Removal of background noise from images

Abstract

A computer-implemented method of removing background noise from a medical image comprising voxels, each voxel having a voxel intensity, the method comprising: generating a mask based on the medical image, where the mask comprises a foreground portion specifying a foreground voxel, a background portion specifying a background voxel, and a perimeter separating the foreground portion from the background portion; specifying a perimeter portion of the mask, the perimeter portion surrounding the perimeter, a subset of the foreground portion, and a subset of the background portion; specifying a threshold for the perimeter portion to separate the voxels based on the voxel intensities; filtering the medical image by using the mask and the threshold to obtain a filtered image; and displaying the filtered image on a display.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Application No. 63 / 411,485 filed on September 29, 2022 and U.S. Patent Application No. 18 / 373,735 filed on September 27, 2023, the contents of which are incorporated herein by reference in their entirety. Background Art

[0003] Tumor Treatment Fields (TTFields) are low-intensity alternating electric fields in the medium frequency range (e.g., 50kHz to 1MHz) that can be used to treat tumors, as described in U.S. Patent No. 7,565,205. TTFields are non-invasively directed into a region of interest by placing transducers directly on the subject's body and applying an alternating current (AC) voltage between the transducers. Traditionally, a first pair of transducers and a second pair of transducers are placed on the subject's body. An AC voltage is applied between the first pair of transducers during a first time interval to generate an electric field having field lines extending generally in a front-to-back direction. Then, an AC voltage is applied between the second pair of transducers at the same frequency during a second time interval to generate an electric field having field lines extending generally in a left-to-right direction. The system then repeats this two-step sequence throughout the treatment.

[0004] TTFields treatment planning can include segmenting tissue from background voxels on a medical image (e.g., a magnetic resonance imaging (MRI) image) for determining where to place the transducer on the subject's body and assessing the distribution and quantitative treatment efficacy of the TTFields. Manual segmentation is time consuming, and conventional computer-implemented segmentation can lack accuracy. Furthermore, using conventional methods, the large amount of data and data annotation can lead to noisy markers and intra- and inter-observer variability. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 is a flow chart depicting an example of removing background from a medical image.

[0006] Figure 2 is a flow chart depicting an example of filtering a medical image using a mask.

[0007] Figures 3A-3D An example of defining a perimeter portion is depicted.

[0008] Figures 4A-4C An example of a medical image is depicted.

[0009] Figures 5A-5C Depicted is an example of a medical image with histogram adjustment, showing the presence of background noise.

[0010] Figures 6A-6C An example of a filtered medical image is depicted.

[0011] Figures 7A-7C An example of a filtered medical image is depicted.

[0012] Figures 8A-8C Depicted is an example of a medical image utilizing high discrete level filtering.

[0013] Figures 9A-9C Depicted is an example of a medical image filtered using discrete levels close to the midpoint of the discrete levels.

[0014] Figures 10A-10C Depicted is an example of a medical image utilizing low discrete level filtering.

[0015] Fig.11 Depicted are examples of filtered images generated using multiple discrete levels.

[0016] Figures 12A-12C An example of a visualization of an exemplary mask with different discrete levels is depicted.

[0017] Fig.13A An example of a medical image is depicted, and Fig. 13B Depicted is an example of a medical image after background removal.

[0018] Fig.14 An example computer device is depicted for use with embodiments herein.

[0019] Various embodiments are described in detail below with reference to the drawings, wherein like reference numerals refer to like elements. DETAILED DESCRIPTION

[0020] In order to provide effective TTFields therapy to a subject, precise locations for placing the transducers on the subject's body must be generated and are based on, for example, the type of cancer, the size of the cancer, and the location of the cancer in the subject's body. However, determining these precise locations is challenging and is typically done through computer simulations of many possible locations for placing the transducers.

[0021] Such computer simulations are constructed from images used to model the subject (e.g., magnetic resonance imaging (MRI), computed tomography (CT), etc.). In order to perform the simulation, the computer needs to perform image segmentation to identify the target tissue from the background noise and remove the background noise. One difficulty with such image segmentation is how to accurately separate foreground voxels (e.g., voxels of the target tissue) from background voxels (e.g., noise and / or artifacts). Another difficulty is how to take into account user input to improve the accuracy and personalization of the segmentation for the subject. For example, it may be difficult to identify the boundaries of the target tissue (e.g., skin) from the surrounding background voxels because the size, texture, and shape of the target tissue may vary from subject to subject.

[0022] The inventors recognized these problems and discovered a method for removing background from a medical image, the method generating a mask based on the medical image and filtering the medical image using the mask and a threshold to remove the background from the medical image. By setting a user-adjustable perimeter of the mask and a user-adjustable threshold, the accuracy and efficiency of background removal can be improved, and thus the accuracy and efficiency of tumor treatment planning can be improved.

[0023] Figure 1 is a flow chart describing an example of a computer-implemented method 100 for removing background from a medical image. In some embodiments, the image is not limited to a medical image and can be any type of image. Certain steps of the method 100 are described as computer-implemented steps. The computer can be any device including one or more processors and a memory accessible by the one or more processors, the memory storing instructions that, when executed by the one or more processors, cause the computer to perform the relevant steps of the method 100. Although for illustration purposes only, the method 100 is described in detail below. Figure 1 An order of operations is indicated in the drawings, but the timing and order of such operations may be varied where appropriate without negating the purposes and advantages of the examples detailed throughout the disclosure.

[0024] refer to Figure 1 , at step 102, the method may include obtaining a medical image having voxels. The medical image may, for example, include at least one of a magnetic resonance imaging (MRI) image, a computed tomography (CT) image, an X-ray image, an ultrasound image, a nuclear medicine image, a positron emission tomography (PET) image, an arthrography image, a myelography image, or any image of a subject's body that provides an internal view of the subject's body. Each image may include an external shape of a portion of the subject's body and an area corresponding to an area of ​​interest (e.g., a tumor) within the subject's body. In one example, the medical image may be a three-dimensional (3D) MRI image.

