Determination of the conductivity of medical images based on measured resistance values.

Abstract

The present invention provides a computer implementation method for generating at least one transducer location for delivering a tumor treatment field to a subject. The method comprises acquiring a medical image of the subject, the medical image having multiple voxels, the medical image representing multiple tissue types of the subject, and at least one voxel associated with each tissue type. The method further comprises determining the conductivity for the subject's tissue types in the medical image using a trained machine learning model and the subject's medical image, the trained machine learning model being trained using medical images of multiple other subjects and resistance values ​​obtained by applying a tumor treatment field to other subjects. The method further comprises identifying the location of a tumor in the medical image and generating at least one transducer location for delivering a tumor treatment field to the subject.

Description

Technical Field

[0001] Cross - reference to Related Applications This application claims priority to U.S. Patent Application No. 18 / 745,184, filed on June 17, 2024, and U.S. Provisional Application No. 63 / 524,511, filed on June 30, 2023, the contents of which are hereby incorporated by reference in their entirety.

Background Art

[0002] A Tumor Treatment Field (TT field) is a low - intensity alternating electric field within an intermediate frequency range (e.g., 50 kHz to 1 MHz) and may be used for the treatment of tumors as described in U.S. Patent No. 7,565,205. The TT electric field is non - invasively induced into the region of interest by applying an alternating current (AC) voltage between transducers placed on the body of a subject (i.e., a patient). Conventionally, a first pair of transducers and a second pair of transducers are placed on the body of the subject. An AC voltage is applied between the first pair of transducers for a first time interval to generate an electric field with electric field lines running generally in the anteroposterior direction. Next, an AC voltage is applied between the second pair of transducers at the same frequency for a second time interval to generate an electric field with electric field lines running generally in the left - right direction. The system repeats this two - step sequence over the course of treatment.

Summary of the Invention

Means for Solving the Problems

[0003] One aspect of the present disclosure provides a computer - implemented method for generating at least one transducer position for delivering a Tumor Treatment Field to a subject. The method includes obtaining a medical image of the subject, the medical image having a plurality of voxels and representing a plurality of tissue types of the subject, wherein at least one voxel is associated with each tissue type; The method involves determining the conductivity of a subject's tissue type in a medical image using a trained machine learning model and the subject's medical images, wherein the trained machine learning model is trained using medical images of multiple other subjects and resistance values ​​obtained by applying a tumor treatment field to other subjects. Identifying the location of tumors in the medical images of the subjects, The method includes generating at least one transducer position for delivering a tumor treatment field to a subject based on the conductivity of the subject's tissue type in the medical image and the location of the tumor in the medical image.

[0004] Another aspect of this disclosure provides a computer implementation method for obtaining a trained machine learning model for identifying conductivity in medical images. This method Acquiring multiple medical images of multiple subjects, wherein each medical image has multiple voxels, each medical image includes multiple tissue types of the subject, and at least one voxel is associated with each tissue type. By applying a tumor treatment area to each subject, the measured resistance value of each subject is obtained, Training a machine learning model to determine conductivity in medical images, the machine learning model being trained using multiple medical images of multiple subjects and measured resistance values ​​of each subject obtained by applying a tumor treatment field to each subject.

[0005] Another aspect of this disclosure provides an apparatus for selecting transducer positions for delivering a tumor treatment site to a subject. The apparatus comprises one or more processors and a memory accessible by one or more processors, the memory, when executed by one or more processors, The method involves determining the conductivity of a subject's tissue type in a medical image using a trained machine learning model and the subject's medical images, wherein the trained machine learning model is trained using medical images of multiple other subjects and resistance values ​​obtained by applying a tumor treatment field to other subjects. Identifying the location of tumors in the medical images of the subjects, The system includes a memory for storing instructions to generate at least one transducer position for delivering a tumor treatment field to a subject, based on the conductivity of the subject's tissue type in a medical image and the location of the tumor in a medical image. [Brief explanation of the drawing]

[0006] [Figure 1] This specification describes an exemplary method for training a machine learning model to determine conductivity in medical images, according to one or more embodiments described herein. [Figure 2] This specification describes an exemplary method for determining the conductivity of a medical image based on measured resistance values, according to one or more embodiments described herein. [Figure 3] This specification illustrates an exemplary apparatus for applying an alternating electric field to a subject's body, according to one or more embodiments described herein. [Figure 4A] A schematic diagram of an exemplary design of a transducer for applying an alternating electric field according to one or more embodiments described herein is shown. [Figure 4B] A schematic diagram of an exemplary design of a transducer for applying an alternating electric field according to one or more embodiments described herein is shown. [Figure 5] Examples of transducer placement on a subject's head according to one or more embodiments described herein are shown. [Figure 6] This specification shows an exemplary computer device according to one or more embodiments described herein. [Modes for carrying out the invention]

[0007] This application describes an exemplary technique for training a machine learning model to predict conductivity measurements for different tissue types in medical images, and for using the trained machine learning model to identify locations on a subject's body for placing transducers to apply a TT field.

