Automatic generation of regions of interest in medical images for making tumor therapy electric field therapy plans

Abstract

A computer-implemented method for formulating a treatment plan for administering a tumor treatment electric field to a subject, the method comprising: presenting a slice on a display through a medical image of the subject, where the medical image comprises voxels; and determining a segmentation of a minimum voxel number of at least one tissue type in the slice of the medical image; receiving a user selection to automatically generate a region of interest (ROI) in the medical image for applying a tumor therapy electric field to the subject; and automatically generating the region of interest in the medical image for applying a tumor therapy electric field to the subject based on the segmentation of the minimum voxel number of at least one tissue type in the slice of the medical image.

Description

[0001] Cross-references to related applications This application claims priority to U.S. Provisional Application No. 63 / 609,202, filed December 12, 2023, and U.S. Provisional Application No. 63 / 524,470, filed June 30, 2023, the entire contents of each of these applications are incorporated herein by reference. This application also relates to U.S. Patent Application No. 18 / 750,582, filed June 21, 2024; U.S. Patent Application No. 18 / 675,714, filed May 28, 2024; and U.S. Provisional Application No. 63 / 523,853, filed June 28, 2023, the entire contents of each of these applications are incorporated herein by reference. Technical Background

[0002] Tumor therapeutic electric fields (TTFields) are low-intensity alternating electric fields in the mid-frequency range (e.g., 50 kHz to 1 MHz) that can be used to treat tumors, as described in U.S. Patent No. 7,565,205. In current commercial systems, TTFields are non-invasively sensed to the region of interest by applying an alternating current (AC) voltage between electrode assemblies (e.g., capacitively coupled electrode arrays, also referred to as electrode arrays, transducer arrays, or simply “transducers”) placed on the patient’s body. 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 with field lines extending generally in a front-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 with field lines extending generally in a left-right direction. The system then repeats this two-step sequence throughout the treatment process. Attached Figure Description

[0003] Figure 1 This is a flowchart depicting an example computer implementation of a method for developing a treatment plan for administering TTFields to a subject.

[0004] Figure 2 This is a flowchart depicting an example computer-implemented method for automatically generating ROIs in medical images of subjects.

[0005] Figure 3 This is a flowchart depicting an example computer implementation of a method for generating a transducer layout for applying TTFields to a subject.

[0006] Figure 4 An example user interface for an application used to manually segment medical images of subjects is depicted.

[0007] Figure 5 An example user interface for an application for automatically generating a region of interest in a medical image of a subject is depicted.

[0008] Figure 6 An example user interface for an alert for automatically generating a region of interest in a medical image of a subject is depicted.

[0009] FIG. 7A, FIG. 7B, and Figure 7C Various stages of automatically generating a ROI in an example medical image of a subject are depicted.

[0010] Figure 8 An example system for applying an alternating electric field to a subject is depicted.

[0011] Figure 9 An example placement of a transducer on a head of a subject is depicted.

[0012] Figure 10 An example computer device in accordance with one or more embodiments described herein is depicted.

[0013] Various embodiments are described in detail below with reference to the attached drawing figures, wherein the same elements exhibit the same reference numerals throughout the several views. DETAILED DESCRIPTION

[0014] This application describes example techniques for formulating a treatment plan for administering TTFields to a subject.

[0015] When formulating a treatment plan for administering TTFields to a subject, it can be necessary to identify a region of interest (ROI) on the subject to determine a dosage of TTFields to administer to the subject. Typically, a user reviews a medical image of the subject to manually identify the ROI on the subject, where the medical image of the subject can have multiple slices and even more voxels. As an example, the user can review each slice of the medical image of the subject and manually identify the ROI in each slice. The ROI of the medical image can be a combination of the ROI of each slice. This manual process requires a significant amount of time to be paid by the user, resulting in increased costs. Additionally, this manual process can result in a delay in formulating the TTFields treatment plan when the user is scheduled to review and manually process the medical image of the subject. Furthermore, if the subject has several medical images to be used to formulate the TTFields treatment plan, each medical image can need to be reviewed and the ROI can need to be manually identified by the user in each medical image, which can result in the user spending even more time, incurring more costs associated therewith, and in more delays in formulating the TTFields treatment plan.

[0016] One or more embodiments described herein provide a technical solution to the technical problem of identifying a ROI on a subject to determine a dosage of TTFields to be administered to the subject. Specifically, the inventors have discovered a technique for automatically generating a ROI in a medical image for use in formulating a TTFields treatment plan. With the present technique, a user no longer needs to manually review a medical image or each slice of a medical image to determine a ROI for the medical image. Rather, in some embodiments, the user only needs to segment a few voxels in the medical image, and the computer automatically generates a ROI for the medical image. With the present technique, the amount of time required by a user to generate a ROI for a medical image can be significantly reduced compared to a typical manual process, which can also make the present technique less costly compared to a typical manual process. Additionally, because the user does not need to spend as much time processing the medical image using the present technique, delays in formulating a TTFields treatment plan due to the need to schedule a user to review and manually process a subject's medical image can be significantly reduced using the present technique. Moreover, if a subject has several medical images to be used in formulating a TTFields treatment plan, the amount of time required by a user to generate a ROI for each medical image according to the present technique can be significantly less than the amount of time required by a user using a typical manual process, which can significantly reduce the costs associated therewith and reduce delays in formulating a TTFields treatment plan. Because of the amount of data involved and the computational complexity, this technical solution cannot be performed by the human brain and needs to be performed by computer-based techniques described herein.

