magnet 304 is an imaging zone 308 where the magnetic field is strong and uniform enough to perform magnetic resonance imaging. Shown within the imaging zone 308 is a field of view 309. Magnetic resonance data is generally collected about the field of view 309. The field of view 309 is shown to image a brain volume of a subject 318, which is shown as being supported by a subject support 320.

[0067] Within the imaging zone 308, the head of the subject 318 is within the head coil 314. This allows imaging of the field of view 309. In some cases, the brain volume is identical to the field of view 309. In other cases, the brain volume is within the field of view 309. A T1 weighted image may have a larger field of view than, for example, an EPI multi-echo T2 star weighted image.

[0068] Above the head of subject 318 is a stimulation system 322, which in this example is a display. Various types of displays can be used, such as a mirror that reflects a view of a projection outside the bore 306 of magnet 304, as well as a magnetic resonance compatible display directly above the eyes of subject 318. Other types of stimulation can also be used, such as sound stimulation and also tactile stimulation.

[0069] Also within the magnet bore 306 are a set of magnetic field gradient coils 310 used for preliminary magnetic resonance data acquisition for spatially encoding magnetic spins within the imaging zone 308 of the magnet 304. The magnetic field gradient coils 310 are connected to a magnetic field gradient coil power supply 312. The magnetic field gradient coils 310 are representative. Typically, the magnetic field gradient coils 310 include three separate sets of coils for spatial encoding in three orthogonal spatial directions. The magnetic field gradient power supply supplies current to the magnetic field gradient coils 310. The current supplied to the magnetic field gradient coils 310 is controlled as a function of time and may be ramped or pulsed.

[0070] Adjacent to the imaging zone 308 is a radio frequency coil 314 for manipulating the orientation of magnetic spins within the imaging zone 308 and for receiving radio transmissions from spins also within the imaging zone 308. In this case, the radio frequency coil 314 is a head coil. A radio frequency antenna may include multiple coil elements. A radio frequency antenna may also be referred to as a channel or antenna. The radio frequency coil 314 is connected to a radio frequency transceiver 316. The radio frequency coil 314 and the radio frequency transceiver 316 may be replaced with separate transmit and receive coils and separate transmitters and receivers. It should be understood that the radio frequency coil 314 and the radio frequency transceiver 316 are representative. The radio frequency coil 314 may also represent a dedicated transmit antenna and a dedicated receive antenna. Similarly, the transceiver 316 may also represent a separate transmitter and receiver. The radio frequency coil 314 may also have multiple receive / transmit elements, and the radio frequency transceiver 316 may have multiple receive / transmit channels.

[0071] The transceiver 316 and gradient controller 312 are shown connected to the hardware interface 106 of the computer system 102. Both of these components, as well as other components such as a subject support that provides position data, may provide sensor data 126.

[0072] Memory 110 is further shown as containing EPI multi-echo pulse sequence commands configured to acquire EPI multi-echo T2 star-weighted k-space data. Memory 110 is further shown as containing T1 weighted pulse sequence commands 332 configured to control the magnetic resonance imaging system 302 to acquire T1 weighted k-space data. Pulse sequence commands are generally commands used to control the magnetic resonance imaging system 302 to acquire k-space data according to a particular magnetic resonance imaging protocol.

[0073] Memory 110 is further shown as containing EPI multi-echo T2 star-weighted k-space data 334 acquired by controlling magnetic resonance imaging system 302 with EPI multi-echo pulse sequence commands 330. Memory 110 is further shown as containing T1 weighted k-space data 336 acquired by controlling magnetic resonance imaging system 302 with T1 weighted pulse sequence commands 332. Memory 110 is further shown as containing T1 weighted images 338 of a brain volume reconstructed from T1 weighted k-space data 336. Memory 110 is further shown as containing T2 star weighted images 340 for each echo of EPI multi-echo T2 star-weighted k-space data 334. Memory 110 is further shown as containing registered T2 star weighted images 342 for each echo.

[0074] 4 shows a flowchart illustrating how the medical system 300 of FIG. 3 operates. Initially, in step 400, T1-weighted k-space data 336 is acquired by controlling the magnetic resonance imaging system 302 using a T1-weighted pulse sequence command 332. Then, in step 402, multiple acquisitions of EPI multi-echo T2 star-weighted k-space data 334 are acquired by controlling the magnetic resonance imaging system multiple times using an EPI multi-echo pulse sequence command 330. During the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data 334, step 404 is performed. In step 404, the stimulation system 332 is controlled using a stimulation signal 124 to provide a sensory stimulus to the subject 318.