[0008] Generally, one or more transducers are placed on the subject's body and used to alternately apply AC voltage (e.g., TT field) to the subject's body. Generally, it is preferable to have at least two pairs of transducers positioned to target specific locations or structures (e.g., tumors) within the subject. Therefore, proper placement of the transducers is crucial for the subject's treatment. To provide effective TT field therapy to a subject, it is necessary to generate precise locations for transducer placement on the subject's body, these precise locations based, 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 difficult, and this determination is usually achieved by computer simulation of numerous possible positions for transducer placement on a 3D model of the subject. Deriving precise locations is time-consuming and laborious.

[0009] Conventional treatment plans use a series of measurements obtained from a subject's magnetic resonance imaging (MRI) scan to measure aspects of the subject (e.g., the subject's head size, tumor location, tumor size, and / or similar aspects, and combinations thereof, and / or multiple aspects). These measurements are used to generate a customized transducer layout. Generating the transducer layout requires consideration of electrical properties, such as conductivity, in the subject's tissue type. Healthcare professionals must manually identify various tissue types, which is a very cumbersome task and can be quite time-consuming due to the enormous number of voxels in medical images. Once the tissue type is determined for the subject's medical images, each voxel in each medical image can be assigned a tissue type, and then conductivity can be assigned for each tissue type of voxel.

[0010] The inventors recognized the need to determine the conductivity of tissue types in medical images of subjects. They further discovered that the conductivity of tissue types in one subject can be predicted using resistance values ​​measured from another subject.

[0011] Embodiments described herein provide a method for determining the conductivity of a subject's tissue type in a medical image using a model trained on medical images of other subjects and resistance values ​​obtained from the application of a TT field to other subjects. For example, a medical image of a subject is acquired, and the conductivity of the subject's tissue type in the medical image is determined using a trained model (e.g., a machine learning model). The location of a tumor in the subject's medical image can be identified, and based on the conductivity of the subject's tissue type in the medical image and the location of the tumor in the medical image, transducer positions for delivering a TT field to the subject can be generated.

[0012] Embodiments described herein also provide training for a model (e.g., a machine learning model) for determining conductivity in medical images. For example, multiple medical images are acquired from multiple subjects, and the medical images include multiple tissue types of the subjects. Measured resistance values ​​for each subject can be obtained by applying a TT field. Using the measured resistance values ​​and multiple medical images, a model can be trained to determine conductivity in the medical images. The acquired model can be used to generate transducer positions for delivering a TT field to a specific subject using the medical images of that subject.

[0013] Embodiments described herein provide a method for determining the conductivity of a subject's tissue type in a medical image without obtaining or using measurements of the subject's electrical conductivity or resistivity. Instead, embodiments described herein use measured resistance values ​​obtained from the application of a TT field to another subject, along with the subject's medical image, to train a machine learning model, and the trained machine learning model is used to determine the conductivity of the tissue type in the subject's medical image.

[0014] The embodiments described herein enable the determination of the conductivity of tissue types in a subject without segmenting the tissue types in the subject's medical images. Instead, the embodiments described herein use a trained machine learning model to determine the conductivity of tissue types in the subject's medical images, and the machine learning model is trained using medical images of other subjects and resistance values ​​measured from the application of a TT field to those subjects. In some embodiments, segmentation of tissue types in the subject's medical images may be used to identify the location of tumors or regions of interest within the subject, but it is not necessary to perform segmentation of tissue types in the subject's medical images to determine the conductivity of tissue types in the subject's medical images.

[0015] Embodiments described herein further provide practical applications for generating transducer positions for delivering TT fields to a subject by using a trained model to determine the conductivity of the subject's tissue type using the subject's medical images. The conductivity of the subject's tissues is taken into consideration when generating transducer layouts for treating the subject using medical images such as MRI and / or computed tomography (CT) images. This allows for individualized treatment based on the conductivity of the subject's tissue type, thereby improving the subject's treatment response. In some embodiments, using a trained model to determine the conductivity of the subject's tissue type provides the technical advantage of efficiently reducing the computation time required to determine the position of the transducer to deliver the appropriate TT field dose to the subject by generating transducer positions for delivering TT fields to the subject's medical images (e.g., MRI or CT scans). For example, instead of the user manually determining the tissue type by examining the subject's medical images, the conductivity of the tissue type can be determined using a trained model. This makes generating the transducer position(s) faster and more accurate. These and other technical improvements may be realized using one or more embodiments described herein.