[0017] The embodiments described herein also provide a practical application that generates a transducer layout for delivering TTFields treatment to a subject by avoiding the need to perform manual generation of regions of interest in a medical image of the subject. With the present technique, the typical manual process for generating regions of interest in a medical image of a subject can be avoided, thereby saving time, reducing costs, and reducing delays in providing formulating TTFields treatment plans for a user (e.g., a healthcare provider). These and other technical improvements can be achieved using one or more embodiments described herein.

[0018] In some embodiments, the computer-based inventive technique can include automatically generating ROIs for formulating a TTFields treatment plan from a medical image. As an example, a user can first manually segment a minimum number of voxels (e.g., 10 voxels) of at least one tissue type (e.g., tumor, resection, necrosis, etc.) in a medical image, and then select an “Automatic ROI Generator” button in a user interface. In some embodiments, the user can also select a value for a margin (e.g., a proximal boundary zone (PBZ)) to be added to each segmented region of the medical image. Next, a computer can automatically determine a segmented region for each tissue type based on the manually segmented voxels, add the margin to each segmented region to obtain an extended segmented region, and combine the extended segmented regions to obtain a region of interest of the medical image.

[0019] Figure 1 An example computer-implemented method 100 for formulating a treatment plan for administering TTFields to a subject is depicted. Certain steps of the method 100 are described as computer-implemented steps. The method 100 can be implemented by any suitable system or device, such as the device 1000 of Figure 10 Although the order of the operations is indicated in Figure 1 for purposes of illustration, the time and order of such operations can be altered in appropriate circumstances without negating their intended purpose and advantages as set forth in the examples detailed herein.

[0020] At step 102, the method 100 can include presenting a slice on a display by a medical image of a subject. In some embodiments, the medical image includes voxels and includes one or more slices. The medical image can include at least one of a computed tomography (CT) medical image, a magnetic resonance imaging (MRI) medical image, or a positron emission tomography (PET) medical image. In some embodiments, the medical image of the subject can include an image of a torso and / or a head of the subject.

[0021] At step 104, the method 100 can include determining a segmentation of a minimum number of voxels of at least one tissue type in a slice of the medical image. The at least one tissue type can include at least one of a tumor, a gross tumor volume (GTV), a resection cavity, a necrosis region, or an enhanced tumor. The GTV can represent a grossly visible tumor volume, which can be used as a central target volume for treatment. The resection cavity can refer to a cavity resulting from a resection of a tissue, structure, or organ. The necrosis region can refer to cells that have died due to injury or disease. The enhanced tumor can refer to a remaining tumor region that is injected with a contrast agent, making such region more easily detectable in a medical image (e.g., in an MRI medical image). In some embodiments, a user interface can be provided to a user for determining the segmentation of the at least one tissue type in the slice of the medical image. As an example, Figure 4An example user interface of an application for manual segmentation of a medical image of a subject is depicted.

[0022] In some embodiments, step 104 can be based on user input. For example, a user can select voxels in a slice of the medical image as a particular tissue type, and the user-selected voxels can be identified as the minimum number of voxels of the at least one tissue type in the slice of the medical image for the determination of the segmentation. As an example, the minimum number of voxels of the medical image for the segmentation can be between 5 voxels and 25 voxels. As an example, the minimum number of voxels of the medical image for the segmentation can be 10 voxels. In some embodiments, if the minimum number of voxels of the medical image for the segmentation is not met, a warning can be provided to the user.

[0023] At step 106, method 100 can include receiving a user selection to automatically generate a ROI in the medical image for applying TTFields to the subject. In some embodiments, a user can select a button in the user interface, such as an “Automatic ROI Generator” button in the user interface (e.g., “Automatic ROI Generator” button 502 in FIG. 5B), and the selection can be received, and then used to automatically generate a ROI in the medical image. In some embodiments, as part of the automatic generation of the ROI, a user can also input a margin (e.g., a proximal boundary zone (PBZ)) to add to the segmented region. Figure 5

[0024] At step 108, method 100 can include performing the automatic generation of a ROI in the medical image for applying TTFields to the subject based on the segmentation of the minimum number of voxels of the at least one tissue type in the slice of the medical image in step 104. In some embodiments, the segmented region in the medical image can be automatically generated based on the segmentation of the minimum number of voxels of the at least one tissue type in step 104. In some embodiments, the automatic segmentation can be performed using techniques discussed in commonly owned U.S. Patent Application Publication No. 2021 / 0201572, entitled “METHODS, SYSTEMS, AND APPARATUSES FOR IMAGE SEGMENTATION,” the contents of which are incorporated by reference herein in their entirety. In some embodiments, the ROI can include a clinical tumor volume (CTV). The CTV can refer to a volume of tissue that includes a gross tumor volume (GTV) and subclinical microscopic malignant disease. In some embodiments, generating the ROI in the medical image can include adding a proximal boundary zone (PBZ) to the GTV to obtain a CTV in the medical image. In some embodiments, a user interface can be provided to the user for initiating the automatic generation of the ROI. As an example, Figure 5 ​An example user interface for an application used to automatically generate ROIs in medical images of subjects is depicted.