[0025] At step 104, method 100 may include performing one or more pre-processing procedures on the medical image. In some embodiments, the pre-processing procedure may include at least one of Gaussian smoothing or bias correction. In one example, the bias correction is an N4 bias correction. In some embodiments, the pre-processing procedure produces a smoother and bias-free image. In some embodiments, the pre-processing procedure may include scaling the voxel intensities of voxels in the medical image to obtain a scaled medical image. In one example, the following steps 106 to 118 are performed on the scaled medical image.

[0026] At step 106, method 100 may include generating a mask based on the medical image. In some embodiments, the mask includes a foreground portion specifying foreground voxels, a background portion specifying background voxels, and a perimeter separating the foreground portion from the background portion. In some embodiments, the mask is generated by at least one of a multi-Otsu thresholding operation, a k-means clustering operation, or a morphological segmentation operation.

[0027] In some embodiments, a foreground portion having foreground voxels represents a desired tissue (e.g., a target tissue) in a medical image, and a background portion having background voxels represents one or more regions in the medical image where the desired tissue is not present. In some embodiments, the foreground portion represents a region of interest in a medical image, and the background portion represents one or more regions in the medical image that are not in the region of interest. In one example, the desired tissue is at least one of skin, bone, skull, organ, brain, or any tissue in the human body. In a more specific example, the foreground portion corresponds to a subject's head. As another example, the foreground portion corresponds to a subject's torso.

[0028] At step 108, method 100 may include specifying a perimeter portion of a mask that encompasses a perimeter, a subset of a foreground portion, and a subset of a background portion. In some embodiments, the perimeter portion includes a foreground perimeter that separates the perimeter portion from the remainder of the foreground portion, and the remainder of the foreground portion does not include the subset of the foreground portion. In some embodiments, the perimeter portion also includes a background perimeter that separates the perimeter portion from the remainder of the background portion, and the remainder of the background portion does not include the subset of the background portion. In some embodiments, the perimeter is approximately equidistant from the foreground perimeter and the background perimeter. In some embodiments, the width of the perimeter portion between the foreground perimeter and the background perimeter is approximately 5 mm, 10 mm, 15 mm, 20 mm, or 25 mm. The width of the perimeter portion between the foreground perimeter and the background perimeter may depend on the subject and / or imaging parameters. Figures 4A-4C An example of specifying a perimeter portion of a mask is shown.

[0029] At step 110, method 100 may include specifying a threshold for the perimeter portion to separate voxels based on voxel intensity. In some embodiments, the threshold is a weighted threshold. In some embodiments, the threshold is user-defined. As an example, the threshold may be a user-defined weighted threshold.

[0030] At step 112, method 100 may include filtering the medical image using the mask to obtain a filtered image. In some embodiments, filtering the medical image is based on voxel intensity. For example, for voxels in the medical image that are located in the remainder of the foreground portion of the mask, their voxel intensity remains unchanged in the filtered image. For voxels in the medical image that are located in the remainder of the background portion of the mask, background voxel intensity is assigned. In some embodiments, for voxels in the medical image that are located in the perimeter portion of the mask, voxel intensity is assigned in the filtered image based on a threshold and the position of the voxel in the perimeter portion. For example, if the voxel intensity of a particular voxel is higher than the threshold specified at step 110, the voxel intensity of the particular voxel remains unchanged in the filtered image; if the voxel intensity of the particular voxel is lower than the threshold, the voxel intensity of the particular voxel is assigned the background voxel intensity. Figure 2 An example of filtering a medical image using a mask is shown, which will be discussed further below.

[0031] In some embodiments, for voxels in a medical image located in a perimeter portion, the medical image may be filtered based on the voxel intensity and the distance of the voxel from the perimeter portion. In one example, filtering of the medical image may include averaging the voxel intensity of the voxel and the intensity assigned to the voxel based on the distance of the voxel from the perimeter portion. In one example, filtering of the medical image may include assigning an intensity to the voxel based on the distance relative to a position in the foreground portion. In other words, the voxel intensity of a corresponding position in the perimeter portion may be based on the distance of the position from the foreground perimeter of the perimeter portion or the background perimeter of the perimeter portion. In some embodiments, the intensity assigned to the voxel is related to the distance between the voxel and the position in the foreground portion. In some embodiments, the intensity assigned to the voxel is proportional to the distance between the voxel and the position in the foreground portion. In some embodiments, the intensity assigned to the voxel is proportional to the distance between the voxel and the perimeter of the perimeter portion. In some embodiments, the intensity assigned to the voxel is proportional to the distance between the voxel and the position on the foreground perimeter of the perimeter portion. As an example, if the distance relative to the position in the foreground portion is small, the intensity assigned to the voxel is large. As an example, the intensity assigned to a voxel may be an average of the intensity of the voxel and an intensity assigned to the voxel based on the distance of the voxel relative to the position in the foreground portion, and this assigned intensity may be represented by the following equation:

[0032] p 绥别=(m1+m2) / 2 Equation (1)

[0033] where p 级别 is the voxel intensity assigned to the voxel, m1 is the voxel intensity of the voxel in the medical image, and m2 is the intensity assigned to the voxel as proportional to the distance between the voxel and the position in the foreground portion.

[0034] In some embodiments, for voxels in the medical image located in the peripheral portion, filtering of the medical image includes weighted averaging of the voxel intensity of the voxel and the intensity assigned to the voxel based on the distance of the voxel from the peripheral portion. In other words, the voxels in the medical image located in the peripheral portion of the mask are assigned voxel intensities in the filtered image based on a weighted combination of the voxel intensity of the voxel in the medical image and another voxel intensity based on the position of the voxel in the peripheral portion. In some embodiments, for voxels in the medical image located in the peripheral portion, filtering of the medical image includes voxel-by-voxel weighted summation of the normalized voxel intensity of the voxel in the medical image and the voxel intensity of the corresponding position in the peripheral portion of the mask. As an example, the filtered image is generated based on the following equation:

[0035] p 级别 =w1*m1+w2*m2, w2=1-w1 Equation (2)

[0036] Wherein m1 is the voxel intensity of a voxel in the medical image, m2 is the voxel intensity of the corresponding position in the perimeter portion of the mask, and w1 and w2 are weighting parameters.