[0016] FIG. 1 is a flowchart showing a method 100 for training a machine learning model to determine conductivity in medical images. The method 100 can be implemented by any suitable system or device, such as the system of FIG. 3 and / or the device of FIG. 6. Although the order of operations is shown in FIG. 1 for illustrative purposes, such timing and order of operations can be changed where appropriate without negating the objectives and advantages of the embodiments described in detail herein.

[0017] In block 102, the method 100 includes obtaining a plurality of medical images for a plurality of subjects, each medical image having a plurality of voxels, and each medical image including a plurality of tissue types of the subject. The medical images can include, for example, at least one of an MRI image, a CT image, an X-ray image, an ultrasound image, a nuclear medicine image, a positron emission tomography (PET) image, an arthrogram image, a myelogram image, or any image of the subject's body that provides an internal image of the subject's body. Each image may include an outer shape of a part of the subject's body and a region corresponding to a region 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. In certain embodiments, two or more of the medical images are associated with a similar region of one of the plurality of subjects. The plurality of medical images can be obtained from a memory (e.g., the memory 314 of FIG. 3, the memory 603 of FIG. 6).

[0018] In block 104, the method 100 includes applying a TT field to the plurality of subjects at an associated voltage and obtaining a measured current. The measured current is obtained for each of the plurality of subjects. As described herein, one or more pairs of transducers are placed on the subject's body and used to alternately apply the TT field to the subject's body.

[0019] According to one or more embodiments described herein, before obtaining a medical image of a subject (block 102), an electric current can be obtained by authorizing a TT field to the subject (block 104). In other embodiments, a medical image can be obtained (block 102) before applying the TT field to the subject and obtaining the measured electric current (block 104).

[0020] In block 106, method 100 includes calculating a resistance value measured using the measured current and associated voltage applied to the subject. The measured resistance value can be stored and / or read using a memory (e.g., memory 314 of FIG. 3, memory 603 of FIG. 6). The measured current can be obtained from a position on the subject(s) receiving the TT field. The measured resistance value can be associated with the voltage range and frequency range when applying the tumor treatment field to multiple subjects. According to one or more embodiments described herein, the current used to calculate the measured resistance value is in the range of about 0.1 ampere to 2.0 amperes. According to one or more embodiments described herein, the current used to calculate the measured resistance value is in the range of about 0.5 ampere to 1.0 amperes. Other current ranges can be implemented in other embodiments. The associated voltage can be generated by a voltage generator (e.g., voltage generator 308 of FIG. 3) used to generate the TT field. According to one or more embodiments described herein, using an MRI medical image, the resistance value at a specific electrode position on the subject can be evaluated. The evaluation result can be compared with the actual resistance measurement value obtained using the same electrode position(s), which can be performed either before or after the MRI medical image acquisition. If necessary, the accuracy of the result can be improved by recalculating the calculated resistance value.

[0021] In block 108, method 100 includes training a machine learning model to determine conductivity in medical images, the machine learning model being trained using multiple medical images of multiple other subjects and the resistance values ​​of each subject obtained by applying a TT field to multiple other subjects. For example, an untrained machine learning model or model form can be trained using training data. The training data may include multiple medical images of multiple subjects and the measured resistance values ​​of each subject by applying a TT field to each subject. In one example, a training engine may receive training data and a model form. The model form may represent an untrained base model. The model form may include pre-set weights and biases, which are adjustable during training. Training can be supervised learning, semi-supervised learning, unsupervised learning, reinforcement learning, and / or similar forms, and combinations thereof and / or a combination thereof. Training may be performed over multiple iterations (called "epochs") until a suitable model is learned.

[0022] The machine learning model may be a software program stored in memory (e.g., memory 603 in Figure 6) and executable by a processor (e.g., one or more processors 602 in Figure 6). Medical images may also be stored in memory and accessible from the machine learning model.