[0025] In some embodiments, automatically generating the ROI at step 108 may include: determining a segmentation region for a specific tissue type in the medical image based on the minimum number of voxels in a slice of the medical image; determining the PBZ of the segmentation region of the medical image; and then adding the PBZ to the segmentation region of the medical image to obtain an ROI in the medical image for applying TTFields to the subject. Specifically, the segmentation region may be a tumor, GTV, resection cavity, necrotic area, enhanced tumor, or non-enhanced tumor. As used herein, the phrase "segmentation region" may refer to a three-dimensional volume in a medical image. In some embodiments, step 106 may use... Figure 2 This will be implemented using method 200, which will be discussed further below.

[0026] In some embodiments, computational warnings may be provided via a user interface during the automatic generation of ROI. In some embodiments, a user interface may be provided to the user to provide warnings for the automatic generation of ROI. As an example, Figure 6 An example user interface is depicted for warnings regarding Regions of Interest (ROIs) in medical images of a subject for automatic generation. In some embodiments, the warning may indicate that a minimum number of voxels has not been selected (e.g., in step 104), and therefore, automatic ROI generation cannot be performed. In some embodiments, the warning may indicate at least one of the following: the ROI in the medical image does not contain a minimum number of gray matter and / or white matter voxels; the ROI in the medical image does not contain a minimum number of enhancing tumor voxels; and / or only regions of interest containing gray matter and / or white matter voxels or enhancing tumor voxels will be presented, including combinations and / or multiple composites thereof. If one or more of these warnings are received, the computer system may provide the user with an opportunity to resolve the issue. For example, to resolve the warning that the ROI in the medical image does not contain a minimum number of gray matter and / or white matter voxels, the user may need to modify the size of the segmentation in step 104 to include additional gray matter and / or white matter voxels. For example, to resolve the warning that the ROI in the medical image does not contain a minimum number of enhancing tumor voxels, the user may need to modify the size of the segmentation in step 104 to include additional enhancing tumor voxels. For example, to resolve the warning that only regions of interest with gray matter and / or white matter voxels or enhanced tumor voxels will be displayed, the user may need to modify the size of the segmentation in step 104 to display voxels that are not in the ROI.

[0027] At step 110, the method 100 can comprise generating a plurality of transducer layouts for applying TTFields to the subject based on the automatically generated ROI in the medical image. In some embodiments, at least one of the transducer layouts can comprise two pairs of transducers for placement at transducer locations on the subject. In some embodiments, at least one of the transducer layouts can comprise one pair of transducers for placement at transducer locations on the subject. In some embodiments, the transducer layouts can be generated based on at least one of: the three-dimensional model of the subject, a three-dimensional model of a generic subject, or a geometric calculation of the subject. In some embodiments, step 110 can be implemented using the method 300 of Figure 3 , which is discussed further below.

[0028] Figure 2 A flowchart depicting an example method 200 for automatically generating a ROI in a medical image of a subject is illustrated. The method 200 can be used to implement step 108 of Figure 1 . Certain steps of the method 200 can be described as computer- implemented steps. The method 200 can be implemented by any suitable system or device, such as the device 1000 of Figure 10 . While the order of the operations is indicated in Figure 2 for illustrative purposes, the timing and order of such operations can be altered as appropriate without negating the objects and advantages of the examples set forth in detail herein.

[0029] At step 202, the method 200 can comprise determining at least two segmented regions of different tissue types in the medical image based on segmentation of a minimum number of voxels of at least one tissue type in a slice of the medical image. As an example, the at least two segmented regions can comprise at least two of: a tumor, a gross tumor volume, a resection cavity, a necrotic region, or an enhanced tumor or non-enhanced tumor. The segmentation of the medical image having the minimum number of voxels can be selected by a user in step 104.

[0030] At step 204, method 200 may include: determining the margin of each segmented region in the medical image. This margin can be used to extend the segmented region, which may be useful if there is uncertainty regarding where the segment actually lies on the subject (e.g., voxels are located on edges or separating tumor tissue from non-tumor tissue). In some embodiments, the margin may be the PBZ of each segmented region in the medical image. In some embodiments, the margin (or PBZ) may be the same or different values ​​for each segmented region in the medical image. In some embodiments, the margin (or PBZ) may be a user-selectable value for the segmented region in the medical image. In some embodiments, the margin may be a distance measurement result. As an example, the margin may be between 1 mm and 15 mm, or between 1 mm and 20 mm. As an example, the margin may be 3 mm. In some embodiments, the margin may be a percentage of the size of the segmented region. As an example, the margin may be between 0.1% and 5.0% of the size of the segmented region, or between 0.01% and 10.0%. As an example, the margin may be 0.01%, 0.1%, 0.5%, or 1.0%.

[0031] At step 206, method 200 may include: for each segmented region, adding the margin to the segmented region of the medical image to obtain an expanded segmented region. In some embodiments, each expanded segmented region corresponds to a segmented region.

[0032] At step 208, method 200 may include: combining extended segmented regions to obtain a region of interest (ROI) in a medical image for applying TTFields to a subject. The ROI may include overlapping and non-overlapping portions of the extended segmented regions.

[0033] In some embodiments, the order of steps 206 and 208 can be reversed. As an example, the segmented regions can be combined into a combined segmented region first, and then margins can be added to the combined segmented region to obtain the ROI in the medical image.

[0034] As examples of various aspects of implementing method 200, Figures 7A to 7B are shown. Figure 7C Various stages of automatically generating ROIs in example medical images of subjects are depicted.