[0037] In some embodiments, the threshold value specified at step 110 is a weighted threshold value. As an example, for voxels in the medical image located in the perimeter portion whose intensity is greater than the weighted threshold value, filtering of the medical image includes assigning an intensity to the voxel based on the intensity of the voxel in the medical image or the preprocessed medical image. As an example, for voxels in the medical image located in the perimeter portion whose intensity is greater than the weighted threshold value, filtering of the medical image includes assigning an intensity to the voxel based on a weighted average of the intensities. In some embodiments, the threshold value is a user-defined weighted threshold value. In one example, method 100 may include filtering the medical image using a mask and a user-defined weighted threshold value to obtain a filtered image.

[0038] At step 114, method 100 may include performing one or more post-processing procedures on the filtered image. In some embodiments, the post-processing procedures include at least one of a morphological closing operation, a three-dimensional (3D) hole filling operation, or a smoothing operation. Step 114 may be performed optionally.

[0039] At step 116, method 100 may include displaying the filtered image. As an example, the filtered image generated at step 112 is displayed. As an example, the filtered image generated at step 114 is displayed. In some embodiments, the method further includes displaying voxel intensities of the medical image and / or displaying weighted average voxel intensities of voxels based on the distance of the voxels from the perimeter portion.

[0040] At step 118, the method 100 may include adjusting the size and / or threshold of the perimeter portion. In some embodiments, the method includes receiving user input to adjust the filtering parameters to obtain adjusted filtering parameters, wherein the adjusted filtering parameters include at least one of an adjusted size of the perimeter portion or an adjusted threshold of the perimeter. As an example, the method includes adjusting the size of the perimeter portion to obtain a modified mask. As an example, the method includes adjusting the threshold to obtain a modified threshold. As an example, the threshold is a weighted threshold. As discussed above in step 112, for voxels in the medical image located in the perimeter portion of the mask, a voxel intensity is assigned in the filtered image based on the threshold and the location of the voxel in the perimeter portion. For example, if the voxel intensity is above the threshold, the voxel intensity remains unchanged in the filtered image; if the voxel intensity is below the threshold, the assigned voxel intensity is the background voxel intensity. In this way, the user can adjust the threshold to adjust the perimeter (e.g., skin boundary) in the filtered image. As an example, when the user adjusts the threshold, the user can be able to see the modified filtered image in real time. As an example, the method may include adjusting a size of the perimeter portion to obtain a modified mask, and adjusting a threshold to obtain a modified threshold.

[0041] In some embodiments, one or more users (e.g., physicians, nurses, assistants, staff, physicists, dosimetrists, etc.) can use a user interface to adjust the perimeter portion of the mask and / or the threshold value through user-adjustable levels. In some embodiments, the user-adjustable level is an interactive slider in the user interface, and the interactive slider defines a removal area close to the outer surface (e.g., perimeter) of the tissue in the medical image. As an example, the interactive slider has discrete levels that are user-adjustable.

[0042] In some embodiments, the interactive slider (or other user interface) may be user adjustable to adjust the discrete levels and / or select the level of the perimeter portion within a predetermined range of discrete levels. In some embodiments, the user adjustable discrete levels have an initial setting. As an example, the user adjustable discrete levels may be user adjustable from the initial setting to increase the size of the perimeter portion, and may be user adjustable from the initial setting to decrease the size of the perimeter portion. As an example, the initial setting is the perimeter of a mask that separates foreground voxels from background voxels. As an example, the user adjustable discrete levels have between about 8 and about 512 discrete levels, wherein the initial setting of the user adjustable levels is approximately the middle of the discrete levels (e.g., the perimeter of the mask). As an example, the user adjustable discrete levels have 255 discrete levels, and the initial setting of the user adjustable levels is 126. In this example, the value 126 defines the perimeter of the mask that separates foreground voxels from background voxels. Figure 3D An example of user adjustable discrete levels is shown, which will be discussed further below.An example of these embodiments regarding user customized perimeter portions is shown in FIG. 12, which will be discussed further below.

[0043] After adjusting the filtering parameters, the flow of method 100 continues to step 112 and repeats the cycle. In some embodiments, the method further includes filtering the medical image using the modified mask and / or the modified threshold to obtain a modified filtered image, and displaying the modified filtered image on a display.

[0044] In some embodiments, method 100 further comprises generating and outputting one or more recommendations regarding locations on the subject's body based on the filtered image to place one or more transducers for applying the tumor treatment field to the subject's body. In some embodiments, the selection of locations to place the one or more transducers can be further based on, for example, a region of interest of the subject's body corresponding to the tumor.

[0045] Figure 2 is a flowchart depicting an example of filtering a medical image using a mask. Certain steps of method 200 are described as computer-implemented steps. A computer may be any device including one or more processors and a memory accessible by the one or more processors, the memory storing instructions that, when executed by the one or more processors, cause the computer to perform the relevant steps of method 200. Although for illustration purposes, Figure 2 An order of operations is indicated in the drawings, but the timing and order of such operations may be varied where appropriate without negating the purposes and advantages of the examples detailed throughout the disclosure.

[0046] At step 202, for voxels in the remainder of the foreground portion, filtering the medical image includes maintaining the voxel intensity of the voxels. At step 204, for voxels in the remainder of the background portion, filtering the medical image includes assigning background voxel intensities to the voxels.

[0047] For voxels in the perimeter portion, if the voxel intensity is above the threshold, the voxel intensity of the voxel is maintained at step 206, and if the voxel intensity is below the threshold, the voxel is assigned a background voxel intensity at step 208. In one example, the threshold is a weighted threshold.