[0075] The stimulation signal 124 is then synchronized with the various acquisitions of EPI multi-echo T2 star-weighted k-space data 334. Then, in step 406, a T1-weighted image 338 of the brain volume is reconstructed from the T1-weighted k-space data 336. Next, in step 408, a T2 star-weighted image is reconstructed for each of the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data 334. Then, in step 410, a matched T2 star-weighted image 342 for each echo is calculated for each echo of the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data by performing pre-processing to match the T2 star-weighted image for each echo of the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data with the T1-weighted image of the brain volume. Next, in step 412, a time series of R2 star maps 122 for the subject's brain volume for each voxel is calculated by fitting attenuation curves to the aligned T2 star-weighted images for each echo of the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data. After step 412 is performed, the method then proceeds to steps 200-214 shown in FIG.

[0076] Functional fMRI is an important tool in neuroscience, and there is growing interest and evidence that fMRI can be used for diagnostic purposes. Particularly in the field of psychiatric disorders, fMRI could prove to be a game-changer. For example, in many cases, anatomical scans do not reveal any abnormalities in these patients. However, functional MRI shows that these patients have abnormal responses, i.e., stronger or weaker responses, in other areas of the brain. The fMRI responses are the result of hemodynamic fluctuations, increased blood flow, and different ratios of oxygenated to deoxygenated hemoglobin in the responding areas.

[0077] Variations in blood flow are described by the hemodynamic response function (HRF). Most analytical tools use a constant HRF for all subjects. As will be shown below, HRF can vary from subject to subject, and the HRF may have additional diagnostic value.

[0078] As disclosed herein, no a priori knowledge is used to determine the HRF, and high time resolution measurements utilizing multi-echo EPI are performed.

[0079] In some examples, the methods rely on fMRI data recorded using multi-echo EPI, which increases the contrast-to-noise ratio (CNR) by approximately 30%. Some example benefits include one or more of the following: Analysis of HRF without prior knowledge Increasing SNR by using multi-echo fMRI High spatial resolution.

[0080] fMRI Task: In the following examples, subjects performed visual tasks: a block design and an event-relationship task. In all experiments, the visual stimulus was a flickering checkerboard. For the block design, the duration was 30 seconds and the flickering frequency was 8 Hz. A total of eight task blocks were recorded, separated by 30-second rest blocks. The experiment ended with the rest block. For the event-relationship experiment, a checkerboard with a 1-second duration and a 10-Hz frequency was used. The time between event onsets was either 20 seconds or 30 seconds, and was randomly varied. The start times were 20 seconds, 50 seconds, 80 seconds, 100 seconds, 130 seconds, 150 seconds, 170 seconds, 200 seconds, 230 seconds, 250 seconds, 280 seconds, 310 seconds, 330 seconds, 350 seconds, 380 seconds, 400 seconds, 420 seconds, 450 seconds, 470 seconds, and 490 seconds. A total of 20 events were displayed. The event-related behavior of both was identical.

[0081] The paradigm was visually presented using a beamer or projector, and the exact onset of all stimuli was controlled by a TLL pulse from the scanner.

[0082] All MRI experiments were recorded on a 3T MRI system using a 32-channel head coil. All fMRI experiments had the same parameter settings.

[0083] Single-shot EPI: Tr = 500 ms, echo time = 12, 28, 44 ms, multiband acceleration (MB) = 3 slices, voxel size = 2.75 × 2.75 × 3, 21 slices, 1020 kinetics. Since a short TR was the "holy grail," the number of slices was the limiting factor. As a result, fMRI scans did not cover the entire brain in the FH direction, field of view (fov) 63 mm.

[0084] FIG. 5 shows an example of a T1 weighted (T1w) image 338. Within the T1 weighted image 338, a brain volume 500 is shown. In this example, the brain volume 500 includes the visual cortex. In the illustrated example, a visual stimulus was used. Therefore, time can be saved by limiting the brain volume 500 to the visual cortex. The T1 weighted scan was performed using multi-shot TFE, Tr / TE=8.3 / 3.8 ms, 1 mm 3 The voxel size was 500. Single-shot EPI data were collected in the brain volume.