[0035] Figure 3 A flowchart illustrating an example method 300 for generating a transducer layout for applying TTFields to a subject is shown. Method 300 can be used to implement... Figure 1 Step 110. Some steps of method 300 can be described as computer-implemented steps. Method 300 can be implemented by any suitable system or device (such as...). Figure 10 This is achieved using device 1000. Although for illustrative purposes...Figure 3 The order of the operations is indicated, but the time and order of such operations can be changed as appropriate without negating the purpose and advantages of the examples set forth in detail herein.

[0036] At step 302, the method 300 can include creating a three-dimensional (3D) model of the subject based on the medical images, where the 3D model of the subject includes automatically generated ROIs. In some embodiments, the 3D model can include a 3D conductivity map. The 3D conductivity map can depict the electrical conductivity of the subject’s body tissue. In some embodiments, creating the 3D model can include performing calculations based on the medical images and tissue types in the medical images to determine the electrical conductivity of the subject’s tissue. As one example, creating the 3D model can include assigning tissue types and associated electrical conductivities to voxels of the 3D model of the subject. In some embodiments, creating the 3D model of the subject can include automatically segmenting normal tissue in the medical images. In some embodiments, after creating the 3D model of the subject, the method 300 can include receiving user confirmation of the 3D conductivity map associated with the 3D model. In some embodiments, automatically segmented normal tissue, such as gray and white matter, skull, scalp, and cerebrospinal fluid (CSF), can be automatically added to the 3D model. In some embodiments, creating the 3D model of the subject at step 302 can be performed using the techniques in commonly owned U.S. Patent Application Publication No. 2021 / 0201572, entitled METHODS, SYSTEMS, AND APPARATUSES FOR IMAGE SEGMENTATION, the contents of which are incorporated herein by reference in their entirety.

[0037] At step 304, the method 300 can include generating a plurality of transducer layouts for applying TTFields to the subject based on the 3D model of the subject. The transducer layouts can define one or more locations relative to the subject for placing transducers. In some embodiments, the plurality of transducer layouts can include four locations on the subject for placing four respective transducers, such as on the head or torso of the subject. In some embodiments, the plurality of transducer layouts can include two locations on the subject for placing two respective transducers, such as on the head or torso of the subject. In some embodiments, each of the transducers can include one or more electrode elements. The electrode elements can be of any suitable type or material. For example, at least one electrode element can include a ceramic dielectric layer and / or a polymer film, among others, including combinations and / or multiple composites thereof. Generating the plurality of transducer layouts can be performed after the user receives a selection from the user interface to begin the generation. In some embodiments, generating the plurality of transducer layouts at step 304 can be performed using the techniques in commonly owned U.S. Patent Application Publication No. 2021 / 0201572, entitled METHODS, SYSTEMS, AND APPARATUSES FOR IMAGE SEGMENTATION, the contents of which are incorporated by reference herein in their entirety.

[0038] At step 306, the method 300 can include selecting at least two of the transducer layouts as recommended transducer layouts to present to the user. In some embodiments, at least one of the recommended transducer layouts can deliver the highest dose of TTFields to the ROI, and / or to a tumor progression area, among others, including combinations and / or multiple composites thereof. In some embodiments, at least one of the recommended transducer layouts can be a transducer layout that is in an offset or rotated position as compared to a transducer layout that delivers the highest dose of TTFields to the ROI. In some embodiments, at least three of the recommended transducer layouts can deliver the three highest doses of TTFields to the ROI.

[0039] At step 308, the method 300 can include presenting the recommended transducer array layouts. For example, the method 300 can include presenting at least four recommended transducer layouts, although in other examples more or fewer transducer layouts can be presented. In some embodiments, presenting the recommended transducer layouts can include presenting information about the recommended transducer layouts via a user interface. The information can include one or more of: a dose of TTFields delivered to the ROI for each of the recommended transducer layouts; a medical image slice overlaid with a dose of TTFields for at least one of the recommended transducer layouts; a two-dimensional plot comparing a volume percentage of the ROI to a dose percentage of TTFields for at least one of the recommended transducer layouts; an image of the subject depicting a location of the electrode elements for at least one of the recommended transducer layouts; a two-dimensional plot depicting a cumulative dose of TTFields across the ROI for at least one of the recommended transducer layouts; a two-dimensional plot depicting a dose of TTFields across the ROI for at least one of the recommended transducer layouts; a percentage of overlap between electrode elements of two of the recommended transducer layouts; and / or a percentage of overlap between adhesive portions of two of the recommended transducer layouts, including combinations and / or multiple composites thereof.

[0040] At step 310, the method 300 can include receiving a user selection of at least one of the recommended transducer array layouts. In some embodiments, the user can select a primary transducer layout and an alternative transducer layout. To make the selection, the user can accept the primary layout as a first layout, then review and evaluate the alternative layout and select a certain alternative layout as a second layout. For example, the user can select one of the transducer layouts (e.g., the primary transducer layout) for use in a first time period. The user can select another (e.g., alternative) transducer layout of the transducer layouts for use in a second time period after the first time period.

[0041] At step 312, the method 300 can include providing a report for at least one of the selected recommended transducer layouts. In some embodiments, the report can depict a location of the transducers of the selected recommended transducer layout on the subject in multiple views. In some embodiments, the report can provide a dose of TTFields applied to the subject. It should be appreciated that, for example, different reports can be provided depending on an intended target of the report (e.g., a first report type for the subject, a second report type for inclusion in the subject’s medical record).