[0048] In some embodiments, when the voxel intensity is higher than the threshold value as in step 206, the voxel intensity is set to be the same as the voxel intensity in the medical image. In other words, when the voxel intensity is higher than the threshold value, the voxel intensity of the medical image is maintained. In some embodiments, when the voxel intensity is higher than the threshold value, the voxel intensity is set to the voxel intensity of the image after the post-processing procedure. In some embodiments, when the voxel intensity is higher than the threshold value, the voxel intensity is set to the average value of the intensity of the voxel in the medical image and the intensity based on the distance of the voxel from the peripheral portion calculated by the above equation (1). In some embodiments, when the voxel intensity is higher than the threshold value, the voxel intensity is set to the weighted average value of the intensity of the voxel in the medical image and the intensity based on the distance of the voxel from the peripheral portion calculated by the above equation (2).

[0049] Figures 3A-3D An example of defining a perimeter portion is depicted. Figure 3A A mask image generated based on a medical image or a preprocessed medical image is depicted. In this example, the mask includes a foreground portion 302 specifying foreground voxels, a background portion 304 specifying background voxels, and a perimeter 306 separating the foreground portion from the background portion. The foreground portion 302 having foreground voxels represents desired tissue in the medical image, while the background portion 304 having background voxels represents one or more areas in the medical image where there is no desired tissue. Figure 3A In the depicted example, foreground portion 302 represents the subject's head.

[0050] Figure 3B and 3CAn example of a perimeter portion 312 of a specified mask is depicted. Perimeter portion 312 encompasses perimeter 306, a subset of foreground portions (e.g., a portion between foreground perimeter 308 and perimeter 306), and a subset of background portions (e.g., a portion between background perimeter 310 and perimeter 306). Perimeter portion 312 includes foreground perimeter 308 separating perimeter portion 312 from the remainder of the foreground portions, wherein the remainder of the foreground portions does not include the subset of the foreground portions. Additionally, perimeter portion 312 also includes background perimeter 310 separating perimeter portion 312 from the remainder of the background portions, wherein the remainder of the background portions does not include the subset of the background portions. As an example, perimeter 306 is approximately equidistant from foreground perimeter 308 and background perimeter 310. As an example, the width of perimeter portion 312 between foreground perimeter 308 and background perimeter 310 is approximately 5 mm, 10 mm, 15 mm, 20 mm, or 25 mm. In some embodiments, the perimeter portion of the mask is user-defined. As an example, a user interface with an interactive slider can be provided to specify the perimeter portion.

[0051] Figure 3D An example of adjusting a perimeter portion using a user interface 318 to adjust a discrete level of the perimeter portion is depicted. In this example, the user interface 318 may include a plurality of discrete levels 320 for a user to select from and a slider to assist the user in selecting the discrete level. Thus, the user may use the slider to increase or decrease the size of the perimeter portion. As an example, the lowest setting of the discrete levels may correspond to the background perimeter 310, while the highest setting of the discrete levels may correspond to the foreground perimeter 308.

[0052] As an example, adjusting the slider 318 may move both the foreground perimeter 308 and the background perimeter 310. For this example, the slider 318 may have two sliders: a first slider 322 for moving the background perimeter 310 and a second slider 324 for moving the foreground perimeter 308. Figure 3D The first slider 322 is shown in the position, the background perimeter 310 is moved to the position 316, and for Figure 3D 314. With the position of the second slider 324 shown, the foreground perimeter 308 is moved to position 314. Thus, the perimeter portion is adjusted between positions 316 and 314.

[0053] As an example, adjusting the slider 318 may move the background perimeter 310 but keep the foreground perimeter 308 unchanged. For this example, the slider 318 may have only one slider: a first slider 322 for moving the background perimeter 310. Figure 3D In the illustrated position of the first slider 322, the background perimeter 310 is moved to the position 316, while the foreground perimeter 308 is not moved. Thus, the perimeter portion is adjusted between the positions 316 and 308.

[0054] As an example, adjusting the slider 318 may move the foreground perimeter 308, but keep the background perimeter 310 unchanged. For this example, the slider 318 may have only one slider: a second slider 324 for moving the foreground perimeter 308. Figure 3D In the position of the second slider 324 shown, the foreground perimeter 308 is moved to the position 314, while the background perimeter 310 is not moved. In this way, the perimeter portion is adjusted to be between the positions 314 and 310.

[0055] Figures 4A-4C Examples of medical images are depicted. In these examples, the medical image is an MRI image of a subject's head. Figure 4A An MRI image of a subject's head at an axial angle is shown; Figure 4B showing an MRI image of the subject's head at a sagittal angle; and Figure 4C An MRI image of a subject's head at a coronal angle is shown.

[0056] Figures 5A-5C Depicts a histogram with adjustment Figures 4A-4C An example of a medical image is shown, showing the presence of background noise.

[0057] Figures 6A-6C Depicts filtering according to an exemplary embodiment of the present invention. Figures 4A-4C An example of a medical image is shown. Figures 6A-6C In the depicted example, a medical image is filtered using a mask that includes a perimeter that separates voxels considered to be foreground voxels from voxels considered to be background voxels. When filtering using the mask, voxels considered to be foreground voxels retain their values, while voxels considered to be background voxels are assigned a constant background value (here "0", corresponding to black). Thus, in these filtered medical images, foreground voxels are depicted in their retained gray color, while background voxels are depicted in black. The mask is based on Figure 1 generated in step 106.

[0058] Figures 7A-7C Depicts filtering according to an exemplary embodiment of the present invention. Figures 4A-4C Another example of a medical image is shown. Figures 6A-6CSimilar to the depicted example, a medical image is filtered using a mask that includes a perimeter that separates voxels considered to be foreground voxels from voxels considered to be background voxels. When filtering using the mask, voxels considered to be foreground voxels retain their values, except that voxels within the perimeter that have a background color (e.g., "0" for black) are assigned values ​​that are unrelated to the background color (e.g., non-zero values ​​that are not black). Voxels considered to be background voxels are assigned a constant background value (here "0", corresponding to black). Thus, in these filtered medical images, foreground voxels are depicted as the gray color they retain or are adjusted to not be the background color, while background voxels are depicted as black. The mask is based on Figure 1 generated in step 106.