[0085] Pre-processing of the fMRI data comprises one or more of the following steps performed on the T1w images (fMRI data) and the T1w images: 1. Co-register the T1w image to the first dynamics of the first echo (of three) (first acquisition of block MRI data). (The choice to use the first dynamics is optional.) 2. Segment the T1-weighted image for gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) (SPM segmentation) 3. Reslicing T1w images and segmentation to match the first kinetics of echo 1 (SPM Reslicing) 4. Construct masks from the segmentations so that there are masks for the GM and the whole brain 5. Realign each echo of the fMRI data separately to the first kinetic (SPM realignment) 6. Perform linear detrending of fMRI data 7. Smoothing fMRI data using a Gaussian kernel, FWHM5 (SPM) 8. fMRI: Perform a log-linear fit of the echoes using the fMRI data to create 4D-R2 star and D-S0 maps. Data outside the whole brain mask is excluded. 10. fMRI: Constructing functional activation maps for the blocking task as PSC based on the R2 star map

[0086] Figure 6 shows three views 600, 602, 604 of the 3D R2 star image showing percent change in activation for task blocks. These images were used to manually select a seed voxel 126. This seed voxel 126 was marked in all three images. Figure 6 shows the percent signal change (PSC) of activation for task block 4 from the R2 star data.

[0087] The HRF reactions were selected as follows. 1) Manually select one or two seed voxels (block paradigm) in the activated region (see Figure 6). 2) Select the time course of changes in selected voxels in the event-related data. 3) Apply a smoothing spline to these seed voxels (Figure 7).

[0088] Figure 7 shows the event-related R2 star data for the seed voxel 126 selected in Figure 6. Curve 700 shows the raw data. Curve 702 shows the same data in curve 700, but with a translation or noise removal algorithm applied to the data. Curve 704 shows the difference between curves 700 and 702, indicating the noise that has been removed.

[0089] Next, the Pearson correlation coefficient is calculated between the smooth voxel 702 and all smooth voxels in the four-dimensional dataset of fMRI data to determine the activated regions 132 of the brain volume 500 . 4) Determine the Pearson correlation coefficient between the smoothed voxel and every other smoothed voxel in the 4D dataset of the event-related data in Figures 8 and 9 below.

[0090] Figure 8 shows an image of the Pearson correlation coefficient 800 for one slice in one time period of an R2 star map for a brain volume. The Pearson correlation coefficient 800 is visible in the center, and the surrounding region 802 is masked. By thresholding the Pearson correlation value 800, the activation region 132 was identified and is displayed in Figure 9. Figure 9 was generated by thresholding Figure 8. 5) Apply a threshold (0.75) to select voxels from the original unsmoothed data. Detrending (order 5) and averaging over time. Figure 10 shows the sum of the R2 star values ​​for all activated regions 132 shown in Figure 9. This is the spatially averaged signal 1000. 6) - Select HRF responses using a 5- to 20-second window. The onset of the stimulus is at 0 seconds. Average all selected HRF responses (Figure 11). All individual responses (unsmoothed) are saved for further analysis.

[0091] Figure 11 shows a hemodynamic response function 1100 calculated by summing HRFs taken from curve 1000 in Figure 10 using the selected criteria just described. There are 106 voxels with 20 HRF responses per voxel. This results in 4100 responses that are summed to construct HRF 1100. 7) (Again) fit a smoothing spline to all individual responses saved in step 5. Determine the time maxima and create a histogram (see Figure 12).

[0092] Figure 12 shows a histogram 1200 of the maxima of the 4100 individual hemodynamic response functions. Within this figure, a maximum value 1202 and several outliers 1204 can be seen. Before proceeding, the hemodynamic response functions in the region labeled 1204 are excluded from the analysis, thereby eliminating the outliers. In this example, the histogram was used to select the outliers, but other values, such as those within a certain number of standard deviations of the maximum value 1202, would also work. 8) Select the HRF response around the maximum value in the histogram (window: -1.5 to +2.0 s). Sum the selected unsmoothed HRF responses. Save the sum (mean) for further analysis (Figure 13).