[0042] Turning to Figure 4 , Figure 4 An example user interface depicting an application for manual segmentation of a medical image of a subject. In particular, Figure 4A plurality of user-selectable options 402 (e.g., user-selectable icons) are shown for manually segmenting the slices presented in the drop-down menu. Examples of user-selectable options 402 can include: a user-selectable icon for automatically filling regions (e.g., a "polygonal brush" option); a user-selectable icon for selecting regions without automatically filling (e.g., a "paintbrush" option); a user-selectable icon for erasing segmentation (e.g., an "erase" option); a user-selectable icon for assigning an organization type (e.g., an "assign" option); a user-selectable icon for expanding the boundaries of a region (e.g., an "expand and margin" option); a user-selectable icon for cleaning up segmentation slices; and / or a user-selectable icon for splitting segmentation, including combinations and / or multiple composites thereof. In some embodiments, user-selectable options 402 provide the user with a selection of tools. These tools can be used to segment abnormal tissue. For example, a polygonal brush or paintbrush can be used to outline an abnormal tissue region. The abnormal tissue can be any undesirable tissue type, such as a tumor, necrotic tissue, and / or a prior surgery region (e.g., a resection cavity), including combinations or and / or multiple composites thereof.

[0043] Figure 5 An example user interface for an application for automatically generating ROIs in medical images of a subject is depicted. As Figure 5 As shown in the example, a user can begin automatic ROI generation by clicking the "Auto ROI Generator" button at 502, and an "Auto ROI Generator" window can appear. The window can display a user-adjustable value for a PBZ margin at 504, which is used to add to a segmented region in the medical image and / or to expand the segmented region to obtain an expanded segmented region in the medical image. The window can display a user-selectable tissue type for the segmented region to which the PBZ margin is to be added at 506. Examples of user-selectable tissue types can include a resection cavity, a necrotic core, or an enhanced tumor. The window can display an indication of how the margin will be combined with different tissue types at 508. As an example, the enhanced tumor can be assigned to a GTV, which can be assigned to a CTV in the ROI. As an example, the PBZ can be assigned to a CTV in the ROI. Upon making these selections, the user can click a button at 510 to accept the parameters and begin automatic ROI generation using the selected parameters.

[0044] Figure 6 An example user interface for warnings for automatically generating ROIs in medical images of a subject is depicted. Figure 6 As shown, if a minimum number of voxels for segmentation of a medical image is not met, a warning is provided to the user. As an example, Figure 6The warning in the CTV and PBZ will be added to complete the generation of the ROI, and the CTV will be the main ROI.

[0045] FIGS. 7A to Figure 7C depicts an example medical image of a subject according to Figure 2 the method 200 automatically generating an ROI in the medical image. In FIG. 7A, at step 202, two segmented regions are generated in the medical image, an enhanced tumor segmented region 712 and a resection cavity segmented region 714. In FIG. 7B, at step 206, a PBZ margin 716 (e.g., 3 mm) is added to the enhanced tumor segmented region 712, and it is also added to the resection cavity segmented region 714, resulting in two enhanced segmented regions. In Figure 7C at step 208, the two enhanced segmented regions are combined to obtain the ROI 718.

[0046] Figure 8 depicts an example system 800 for applying an alternating electric field (e.g., TTFields) to a subject’s body. The system can be used to treat a target region of a subject’s body with an alternating electric field. In examples, the target region can be in the subject’s brain, and the alternating electric field can be delivered to the subject’s body via two pairs of transducer arrays positioned on the subject’s head, such as, for example, in Figure 9 In examples, the target region can be in the subject’s torso, and the alternating electric field can be delivered to the subject’s body via two pairs of transducer arrays positioned on at least one of the subject’s chest, abdomen, or one or both thighs. Other transducer array placements on the subject’s body are also possible.

[0047] The example device 800 has four transducers (or “transducer arrays”) 800A-D. Each transducer 800A-D can include a substantially planar electrode element 802A-D that is positioned on a base 804A-D and electrically and physically connected (e.g., by conductive wiring 806A-D). The base 804A-D can include, for example, cloth, foam, flexible plastic, and / or conductive medical gel. Two transducers (e.g., 800A and 800D) can be a first pair of transducers configured to apply an alternating electric field to a target region of a subject’s body. The other two transducers (e.g., 800B and 800C) can be a second pair of transducers configured to similarly apply an alternating electric field to the target region.

[0048] The transducers 800A-D can be coupled to an AC voltage generator 820, and the system can further include a controller 810 communicatively coupled to the AC voltage generator 820. The controller 810 can include a computer having one or more processors 824 and a memory 826 accessible by the one or more processors. The memory 826 can store instructions that, when executed by the one or more processors, control the AC voltage generator 820 to induce an alternating electric field between the pairs of transducers 800A-D according to one or more voltage waveforms and / or cause the computer to perform one or more methods disclosed herein. The controller 810 can monitor operations performed by the AC voltage generator 820 (e.g., via the processor 824). One or more sensors 828 can be coupled to the controller 810 for providing measurement values or other information to the controller 810.

[0049] The electrode elements 802A-D can be capacitively coupled. In one example, the electrode elements 802A-D can be ceramic electrode elements coupled to one another via conductive traces 806A-D. The ceramic electrode elements can be circular or non-circular when viewed in a direction perpendicular to their surfaces. In other embodiments, the array of electrode elements can not be capacitively coupled, and there can be no dielectric material (such as a ceramic or high dielectric polymer layer) associated with the electrode elements.