[0059] Figures 8A-10C Depicted are examples of medical images filtered using different user-adjustable discrete levels for adjusting the size of the perimeter portion of the mask. In these examples, foreground voxels at different angles of a subject's head are shown in gray, while background voxels are filtered to appear black. Figures 8A-10C In the depicted example, there are 255 user-adjustable discrete levels. The user-adjustable levels adjust the size of the perimeter portion of the mask. The user-adjustable levels can increase the size of the perimeter portion by lowering the discrete levels, and can decrease the size of the perimeter portion by raising the discrete levels. In this example, the background perimeter of the perimeter portion (similar to Figure 3D The background perimeter 308 in ) remains constant, while the foreground perimeter (similar to Figure 3D The foreground perimeter 310 in FIG. 1 is user adjustable at 255 discrete levels, where level 255 makes the foreground perimeter closest to the background perimeter and level 1 makes the foreground perimeter farthest from the background perimeter.

[0060] Figures 8A-8C An example of a medical image using high discrete level filtering is depicted. In this example, the resulting filtered image has 223 discrete levels. Figures 8A-8C As shown, the foreground voxels (grey) are reduced in size and the background voxels (black) encroach upon the perimeter of the foreground voxels, such as in portion 801 .

[0061] Figures 9A-9C An example of a medical image filtered using discrete levels close to the midpoint of the discrete levels is depicted. In this example, the resulting filtered image has 172 discrete levels. Figures 9A-9C As shown, the perimeter of the foreground voxels (in grey) is sharp, and the background voxels (in black) are removed with little error.

[0062] Figures 10A-10CAn example of a medical image using low discrete level filtering is depicted. In this example, the resulting filtered image has 18 discrete levels. Figures 10A-10C As shown, the size of the foreground voxels (grey) is increased so that the perimeter of the foreground voxels includes unwanted background voxels, such as in portion 1001 .

[0063] Fig.11 Depicts an example of a filtered image generated using multiple discrete levels. Fig.11 In the depicted example, the foreground voxels include voxels of the subject's head and the perimeter of the subject's head. The perimeter portion is the area of ​​the subject's head from the lowest discrete level to the highest discrete level. The perimeter close to the inside of the subject's head has a higher discrete level, while the perimeter close to the background voxels has a lower discrete level. The discrete levels 1101 within the perimeter portion are displayed in different grays.

[0064] Figures 12A-12C Depicted are examples of visualizations of exemplary masks with different discrete levels. Figures 12A-12C In the depicted example, the portion of the mask of voxels considered to be foreground voxels is depicted in the same grey colour, whereas the portion of the mask of voxels considered to be background voxels is the background colour (here black). Figures 12A-12C As shown, the discrete levels of the perimeter portion of the mask capture the varying characteristics of the subject, as indicated by the varying grey values.

[0065] Fig.13A An example of a medical image is depicted, and Fig. 13B An example of a medical image after background removal using a mask according to an exemplary embodiment is depicted. In this example, the mask has 126 discrete levels out of 255 discrete levels. It can be seen that the resulting filtered medical image clearly indicates the perimeter of the subject's head.

[0066] Fig.14 An example computer device for use with embodiments herein is depicted. As an example, device 1400 may be a computer to implement certain inventive techniques disclosed herein, such as removing background noise from medical images. For example, Figure 1 and 2 The method may be performed by a computer such as the apparatus 1400. The apparatus 1400 may include one or more processors 1402, a memory 1403, one or more input devices, and one or more output devices 1405.

[0067] In one example, based on input 1401, one or more processors remove background from an image according to embodiments of the present invention. In one example, input 1401 is user input. In another example, input 1401 may come from another computer in communication with apparatus 1400. Input 1401 may be received together with one or more input devices (not shown) of apparatus 1400.

[0068] The memory 1403 may be accessible by the one or more processors 1402 (e.g., via the link 1404), such that the one or more processors 1402 may read information from and write information to the memory 1403. The memory 1403 may store instructions that, when executed by the one or more processors 1402, implement one or more embodiments of the present invention. The memory 1403 may be a non-transitory computer-readable medium (or a non-transitory processor-readable medium) having a set of instructions for removing background noise from a medical image embodied thereon, wherein the instructions, when executed by a processor (such as the one or more processors 1402), cause the processor to perform one or more methods disclosed herein.

[0069] One or more output devices 1405 may provide the state of the art of computer implementation herein. One or more output devices 1405 may provide visualization data, such as medical images, masks, filtered images, and / or voxel intensities of medical images according to certain embodiments of the present invention. One or more output devices 1405 may display user adjustable levels, which may be controlled using input 1401.

[0070] Device 1400 may be a device for removing background noise from a medical image, the device comprising: one or more processors (such as one or more processors 1402); and a memory (such as memory 1403) accessible by the one or more processors, the memory storing instructions that, when executed by the one or more processors, cause the device to perform one or more methods disclosed herein.

[0071] Illustrative Embodiments

[0072] The invention includes other illustrative embodiments, such as the following.

[0073] Illustrative embodiments 1. A computer-implemented method for removing background noise from a medical image comprising voxels, each voxel having a voxel intensity, the method comprising: generating a mask based on the medical image, wherein the mask comprises a foreground portion of a specified foreground voxel, a background portion of a specified background voxel, and a perimeter separating the foreground portion from the background portion; specifying a perimeter portion of the mask, the perimeter portion enclosing the perimeter, a subset of the foreground portion, and a subset of the background portion; specifying a threshold for the perimeter portion to separate the voxels based on the voxel intensity; filtering the medical image using the mask and the threshold to obtain a filtered image; and displaying the filtered image on a display.

[0074] Illustrative embodiment 2. A method according to illustrative embodiment 1, wherein after filtering the medical image using a mask: voxels in the medical image located in the remainder of the foreground portion of the mask retain their voxel intensity in the filtered image, voxels in the medical image located in the remainder of the background portion of the mask are assigned background voxel intensity, if the voxel intensity is above a threshold, then voxels in the medical image located in the peripheral portion of the mask retain their voxel intensity in the filtered image, and if the voxel intensity is below a threshold, then voxels in the medical image located in the peripheral portion of the mask are assigned background voxel intensity in the filtered image.