[0093] Figure 13 shows an average hemodynamic response function 1300 calculated from the average of 2555 hemodynamic response functions that "survived" selection based on the histogram 1200 shown in Figure 12. The hemodynamic response function 1300 still exhibits some noise and is relatively rough. The hemodynamic response function (HRF) 1300 can be denoised or fitted to a spline function, for example, following a subject analysis of the HRF response.

[0094] A smoothing spline is applied to the sum. All further processing and analysis is performed on the smoothed HRF.

[0095] For each HRF, the maximum R2 value is determined from the smoothed curve. The signal value at the start of the event (t=0) is taken as the reference. The percentage signal change (PSC) is calculated relative to the reference to yield the HRF curve as PSC. An example of a found HRF curve is shown in Figure 14.

[0096] 14 shows subject-specific hemodynamic response functions 1400 for multiple individuals. These subject-specific hemodynamic response functions are labeled HRF1002-HRF1108, HRF1110, and HRF1101-HRF1109. As shown in FIG. 14, it can be seen that the subject-specific hemodynamic response functions 1400 vary significantly from individual to individual, which also illustrates the benefits of using the subject-specific hemodynamic response functions 1400 when performing other functional magnetic resonance imaging protocols.

[0097] Several parameters of these HRF responses can be determined. Maximum amplitude (PSC max value). Time to maximum value Response width (FWHM) Response skewness Reaction Integral Initial maximum uphill gradient (gradient 1) Maximum downward slope (slope 2)

[0098] An example is shown in the table below: The data is an analysis of the HRF response shown in FIG.

[0099] [Table 1]

[0100] The following illustrates the advantages of multi-echo fMRI. Both R2 star data and Echo 2 data (HRF determination using only two echoes) were analyzed. Echo 2 was recorded with a TE of 28 ms, which is representative of studies using single-echo fMRI. For both data sets, the same processing derived above was used. The maximum PSC values ​​for both data sets were determined, as shown in Figure 15 below.

[0101] FIG. 15 illustrates the benefit of using three-point measurements to determine R2 star values. In FIG. 15, an analysis was used in which both data sets had the same data processing as derived above, using the three-point R2 star values ​​1500 described herein and also using a second echo at a time of 28 ms, which is quite common or representative of studies using single-echo functional magnetic resonance imaging. The maximum percent change values ​​for both data sets were determined as shown in FIG. 15. Each R2 star value 1500 is above the second echo 1502 data. It can be seen that there is a systematic error when using single-echo functional magnetic resonance imaging techniques. The average difference between the R2 star data and echo 2 is 34%.

[0102] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description is to be considered illustrative or exemplary and not restrictive. The invention is not limited to the disclosed embodiments.

[0103] Other variations of the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprises" does not exclude other elements or steps, and the singular form of an element does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that several means are recited in mutually different dependent claims does not indicate that a combination of these means cannot be used to advantage. A computer program can be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium, provided together with or as part of other hardware, as well as distributed in other forms, such as via the Internet or other wired or wireless telecommunications systems. Any reference signs in the claims should not be construed as limiting the scope. [Explanation of symbols]