[0050] The structure of the transducers 800A-D can take a variety of forms. The transducers can be affixed to a subject’s body or attached to or incorporated into a garment that covers a subject’s body. The transducers can include suitable materials for attaching the transducers to a subject’s body. For example, these suitable materials can include cloth, foam, flexible plastic, and / or conductive medical gel. The transducers can be electrically conductive or non-conductive.

[0051] The transducer may include any desired number of electrode elements (e.g., one or more electrode elements). For example, the transducer may include 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more electrode elements (e.g., 20 electrode elements). These electrode elements may use a variety of shapes, sizes and materials. Any structure that enables: (a) delivery of TTFields to the subject's body and (b) positioning at the location specified herein can be used to implement the transducer (or electric field generating device) used with embodiments of the invention. The transducer may be conductive or non-conductive. In some embodiments, an AC signal may be capacitively coupled into the subject's body. In some embodiments, at least one electrode element of the first, second, third, or fourth transducer may include at least one ceramic disk adapted to generate an alternating electric field. In some embodiments, at least one electrode element of the first, second, third, or fourth transducer may include a polymer film adapted to generate an alternating electric field.

[0052] Figure 10 An example computer device for use with the embodiments described herein is depicted. As an example, device 1000 may be a computer implementing certain inventive techniques disclosed herein, such as selecting transducer positions for delivering TTFields to a subject. As an example, Figure 1 Method 100 can be executed by a computer (such as device 1000). As another example, Figure 2 Method 200 can be executed by a computer (such as device 1000), which can be used to execute Figure 1 The method 100 may be the same or different for computers. As another example, Figure 3 Method 300 can also be executed by a computer (such as device 1000), which may be the same as or different from the computer used to execute methods 100 and 200. In some embodiments, Figure 8 The controller 810 can be implemented using device 1000. Device 1000 may include: one or more processors 1002; memory 1003; one or more input devices (not shown); and one or more output devices 1005.

[0053] In some embodiments, based on input 1001, one or more processors 1002 may generate control signals for controlling a voltage generator. As an example, input 1001 is a user input. As an example, input 1001 may come from another computer communicating with device 1000. Input 1001 may be received using one or more input devices (not shown) of device 1000.

[0054] Memory 1003 can be accessible by one or more processors 1002 (e.g., via link 1004) such that one or more processors 1002 can read information from and write information to memory 1003. Memory 1003 can store instructions that, when executed by one or more processors 1002, implement one or more embodiments described herein.

[0055] One or more output devices 1005 can provide status of operation of the present application, such as transducer array selection, voltage being generated, and other operational information. According to some embodiments described herein, output device 1005 can provide visualization data.

[0056] Device 1000 can be a device for developing a treatment plan for administering a tumor treating electric field to a subject, the device comprising: one or more processors (such as one or more processors 1002); and a memory (such as memory 1003) 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 described herein.

[0057] Memory 1004 can be a non-transitory processor-readable medium containing a set of instructions thereon for developing a treatment plan for administering a tumor treating electric field to a subject, wherein the instructions, when executed by a processor (such as processor 1002), cause the processor to perform one or more methods described herein.

[0058] Exemplary Embodiments The present application includes the following other illustrative embodiments (“embodiments”).

[0059] Embodiment 1 : A computer-implemented method for developing a treatment plan for administering a tumor treating electric field to a subject, the method comprising: presenting a slice on a display through a medical image of the subject, wherein the medical image comprises voxels; and determining a segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image; receiving a user selection to automatically generate a region of interest (ROI) in the medical image for applying a tumor treating electric field to the subject; and automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject based on the segmentation of the minimum number of voxels of at least one tissue type in the slice of the medical image.

[0060] Embodiment 2: The computer-implemented method of embodiment 1, wherein determining the segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image comprises: determining a segmentation of a minimum number of voxels of a tumor in the medical image.

[0061] Example 3: The computer-implemented method of example 1, wherein determining the segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image comprises determining a segmentation of a minimum number of voxels of a gross tumor volume in the medical image.

[0062] Example 4: The computer-implemented method of example 1, wherein determining the segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image comprises at least one of: determining a segmentation of a minimum number of voxels of a resection cavity in the medical image; determining a segmentation of a minimum number of voxels of a necrotic region in the medical image; or determining a segmentation of a minimum number of voxels of an enhanced tumor in the medical image.

[0063] Example 5: The computer-implemented method of example 1, wherein the region of interest in the medical image for applying a tumor treating electric field to the subject comprises a clinical tumor volume in the medical image for applying a tumor treating electric field to the subject.

[0064] Example 6: The computer-implemented method of example 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises adding a proximal boundary zone (PBZ) to a gross tumor volume (GTV) to obtain a clinical tumor volume (CTV) in the medical image for applying a tumor treating electric field to the subject.

[0065] Example 7: The computer-implemented method of example 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises: determining a segmented region of at least one tissue type in the medical image based on the segmentation of the minimum number of voxels of the at least one tissue type in the slice of the medical image; determining a proximal boundary zone (PBZ) of the segmented region of the medical image; and adding the PBZ to the segmented region of the medical image to obtain the region of interest in the medical image for applying a tumor treating electric field to the subject.