[0075] Illustrative embodiment 3. The method according to illustrative embodiment 1, wherein voxels in the medical image located in the perimeter portion of the mask are assigned voxel intensities in the filtered image based on a threshold and the position of the voxel in the perimeter portion.

[0076] Illustrative embodiment 4. The method according to illustrative embodiment 1, wherein for a voxel in the medical image located in the peripheral portion, filtering of the medical image includes averaging the voxel intensity of the voxel and the intensity assigned to the voxel based on the distance of the voxel from the peripheral portion.

[0077] Illustrative embodiment 5. The method according to illustrative embodiment 1, wherein for voxels in the medical image located in the perimeter portion, the filtering of the medical image includes assigning an intensity to the voxel based on a distance relative to a location in the foreground portion.

[0078] Illustrative embodiment 6. The method according to illustrative embodiment 5, wherein the intensity assigned to the voxel is larger if the distance relative to the position in the foreground portion is smaller.

[0079] Illustrative embodiment 7. A method according to illustrative embodiment 1, wherein for a voxel in the medical image located in the peripheral portion, filtering of the medical image includes weighted averaging the voxel intensity of the voxel and the intensity assigned to the voxel based on the distance of the voxel from the peripheral portion.

[0080] Illustrative embodiment 8. A method according to illustrative embodiment 1, wherein voxels in the medical image located in the peripheral portion of the mask are assigned voxel intensities in the filtered image based on a weighted combination of the voxel intensity of the voxels in the medical image and another voxel intensity based on the position of the voxel in the peripheral portion.

[0081] Illustrative embodiment 9. A method according to illustrative embodiment 1, wherein for voxels in the medical image located in the peripheral portion, filtering of the medical image includes a voxel-by-voxel weighted summation of the normalized voxel intensity of the voxel and the voxel intensity of the corresponding position in the peripheral portion.

[0082] Illustrative embodiment 10. A method according to illustrative embodiment 8, wherein the perimeter portion includes a foreground perimeter separating the perimeter portion from the rest of the foreground portion, wherein the perimeter portion includes a background perimeter separating the perimeter portion from the rest of the background portion, wherein the voxel intensity of a corresponding position in the perimeter portion is based on the distance of the position from the foreground perimeter of the perimeter portion or the background perimeter of the perimeter portion.

[0083] Illustrative embodiment 11. A method according to illustrative embodiment 1, wherein the method further comprises: scaling the voxel intensities of voxels in the medical image to obtain a scaled medical image; and assigning the scaled voxel intensities to the voxels in the perimeter portion of the mask based on the distance of the voxels in the mask from the foreground perimeter or the background perimeter, wherein the foreground perimeter separates the perimeter portion from the rest of the foreground portion, and the background perimeter separates the perimeter portion from the rest of the background portion, wherein filtering the medical image using the mask comprises: filtering the scaled medical image using the mask and a threshold; for the scaled voxels in the rest of the foreground portion of the mask, For voxels in the medical image, maintaining the voxel intensity of the voxel in the filtered image; for voxels in the scaled medical image located in the rest of the background portion of the mask, assigning the background voxel intensity to the voxels in the filtered image; for voxels in the scaled medical image located in the perimeter portion of the mask: if the voxel intensity is above a threshold, assigning the voxel intensity to the voxel in the filtered image based on a weighted combination of the scaled voxel intensity of the scaled medical image and the scaled voxel intensity of the voxels in the perimeter portion of the mask; and if the voxel intensity is below the threshold, assigning the background voxel intensity to the voxel in the filtered image.

[0084] Illustrative embodiment 12. The method according to illustrative embodiment 1, wherein the threshold is a user-defined weighted threshold, and wherein filtering the medical image comprises filtering the medical image using a mask and the user-defined weighted threshold to obtain a filtered image.

[0085] Illustrative embodiment 13. A method according to illustrative embodiment 1, wherein the threshold is a weighted threshold, and wherein for voxels in the medical image located in the peripheral portion whose intensity is greater than the weighted threshold, filtering the medical image includes assigning an intensity to the voxel based on the intensity of the voxel in the medical image.

[0086] Illustrative embodiment 14. The method according to illustrative embodiment 1 also includes preprocessing the medical image before generating the mask to obtain a preprocessed medical image, wherein the threshold is a weighted threshold, and wherein for voxels in the medical image located in the peripheral portion whose intensity is greater than the weighted threshold, filtering of the medical image includes assigning an intensity to the voxel based on the intensity of the voxel in the preprocessed medical image.

[0087] Illustrative embodiment 15. A method according to illustrative embodiment 1, wherein the threshold is a weighted threshold, and wherein for voxels in the medical image located in the peripheral portion whose intensity is greater than the weighted threshold, filtering the medical image includes assigning an intensity to the voxel based on a weighted average of the intensities.

[0088] Illustrative embodiment 16. A method according to illustrative embodiment 1, wherein the perimeter portion includes: a foreground perimeter separating the perimeter portion from the rest of the foreground portion, the rest of the foreground portion not including a subset of the foreground portion; and the perimeter portion includes a background perimeter separating the perimeter portion from the rest of the background portion, the rest of the background portion not including a subset of the background portion.

[0089] Illustrative embodiment 17. The method of illustrative embodiment 16, wherein the perimeter is approximately equidistant from the foreground perimeter and the background perimeter.

[0090] Illustrative embodiment 18. The method of illustrative embodiment 16, wherein a width of the perimeter portion between the foreground perimeter and the background perimeter is approximately 10 mm.

[0091] Illustrative embodiment 19. The method of illustrative embodiment 1, wherein the perimeter portion of the mask is user-defined, and the threshold value of the perimeter portion is user-defined.

[0092] Illustrative embodiment 20. The method of illustrative embodiment 1, further comprising providing a user interface with an interactive slider to specify the perimeter portion.

[0093] Illustrative embodiment 21. The method according to illustrative embodiment 1, wherein the method further comprises: displaying voxel intensities of the medical image and / or displaying weighted average voxel intensities of voxels based on the distances of the voxels from the peripheral portion.