[0104] Reference Code List 100 Healthcare Systems 102 Computer 104 Computing Systems 106 Hardware Interface 108 User Interface 110 memory 120 machine-executable instructions 122 R2 Star Map Time Series 124 Stimulus signal 126 Selection of one or more seed voxels Denoised time series of 128 R2 star maps 130 Correlation Map 132 Activated Areas of Brain Volume 134 Hemodynamic Response Function 136 Subject-specific hemodynamic response functions 200 Receive R2 starmap time series for subject brain volume 202 receive a stimulus signal describing the occurrence of a sensory stimulus repeatedly presented to the subject; 204 Receive a selection of one or more seed voxels identified in the R2 star map time series. 206 Compute a denoised time series of R2 star maps by applying a temporal filter algorithm to the R2 star map time series 208. Compute a correlation map for each voxel of the one or more seed voxels by computing a pixel-by-pixel correlation value between each voxel of the one or more seed voxels and the denoised time series of the R2 star map. 210. Determine activation regions of the brain volume by combining voxels identified in the correlation map for each voxel of one or more seed voxels above a predetermined threshold. Aligning the time series of the 212 R2 star map with the stimulus signal yields a hemodynamic response function for each voxel in the activated region of the brain volume and for each occurrence of sensory stimulation. Averaging the hemodynamic response functions for each voxel in the activated region of the brain volume and each occurrence of sensory stimulation yields a subject-specific hemodynamic response function. 300 Healthcare Systems 302 Magnetic Resonance Imaging System 304 Magnet 306 Magnet Bore 308 Imaging Zone 309 Visibility 310 Magnetic field gradient coil 312 Magnetic field gradient coil power supply 314 Head Coil 316 Transceiver 318 Target 320 Subject Support Platform 322 Stimulation System (Display) 330 EPI multi-echo pulse sequence commands 332 T1 weighted pulse sequence command 334 EPI multi-echo T2 star-weighted k-space data 336 T1-weighted k-space data 338 T1-weighted images of brain volumes 340 T2 star-weighted images for each echo 342 Matched T2 star-weighted images for each echo Acquire T1-weighted k-space data by controlling the magnetic resonance imaging system with 400 T1-weighted pulse sequence commands Collect multiple acquisitions of EPI multi-echo T2 star-weighted k-space data by controlling the magnetic resonance imaging system using 402 EPI multi-echo pulse sequence commands. 404 Controlling the stimulation system using stimulation signals during the acquisition of multiple acquisitions of EPI multi-echo T2 star-weighted k-space data 406 Reconstructing T1-weighted images of brain volumes from T1-weighted k-space data 408 EPI Multi-echo T2 Star-weighted k-space data is used to reconstruct a T2 star-weighted image for each echo of multiple acquisitions. 410. Calculate an aligned T2 star-weighted image for each echo of the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data by performing preprocessing to align the T2 star-weighted image for each echo of the multiple acquisitions of EPI multi-echo T2 star-weighted k-space data with the T1-weighted image of the brain volume. Calculate a time series of R2 star maps for the subject's brain volume for each voxel by fitting attenuation curves to matched T2 star-weighted images for each echo of multiple acquisitions of 412 EPI multi-echo T2 star-weighted k-space data. 500 brain volumes First view of 600 R2 star image 602 R2 Second view of star image Third view of the 604 R2 star image Event-related R2 star time series for 700 seed voxels 702 A smoothed version of the 700 704 Difference between 700 and 702 Pearson correlation coefficient for 800 brain volumes (1 slice) 802 Masked Area 1000 spatially averaged signals 1100 Hemodynamic Response Function 1200 Histogram of maximum values ​​of hemodynamic response function 1202 Maximum 1204 outliers 1300 Averaged Hemodynamic Response Functions 1400 Subject-specific hemodynamic response functions for multiple individuals

Claims

1. a memory storing machine-executable instructions; Computing systems and 10. A medical system comprising: receiving a time series of R2 star maps of a brain volume of the subject; receiving a stimulus signal describing occurrences of sensory stimuli repeatedly presented to the subject, the stimulus signal being synchronized with the time series of the R2 star map; receiving a selection of one or more seed voxels identified in the time series of the R2 star map; and - calculating a denoised time series of the R2 star maps by applying a temporal filter algorithm to the time series of the R2 star maps; calculating a correlation map for each voxel of the one or more seed voxels by calculating a pixel-by-pixel correlation value between each voxel of the one or more seed voxels in the time series of the R2 star map and the denoised time series of the R2 star map; thresholding the correlation map to determine regions of activation in the brain volume by combining voxels identified in the correlation map that have values ​​above a predetermined threshold; aligning the time series of the R2 star map with the stimulation signal to yield a hemodynamic response function for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume; averaging the hemodynamic response functions for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume to yield a subject-specific hemodynamic response function; The medical system causes the computing system to perform the above.

2. Execution of the machine-executable instructions further comprises: calculating the time of maximum value of the hemodynamic response function for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume; calculating statistical characteristics of the time of maximum value of the hemodynamic response function for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume; excluding any hemodynamic response function from the calculation of the subject-specific hemodynamic response function if the hemodynamic response function fails to meet the criteria determined using the statistical properties; The medical system of claim 1 , wherein the computing system performs the following:

3. 3. The medical system of claim 2, wherein the time of maximum value of the hemodynamic response function for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume is calculated using a smoothing function.