[0066] Example 8: The computer-implemented method of example 7, wherein the segmented region is a tumor, a gross tumor volume, a resection cavity, a necrotic region, an enhanced tumor, or a non-enhanced tumor.

[0067] Example 9: The computer-implemented method of Example 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises: determining at least two segmented regions of different tissue types in the medical image based on the segmentation of the minimum number of voxels of at least one tissue type in the slices of the medical image; determining a margin for each segmented region in the medical image; adding the margin to each segmented region of the medical image to obtain an expanded segmented region; and combining the expanded segmented regions to obtain the region of interest in the medical image for applying a tumor treating electric field to the subject.

[0068] Example 10: The computer-implemented method of Example 9, wherein the at least two segmented regions comprise at least two of: a tumor, a gross tumor volume, a resection cavity, a necrotic region, an enhanced tumor, or a non-enhanced tumor.

[0069] Example 10A: The computer-implemented method of Example 9, wherein the region of interest comprises overlapping and non-overlapping portions of the expanded segmented regions.

[0070] Example 10B: The computer-implemented method of Example 9, wherein the PBZ is the same value for each segmented region in the medical image.

[0071] Example 10C: The computer-implemented method of Example 9, wherein the PBZ is a different value for each segmented region in the medical image.

[0072] Example 10D: The computer-implemented method of Example 9, wherein the PBZ is a user-selectable value for the segmented regions in the medical image.

[0073] Example 11: The computer-implemented method of Example 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises: determining a proximal boundary zone (PBZ) for expanding a segmented region in the medical image to obtain an expanded segmented region in the medical image.

[0074] Example 12: The computer-implemented method of Example 15, wherein the PBZ for the segmented region of the medical image is 3 mm.

[0075] Example 12A: The computer-implemented method of Example 15, wherein the PBZ for the segmented region of the medical image is between 1 mm and 15 mm or 1 mm and 20 mm.

[0076] Example 13: The computer-implemented method of Example 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises: providing for display a user-adjustable value of a proximal boundary zone (PBZ) for expanding a segmented region in the medical image to obtain an expanded segmented region in the medical image.

[0077] Example 14: The computer-implemented method of Example 1, wherein the minimum number of voxels for segmentation of the medical image is 10 voxels.

[0078] Example 14A: The computer-implemented method of Example 1, wherein the minimum number of voxels for segmentation of the medical image is between 5 voxels and 25 voxels.

[0079] Example 15: The computer-implemented method of Example 1, wherein if the minimum number of voxels for segmentation of the medical image is not met, a warning is provided to a user.

[0080] Example 16: The computer-implemented method of Example 1, further comprising: generating a plurality of transducer layouts for applying a tumor treating electric field to the subject based on the automatically generated region of interest in the medical image.

[0081] Example 17: The computer-implemented method of Example 1, further comprising: creating a three-dimensional model of the subject based on the medical image, the three-dimensional model of the subject including the automatically generated region of interest; generating a plurality of transducer layouts for applying a tumor treating electric field to the subject based on the three-dimensional model of the subject; selecting at least two of the plurality of transducer layouts as recommended transducer layouts; presenting the recommended transducer layouts; receiving a user selection of at least one recommended transducer layout; and providing a report for at least one selected recommended transducer layout.

[0082] Example 18: The computer-implemented method of Example 1, wherein the medical image comprises at least one of: a computed tomography (CT) medical image, a magnetic resonance imaging (MRI) medical image, or a positron emission tomography (PET) medical image.

[0083] Example 18A: The computer-implemented method of Example 1, wherein the medical image comprises a torso of the subject.

[0084] Example 18B: The computer-implemented method of Example 1, wherein the medical image comprises a head of the subject.

[0085] Example 19: A non-transitory processor-readable medium comprising a set of instructions for developing a treatment plan for administering a tumor treating electric field to a subject, wherein the instructions, when executed by a processor, cause the processor to perform a method comprising: presenting a slice on a display through a medical image of the subject, wherein the medical image comprises voxels; and determining a segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image; receiving a user selection to automatically generate a region of interest (ROI) in the medical image for applying a tumor treating electric field to the subject; and automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject based on the segmentation of the minimum number of voxels of at least one tissue type in the slice of the medical image.

[0086] Example 20: An apparatus for developing a treatment plan for administering a tumor treating electric field to a subject, the apparatus 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 apparatus to perform a method comprising: presenting a slice on a display through a medical image of the subject, wherein the medical image comprises voxels; and determining a segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image; receiving a user selection to automatically generate a region of interest (ROI) in the medical image for applying a tumor treating electric field to the subject; and automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject based on the segmentation of the minimum number of voxels of at least one tissue type in the slice of the medical image.

[0087] Example 21: A method, machine, article of manufacture, and / or means substantially as shown and described.

[0088] Optionally, for each of the embodiments described herein, the voltage generating component supplies an electrical signal to the transducer, the electrical signal having an alternating current waveform with a frequency in a range of about 50 kHz to about 1 MHz and suitable for delivering TTFields treatment to a subject’s body.

[0089] Unless otherwise indicated herein, or otherwise clearly contradicted by context, the embodiments illustrated under any heading of the present disclosure, or in any section, can be combined with the embodiments illustrated under the same or any other heading, or other section, of the present disclosure. For example, and without limitation, embodiments described in dependent claim format for a given embodiment (e.g., a given embodiment described in independent claim format) can be combined with other embodiments (described in independent claim format or dependent claim format).