[0094] Illustrative embodiment 22. A method according to illustrative embodiment 1, wherein after displaying the image, the method further comprises: receiving user input to adjust the filtering parameters to obtain adjusted filtering parameters, wherein the adjusted filtering parameters include at least one of an adjusted size of the perimeter portion or an adjusted threshold of the perimeter portion; filtering the medical image using the adjusted filtering parameters to obtain a modified filtered image; and displaying the modified filtered image on a display.

[0095] Illustrative embodiment 23. A method according to illustrative embodiment 1, wherein after displaying the image, the method further comprises: adjusting the size of the perimeter portion to obtain a modified mask; filtering the medical image using the modified mask and a threshold to obtain a modified filtered image; and displaying the modified filtered image on a display.

[0096] Illustrative embodiment 24. A method according to illustrative embodiment 1, wherein after displaying the image, the method further comprises: adjusting the threshold to obtain a corrected threshold; filtering the medical image using the mask and the corrected threshold to obtain a corrected filtered image; and displaying the corrected filtered image on a display.

[0097] Illustrative embodiment 25. A method according to illustrative embodiment 1, wherein after displaying the image, the method further comprises: adjusting the size of the perimeter portion to obtain a corrected mask; adjusting the threshold to obtain a corrected threshold; filtering the medical image using the corrected mask and the corrected threshold to obtain a corrected filtered image; and displaying the corrected filtered image on a display.

[0098] Illustrative embodiment 26. The method according to illustrative embodiment 1, wherein the foreground portion represents desired tissue in the medical image, and wherein the background portion represents one or more regions in the medical image where the desired tissue is absent.

[0099] Illustrative embodiment 27. The method according to illustrative embodiment 1, wherein the foreground portion represents a region of interest in the medical image, and wherein the background portion represents one or more regions in the medical image that are not in the region of interest.

[0100] Illustrative embodiment 28. The method according to illustrative embodiment 1, wherein the mask is generated by at least one of a multi-Otsu thresholding operation, a k-means clustering operation, or a morphological segmentation operation.

[0101] Illustrative embodiment 29. The method according to illustrative embodiment 1 also includes: performing a post-processing procedure on the filtered image, wherein the post-processing procedure includes at least one of a morphological closing operation, a three-dimensional (3D) hole filling operation or a smoothing operation.

[0102] Illustrative embodiment 30. The method according to illustrative embodiment 1, further comprising: before generating the mask, performing a pre-processing procedure on the medical image, wherein the pre-processing procedure includes at least one of Gaussian smoothing or bias correction.

[0103] Illustrative embodiment 31. The method according to illustrative embodiment 1, wherein the medical image includes at least one of a magnetic resonance imaging image, an ultrasound image, a computed tomography image, or an X-ray image.

[0104] Illustrative embodiment 32. The method according to illustrative embodiment 1 also includes: generating and outputting one or more suggestions about locations on the subject's body to place one or more transducers for applying a tumor treatment field to the subject's body based on the filtered image.

[0105] Illustrative embodiment 33. A computer-implemented method for processing a medical image, the computer comprising one or more processors and a memory accessible by the one or more processors, the memory storing instructions that, when executed by the one or more processors, cause the computer to perform the method, the method comprising: generating a mask based on the medical image, wherein the mask comprises a foreground portion of designated foreground voxels, a background portion of designated background voxels, and a perimeter separating the foreground portion from the background portion; specifying a perimeter portion of the mask, the perimeter portion enclosing the perimeter, a subset of the foreground portion, and a subset of the background portion; specifying a threshold for the perimeter portion to separate the voxels based on voxel intensity; filtering the medical image using the mask and the threshold to obtain a filtered image; receiving user input to adjust filtering parameters to obtain adjusted filtering parameters, wherein the adjusted filtering parameters include at least one of an adjusted size of the perimeter portion or an adjusted threshold of the perimeter portion; filtering the medical image using the adjusted filtering parameters to obtain a modified filtered image; and displaying the modified filtered image on a display.

[0106] Illustrative embodiment 34. An apparatus for removing background noise from a medical image, comprising one or more processors; and a memory storing processor-executable instructions, which, when executed by the one or more processors, cause the apparatus to: generate a mask based on the medical image, wherein the mask includes a foreground portion of specified foreground voxels, a background portion of specified background voxels, and a perimeter separating the foreground portion from the background portion; specify a perimeter portion of the mask that encloses a perimeter, a subset of the foreground portion, and a subset of the background portion, wherein the perimeter portion of the mask is user-defined; specify a threshold for the perimeter portion to separate voxels based on voxel intensity, wherein the threshold for the perimeter portion is user-defined; filter the medical image using the mask and the threshold to obtain a filtered image; and display the filtered image on a display.

[0107] Illustrative embodiment 35. The device of illustrative embodiment 34, comprising a user interface having an interactive slider defining a perimeter portion.

[0108] Embodiments described under any heading of the disclosure or in any section of the disclosure may be combined with embodiments described under the same or any other heading or other section of the disclosure unless otherwise indicated herein or clearly contradicted by context.

[0109] Numerous modifications, changes and variations may be made to the described embodiments without departing from the scope of the invention as defined by the claims. It is intended that the present invention not be limited to the described embodiments, but rather have the full scope defined by the language of the following claims and their equivalents.

Claims

1. A computer-implemented method for removing background noise from a medical image comprising voxels, each voxel having a voxel intensity, the method comprising: generating a mask based on the medical image, wherein the mask includes a foreground portion specifying foreground voxels, a background portion specifying background voxels, and a perimeter separating the foreground portion from the background portion; specifying a perimeter portion of the mask, the perimeter portion encompassing the perimeter, a subset of the foreground portion, and a subset of the background portion; assigning a threshold value to the perimeter portion to separate voxels based on voxel intensity; filtering the medical image using the mask and the threshold to obtain a filtered image; as well as The filtered image is displayed on a display.