4. Execution of the machine-executable instructions further comprises: receiving EPI multi-echo T2 star-weighted k-space data describing the brain volume of the subject from multiple acquisitions, the multiple acquisitions being synchronized to the stimulation signal; receiving T1-weighted k-space data describing the brain volume of the subject; reconstructing a T1-weighted image of the brain volume from the T1-weighted k-space data; reconstructing a T2 star-weighted image for each echo of the EPI multi-echo T2 star-weighted k-space data; calculating a registered T2 star weighted image for each echo by performing pre-processing to register the T2 star weighted image for each echo and the T1 weighted image of the brain volume with each other; calculating the time series of R2 star maps for the subject's brain volume for each voxel by fitting an attenuation curve to the matched T2 star weighted images for each echo; The medical system of claim 1 , wherein the computing system performs the following:

5. wherein said calculation of said aligned T2 star-weighted image for each echo is performed by pre-processing to align said T2 star-weighted image for each echo of said EPI multi-echo T2 star-weighted k-space data with said T1 weighted image of said brain volume; registering the T2 star weighted image corresponding to a first echo of the EPI multi-echo T2 star weighted k-space data with the T1 weighted image; segmenting the T1 weighted image to generate a gray matter segmentation, a white matter segmentation, and a cerebrospinal fluid segmentation; using the registration between the T2 star weighted image and the T1 weighted image corresponding to the first echo, re-slicing the T1 weighted image, the gray matter segmentation, the white matter segmentation, and the cerebrospinal fluid segmentation to match the T2 star weighted image corresponding to the first echo; constructing a brain mask using the gray matter segmentation, the white matter segmentation, and the cerebrospinal fluid segmentation; realigning the T2 weighted image for each echo with the T1 weighted image; The medical system of claim 4 , comprising:

6. The medical system of claim 4 , wherein the EPI multi-echo T2 star-weighted k-space data describing the brain volume of the subject comprises k-space data for three echoes.

7. The memory further stores an EPI multi-echo pulse sequence command and a T1 weighted pulse sequence command, and the medical system further includes: a magnetic resonance imaging system; a stimulation system for providing the sensory stimulation to the subject; Equipped with Execution of the machine-executable instructions further comprises: acquiring the T1 weighted k-space data by controlling the magnetic resonance imaging system using the T1 weighted pulse sequence commands; acquiring EPI multi-echo T2 star-weighted k-space data in multiple acquisitions by controlling the magnetic resonance imaging system with the EPI multi-echo pulse sequence commands; controlling the stimulation system with the stimulation signal during the acquisition of the EPI multi-echo T2 star-weighted k-space data; The medical system of claim 1 , wherein the computing system performs the following:

8. The medical system of claim 7 , wherein the stimulation system is a visual stimulation system and the brain volume includes the visual cortex.

9. The medical system of claim 7 , wherein the EPI multi-echo pulse sequence command is a single-shot EPI pulse sequence command, and the EPI multi-echo pulse sequence command is a multi-band pulse sequence command.

10. 2. The medical system of claim 1, wherein execution of the machine-executable instructions further causes the computing system to calculate at least one of the following parameters from the subject-specific hemodynamic response function: maximum amplitude, time to maximum amplitude, FWHM of the subject-specific hemodynamic response function, skewness of the subject-specific hemodynamic response function, integral of the subject-specific hemodynamic response function, initial maximum upslope, or maximum downslope.

11. Execution of the machine-executable instructions further comprises: receiving functional magnetic resonance imaging k-space data describing regions of the subject's brain; calculating a functional magnetic resonance image using the functional magnetic resonance imaging k-space data and the subject-specific hemodynamic response function; The medical system of claim 1 , wherein the computing system performs the following:

12. Execution of the machine-executable instructions further comprises: constructing a percentage change mapping indicating a percent change in activation from the R2 star map using the stimulation signal; and deriving the one or more seed voxels by searching for voxels above a predetermined threshold in the percentage change mapping; The medical system of claim 1 , wherein the computing system performs the following:

13. the time series of R2 star maps includes a block relation R2 star map, and execution of the machine executable instructions comprises: constructing a percentage change mapping indicating percent change in activation from the block-related R2 star map by calculating the change between rest blocks and stimulation blocks; deriving the one or more seed voxels by searching for voxels above a predetermined threshold in the percentage change mapping; The medical system of claim 1 , wherein the computing system is configured to identify the one or more seed voxels by:

14. 1. A method of medical imaging, said method comprising: receiving a time series of R2 star maps for a brain volume of a subject; receiving a stimulus signal describing the occurrence of a sensory stimulus to be repeatedly presented to the subject, the stimulus signal being synchronized with the time series of the R2 star map; receiving a selection of one or more seed voxels identified in the time series of the R2 star maps; - calculating a denoised time series of the R2 star maps by applying a temporal filter algorithm to the time series of the R2 star maps; calculating a correlation map for each voxel of the one or more seed voxels by calculating a pixel-by-pixel correlation value between each voxel of the one or more seed voxels in the time series of the R2 star maps and the denoised time series of the R2 star maps; thresholding the correlation map to determine regions of activation in the brain volume by combining voxels identified in the correlation map that have values ​​above a predetermined threshold; aligning the time series of the R2 star map with the stimulation signal to yield a hemodynamic response function for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume; averaging the hemodynamic response functions for each voxel and for each occurrence of the sensory stimulus in the activated region of the brain volume to yield a subject-specific hemodynamic response function; A method comprising:

15. 1. A computer program comprising machine-executable instructions stored on a non-transitory computer-readable medium for execution by a computing system, wherein execution of the machine-executable instructions comprises: receiving a time series of R2 star maps of a brain volume of the subject; receiving a stimulus signal describing occurrences of sensory stimuli repeatedly presented to the subject, the stimulus signal being synchronized with the time series of the R2 star map; receiving a selection of one or more seed voxels identified in the time series of the R2 star map; and - calculating a denoised time series of the R2 star maps by applying a temporal filter algorithm to the time series of the R2 star maps; calculating a correlation map for each voxel of the one or more seed voxels by calculating a pixel-by-pixel correlation value between each voxel of the one or more seed voxels in the time series of the R2 star map and the denoised time series of the R2 star map; thresholding the correlation map to determine regions of activation in the brain volume by combining voxels identified in the correlation map that have values ​​above a predetermined threshold; aligning the time series of the R2 star map with the stimulation signal to yield a hemodynamic response function for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume; averaging the hemodynamic response functions for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume to yield a subject-specific hemodynamic response function; a computer program causing the computing system to perform the above.

16. calculating the time of maximum value of the hemodynamic response function for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume; calculating statistical properties of the time of maximum value of the hemodynamic response function for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume; excluding any hemodynamic response function from the calculation of the subject-specific hemodynamic response function if that hemodynamic response function fails to meet the criteria determined using the statistical properties; The method of claim 14 further comprising:

17. 17. The method of claim 16, wherein the time of maximum value of the hemodynamic response function for each voxel and each occurrence of the sensory stimulus in the activated region of the brain volume is calculated using a smoothing function.

18. A method for detecting a brain volume of a subject, comprising: receiving EPI multi-echo T2 star-weighted k-space data describing the brain volume of the subject from multiple acquisitions, the multiple acquisitions being synchronized with the stimulation signal; receiving T1-weighted k-space data describing the brain volume of the subject; reconstructing a T1-weighted image of the brain volume from the T1-weighted k-space data; reconstructing a T2 star weighted image for each echo of the EPI multi-echo T2 star weighted k-space data; calculating a registered T2 star weighted image for each echo by performing pre-processing to register the T2 star weighted image for each echo and the T1 weighted image of the brain volume with each other; calculating the time series of R2 star maps for the subject's brain volume for each voxel by fitting an attenuation curve to the matched T2 star weighted images for each echo; The method of claim 14 further comprising:

19. The method of claim 19, wherein the calculation of the aligned T2 star-weighted image for each echo is performed by pre-processing the T2 star-weighted image for each echo of the EPI multi-echo T2 star-weighted k-space data and the T1 weighted image of the brain volume to be aligned with each other; registering the T2 star weighted image corresponding to a first echo of the EPI multi-echo T2 star weighted k-space data with the T1 weighted image; segmenting the T1 weighted image to generate a gray matter segmentation, a white matter segmentation, and a cerebrospinal fluid segmentation; using the registration between the T2 star weighted image and the T1 weighted image corresponding to the first echo, re-slicing the T1 weighted image, the gray matter segmentation, the white matter segmentation, and the cerebrospinal fluid segmentation to match the T2 star weighted image corresponding to the first echo; constructing a brain mask using the gray matter segmentation, the white matter segmentation, and the cerebrospinal fluid segmentation; realigning the T2 weighted image for each echo with the T1 weighted image; 20. The method of claim 18, comprising:

Images