[0090] Various modifications, alterations, and changes can be made to the described embodiments without departing from the scope of the application as defined in the claims. It is intended that the application not be limited to the described embodiments, but that it has the full scope as defined by the language of the following claims, and equivalents thereof.

Claims

1. A computer-implemented method for formulating a treatment plan for administering a tumor treating electric field to a subject, the method comprising: presenting a slice on a display through a medical image of the subject, wherein the medical image comprises voxels; and determining a segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image; receiving a user selection to automatically generate a region of interest (ROI) in the medical image for applying a tumor treating electric field to the subject; and automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject based on the segmentation of the minimum number of voxels of at least one tissue type in the slice of the medical image.

2. The method of claim 1, wherein determining the segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image comprises: determining a segmentation of a minimum number of voxels of a tumor in the medical image.

3. The method of claim 1, wherein determining the segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image comprises: determining a segmentation of a minimum number of voxels of a gross tumor volume in the medical image.

4. The method of claim 1, wherein determining the segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image comprises at least one of: determining a segmentation of a minimum number of voxels of a resection cavity in the medical image; determining a segmentation of a minimum number of voxels of a necrotic region in the medical image; or determining a segmentation of a minimum number of voxels of an enhanced tumor in the medical image.

5. The method of claim 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises: adding a proximal boundary zone (PBZ) to a gross tumor volume (GTV) to obtain a clinical tumor volume (CTV) in the medical image for applying a tumor treating electric field to the subject.

6. The method of claim 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises: determining a segmented region of a certain tissue type in the medical image based on the segmentation of the minimum number of voxels of at least one tissue type in the slice of the medical image; determining a proximal boundary zone (PBZ) of the segmented region of the medical image; and adding the PBZ to the segmented region of the medical image to obtain the region of interest in the medical image for applying a tumor treating electric field to the subject.

7. The method of claim 6, wherein the segmented region is a tumor, a gross tumor volume, a resection cavity, a necrotic region, an enhanced tumor, or a non-enhanced tumor.

8. The method of claim 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises: ​ determining at least two segmented regions of different tissue types in the medical image based on the segmentation of the minimum number of voxels of at least one tissue type in the slice of the medical image; determining a margin for each segmented region in the medical image; adding the margin to each segmented region of the medical image to obtain an expanded segmented region; and combining the expanded segmented regions to obtain the region of interest in the medical image for applying a tumor treating electric field to the subject.

9. The method of claim 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises: determining a proximal boundary zone (PBZ) for expanding a segmented region in the medical image to obtain an expanded segmented region in the medical image.

10. The method of claim 1, wherein automatically generating the region of interest in the medical image for applying a tumor treating electric field to the subject comprises: providing for display a user-adjustable value for a proximal boundary zone (PBZ) for expanding a segmented region in the medical image to obtain an expanded segmented region in the medical image.

11. The method of claim 1, wherein if the minimum number of voxels for segmentation of the medical image is not met, a warning is provided to a user.

12. The method of claim 1, further comprising: generating a plurality of transducer layouts for applying a tumor treating electric field to the subject based on the automatically generated region of interest in the medical image.

13. The method of claim 1, further comprising: creating a three-dimensional model of the subject based on the medical image, the three-dimensional model of the subject including the automatically generated region of interest; generating a plurality of transducer layouts for applying a tumor treating electric field to the subject based on the three-dimensional model of the subject; selecting at least two of the plurality of transducer layouts as recommended transducer layouts; presenting the recommended transducer layouts; receiving a user selection of at least one recommended transducer layout; and providing a report for at least one selected recommended transducer layout.

14. A non-transitory processor-readable medium having a set of instructions embodied thereon for formulating a treatment plan for administering a tumor treating electric field to a subject, wherein the instructions, when executed by a processor, cause the processor to perform a method comprising: presenting a slice on a display through a medical image of the subject, wherein the medical image comprises voxels; and determining a segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image; receiving a user selection to automatically generate a region of interest (ROI) in the medical image for applying a tumor treating electric field to the subject; and and The segmentation of the minimum number of voxels of at least one tissue type in the slice of the medical image automatically generates the region of interest in the medical image for applying a tumor treating electric field to the subject.

15. An apparatus for developing a treatment plan for administering a tumor treating electric field to a subject, the apparatus comprising: one or more processors; and memory accessible by the one or more processors, the memory storing instructions that, when executed by the one or more processors, cause the apparatus to perform a method comprising: presenting a slice on a display through a medical image of the subject, wherein the medical image comprises voxels; and determining a segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image; receiving a user selection to automatically generate a region of interest (ROI) in the medical image for applying a tumor treating electric field to the subject; and the segmentation of the minimum number of voxels of at least one tissue type in the slice of the medical image automatically generates the region of interest in the medical image for applying a tumor treating electric field to the subject. one or more processors; and memory accessible by the one or more processors, the memory storing instructions that, when executed by the one or more processors, cause the apparatus to perform a method comprising: presenting a slice on a display through a medical image of the subject, wherein the medical image comprises voxels; and determining a segmentation of a minimum number of voxels of at least one tissue type in the slice of the medical image; receiving a user selection to automatically generate a region of interest (ROI) in the medical image for applying a tumor treating electric field to the subject; and the segmentation of the minimum number of voxels of at least one tissue type in the slice of the medical image automatically generates the region of interest in the medical image for applying a tumor treating electric field to the subject.

Images