2. The method of claim 1 , wherein after filtering the medical image using the mask: voxels in the medical image that are in the remainder of the foreground portion of the mask retain their voxel intensities in the filtered image, voxels in the medical image that are in the remainder of the background portion of the mask are assigned a background voxel intensity, If the voxel intensity is above the threshold, a voxel in the medical image located in the perimeter portion of the mask retains its voxel intensity in the filtered image, and If the voxel intensity is below the threshold, voxels in the medical image that are located in the perimeter portion of the mask are assigned the background voxel intensity in the filtered image.

3. The method according to claim 1, wherein, for a voxel in the medical image located in the peripheral portion, filtering of the medical image comprises averaging a voxel intensity of the voxel and an intensity assigned to the voxel based on a distance of the voxel from the peripheral portion. 4 . The method of claim 1 , wherein, for voxels in the medical image located in the perimeter portion, filtering of the medical image comprises assigning an intensity to the voxel based on a distance relative to a location in the foreground portion.

5. A method according to claim 1, wherein voxels in the medical image located in the peripheral portion of the mask are assigned voxel intensities in the filtered image based on a weighted combination of the voxel intensity of the voxel in the medical image and another voxel intensity based on the position of the voxel in the peripheral portion.

6. The method of claim 5, wherein the perimeter portion comprises a foreground perimeter separating the perimeter portion from a remainder of the foreground portion, wherein the perimeter portion includes a background perimeter separating the perimeter portion from a remainder of the background portion, Wherein the voxel intensity of a corresponding position in the perimeter portion is based on a distance of the position from the foreground perimeter of the perimeter portion or the background perimeter of the perimeter portion.

7. The method according to claim 1, wherein the method further comprises: scaling the voxel intensities of the voxels in the medical image to obtain a scaled medical image; as well as assigning scaled voxel intensities to voxels in the perimeter portion of the mask based on a distance of the voxels in the mask from a foreground perimeter that separates the perimeter portion from a remainder of the foreground portion or a background perimeter that separates the perimeter portion from a remainder of the background portion, The filtering of the medical image using the mask comprises: filtering the scaled medical image using the mask and the threshold; for voxels in the scaled medical image that are in the remaining portion of the foreground portion of the mask, maintaining voxel intensities of the voxels in the filtered image; for voxels in the scaled medical image that are in the remaining portion of the background portion of the mask, assigning a background voxel intensity to the voxel in the filtered image; For a voxel in the scaled medical image that is located in the perimeter portion of the mask: If the voxel intensity is above the threshold, assigning a voxel intensity to the voxel in the filtered image based on a weighted combination of the scaled voxel intensity of the scaled medical image and the scaled voxel intensity of the voxels in the perimeter portion of the mask; and If the voxel intensity is below the threshold, the voxel in the filtered image is assigned the background voxel intensity.

8. The method according to claim 1, wherein the threshold is a user-defined weighted threshold, The filtering of the medical image includes filtering the medical image using the mask and the user-defined weighted threshold to obtain the filtered image.

9. The method of claim 1, wherein the threshold is a weighted threshold, Wherein, for voxels in the medical image located in the peripheral portion whose intensities are greater than the weighted threshold, the filtering of the medical image includes assigning intensities to the voxels based on the intensities of the voxels in the medical image.

10. The method of claim 1, wherein the perimeter portion comprises: a foreground perimeter separating the perimeter portion from a remainder of the foreground portion, the remainder of the foreground portion not including the subset of the foreground portion; and The perimeter portion includes a background perimeter separating the perimeter portion from a remainder of the background portion, the remainder of the background portion not including the subset of the background portion.

11. The method according to claim 1, wherein after displaying the image, the method further comprises: receiving a user input to adjust a filtering parameter, thereby obtaining an adjusted filtering parameter, wherein the adjusted filtering parameter comprises at least one of an adjusted size of the perimeter portion or an adjusted threshold of the perimeter portion; filtering the medical image using the adjusted filtering parameters to obtain a modified filtered image; as well as The modified filtered image is displayed on the display.

12. The method according to claim 1, further comprising: A post-processing procedure is performed on the filtered image, wherein the post-processing procedure includes at least one of a morphological closing operation, a three-dimensional (3D) hole filling operation, or a smoothing operation.

13. The method according to claim 1, further comprising: One or more recommendations regarding locations on a subject's body to place one or more transducers for applying a tumor treatment field to the subject's body are generated and output based on the filtered images.

14. A computer-implemented method of processing a medical image, the computer comprising one or more processors and a memory accessible by the one or more processors, the memory storing instructions which, when executed by the one or more processors, cause the computer to perform the method, the method comprising: generating a mask based on the medical image, wherein the mask includes a foreground portion specifying foreground voxels, a background portion specifying background voxels, and a perimeter separating the foreground portion from the background portion; specifying a perimeter portion of the mask, the perimeter portion encompassing the perimeter, a subset of the foreground portion, and a subset of the background portion; assigning a threshold value to the perimeter portion to separate voxels based on voxel intensity; filtering the medical image using the mask and the threshold to obtain a filtered image; receiving a user input to adjust a filtering parameter, thereby obtaining an adjusted filtering parameter, wherein the adjusted filtering parameter comprises at least one of an adjusted size of the perimeter portion or an adjusted threshold of the perimeter portion; filtering the medical image using the adjusted filtering parameters to obtain a modified filtered image; as well as The modified filtered image is displayed on the display.

15. An apparatus for removing background noise from a medical image, comprising one or more processors; and a memory storing processor-executable instructions that, when executed by the one or more processors, cause the apparatus to: generating a mask based on the medical image, wherein the mask includes a foreground portion specifying foreground voxels, a background portion specifying background voxels, and a perimeter separating the foreground portion from the background portion; specifying a perimeter portion of the mask, the perimeter portion enclosing the perimeter, a subset of the foreground portion, and a subset of the background portion, wherein the perimeter portion of the mask is user-defined; specifying a threshold value for the perimeter portion to separate voxels based on voxel intensity, wherein the threshold value for the perimeter portion is user-defined; filtering the medical image using the mask and the threshold to obtain a filtered image; as well as The filtered image is displayed on a display.

Images