[0023] Machine learning models may take several different forms (e.g., neural networks, linear regression, decision trees, support vector machines, etc.). Machine learning models may also be a combination of hardware and software. Machine learning models may also be software programs stored in memory. In some embodiments, machine learning models may be neural networks, such as convolutional neural networks or recurrent neural networks. Examples of convolutional neural network algorithms used to train convolutional neural networks include AlexNet, ResNet, and GoogLeNet. Exemplary algorithms for training recurrent neural networks include Hopfield bidirectional associative memory networks, long short-term memory networks, and recurrent multilayer perceptron networks. Training a machine learning model may involve minimizing a loss function. For example, the loss function for training can be set based on predicted and measured resistance values ​​(e.g., measured voltage and current at a specific frequency, where resistivity is calculated using the measured voltage and current). According to one or more embodiments described herein, the loss function can be any suitable loss function, such as a mean squared error (MSE) loss function or a mean absolute error (MAE) loss function. However, it should be understood that other loss functions are also possible. Once training is complete, the trained machine learning model can determine the conductivity of voxels in medical images related to the subject's tissue.

[0024] According to one or more embodiments described herein, a machine learning model can be trained using one channel (e.g., one pair of electrodes), two channels (e.g., two pairs of electrodes), and / or three or more channels (e.g., three pairs of electrodes). In one example, the machine learning model is trained on a single channel for acquiring conductivity. This can provide cleaner measurements than when the machine learning model is trained on multiple channels. In an example where the machine learning model is trained on two channels, the two channels are roughly orthogonal to each other, and the current passing through the tumor is maximized.

[0025] In block 110, method 100 includes obtaining a trained machine learning model. Obtaining a trained machine learning model may include receiving the trained machine learning model from a processing system that performs training. After acquisition, the machine learning model can be used to perform inference (i.e., the process of determining the conductivity of a medical image based on measured resistance values). This process will be described in detail with reference to Figure 2.

[0026] In particular, Figure 2 is a flowchart of a method 200 for determining the conductivity of a medical image based on measured resistance values, according to one or more embodiments described herein. Method 200 can be carried out by any suitable system or apparatus, such as the system and / or apparatus of Figure 3 and / or Figure 6. Although the sequence of operations is shown in Figure 2 for illustrative purposes, the timing and sequence of such operations may be modified where appropriate without prejudice to the purpose and merits of the embodiments described herein.

[0027] In block 202, method 200 includes acquiring a medical image of a subject. The medical image has multiple voxels and represents multiple tissue types of the subject. According to one or more embodiments described herein, the multiple tissue types include one or more of skin, bone, skull, organs, brain, and / or any other tissue of the human body, and / or combinations thereof. According to one or more embodiments described herein, data in the medical image can be correlated with tissue type (and ultimately conductivity) using segmentation, voxel intensity, relative position within the body, and / or similar techniques, and / or combinations thereof.

[0028] In block 204, method 200 includes determining the conductivity of a subject's tissue type in a medical image using a trained machine learning model and the subject's medical image, the trained machine learning model being trained using medical images of several other subjects and resistance values ​​obtained by applying a TT field to other subjects as described herein. Measuring the conductivity of a subject's tissue type allows for the creation of a conductivity mapping in the subject's medical image. Determining the conductivity of a subject's tissue type may depend on one or more frequencies used to transmit the TT field to the subject. The resistance values ​​obtained by applying the TT field can be based on current values ​​measured when the TT field is applied to other subjects at the associated voltage. That is, the resistance values ​​can be calculated using the current and the associated voltage. The trained machine learning model can be trained over a certain voltage range and a certain frequency range in applying the TT field to other subjects. According to one or more embodiments described herein, the trained machine learning model is for determining the conductivity of a given TT field frequency and a given cancer.

[0029] According to one or more embodiments described herein, a trained machine learning model is for determining conductivity in a given frequency range for delivering a TT field. For example, a trained machine learning model is for determining conductivity at a frequency for delivering a TT field at approximately 250 kHz. In another example, a trained machine learning model is for determining conductivity at a frequency for delivering a TT field at approximately 500 kHz. It should be understood that these frequencies are merely examples, and the trained machine learning model can also be used to determine conductivity for other frequencies, or conversely, conductivity for other frequencies. While the determination of conductivity is described as being performed using a trained machine learning model, it should be noted that conductivity can also be determined by other methods, such as manual operation by a user. When using machine learning, method 200 may include training a machine learning model to obtain a trained machine learning model. The machine learning model can be trained using medical images of other subjects and resistance values ​​obtained from the application of a TT field to other subjects, at least one of which contains a tumor.

[0030] In block 206, method 200 includes locating a tumor in a medical image of a subject. The locating of a tumor in a medical image of a subject can be performed based on user input. For example, a user (e.g., a healthcare professional) can indicate the location of a tumor in one or more medical images of a subject. The locating of a tumor in a medical image of a subject can also be performed based on segmenting the tumor tissue from other tissues within the medical image. For example, the tumor tissue can be automatically and / or manually segmented from other tissues.

[0031] In block 208, method 200 includes generating at least one transducer position for delivering a field to a subject based on the conductivity of the subject's tissue type in the medical image and the location of the tumor in the medical image. According to one or more embodiments described herein, an optimized transducer position layout can be generated in block 208. Non-limiting examples of an optimized layout include maximizing the electric field, maximizing the power density within the tumor, and / or similar optimization methods, and combinations thereof, and / or a combination thereof.

[0032] In block 210, method 200 includes outputting at least one transducer location for delivering a TT field to a subject. A transducer location(s) refers to a location on the subject where a transducer is placed to apply a TT field to the subject. For example, Figure 5 shows an example of an output transducer arrangement with four transducers 500.

[0033] Figure 3 shows an example of a system 300 for applying an alternating current electric field (e.g., a TT field) to a subject's body according to one or more embodiments described herein. This system may be used to treat a target area of ​​the subject's body with the alternating current electric field. In one example, the target area may be in the subject's brain, and the alternating current electric field may be delivered to the subject's body via two pairs of transducer arrays (e.g., four transducers 500 in Figure 5) positioned on the subject's head. In another example, the target area may be in the subject's torso, and the alternating current electric field may 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 arrangements of transducer arrays on the subject's body are also possible.

[0034] The exemplary apparatus 300 shows an example system having four transducers (or "transducer arrays") 300A-D. Each transducer 300A-D may include substantially flat electrode elements 302A-D that are arranged on substrates 304A-D and electrically and physically connected (e.g., via conductive wiring 306A-D). For each substrate 304A-D, the electrode elements 302A-D corresponding to the substrate may be electrically connected to each other and physically connected (e.g., via conductive wiring 307D). For each substrate 304A-D, the electrode elements 302A-D corresponding to the substrate may be electrically connected to each other and physically connected to each of the substrates 304A-D. In one example, the electrode elements 302A-D may be controlled as a group so that the electrode elements 302A-D receive and execute the same command signal. For example, electrode elements 302A to D may be controlled individually, and each electrode element may receive and execute commands different from those received and executed by other electrode elements.

[0035] The substrates 304A-D may include, for example, cloth, foam, flexible plastic, and / or conductive medical gel. Two transducers (e.g., 300A and 300D) may be a first transducer pair configured to apply an alternating electric field to a target area of ​​the subject's body. Two other transducers (e.g., 300B and 300C) may be a second transducer pair similarly configured to apply an alternating electric field to a target area.

[0036] Transducers 300A-D may be coupled to an AC voltage generator 308, and the system may further include a controller 310 communicatively coupled to the AC voltage generator 308. The controller 310 may include a computer including one or more processors 312 and a memory 314 accessible by one or more processors. The memory 314 may store instructions, when executed by one or more processors, for controlling the voltage generator 308 to induce an alternating electric field between transducer pairs 300A-D according to one or more voltage waveforms, and / or causing the computer to perform one or more methods disclosed herein. The controller 310 may monitor the operation performed by the AC voltage generator 308 (e.g., via a processor 312). One or more sensors 316 may be coupled to the controller 310 to provide the controller with measurements or other information.

[0037] Electrode elements 304A to D may be capacitively coupled. In one example, electrode elements 304A to D are ceramic electrode elements coupled to each other via conductive wiring 307A to D. The ceramic electrode elements may be circular or non-circular when viewed from a direction perpendicular to their surface. In other embodiments, the array of electrode elements is not capacitively coupled, and there is no dielectric material (such as a ceramic or high-dielectric polymer layer) associated with the electrode elements.

[0038] The structure of transducers 300A-D can take various forms. The transducer may be fixed to the subject's body, or attached to or incorporated into clothing covering the subject's body. The transducer may include suitable materials for attaching it to the subject's body. For example, suitable materials may include cloth, foam, flexible plastic, and / or conductive medical gel. The transducer may be conductive or non-conductive.

[0039] A transducer may include any desired number of electrode elements (e.g., one or more electrode elements). For example, a transducer may include one, two, three, four, five, six, seven, eight, nine, ten, or more electrode elements (e.g., twenty electrode elements). Electrode elements may be of various shapes, sizes, and materials. Any structure for implementing a transducer (or electric field generator) used in conjunction with embodiments of the present invention may be used, insofar as they are capable of (a) delivering a TT electric field to the body of a subject and (b) being positioned at locations specified herein. In certain embodiments, at least one electrode element of a first, second, third, or fourth transducer may include at least one ceramic disc adapted to generate an alternating electric field. In non-limiting embodiments, at least one electrode element of a first, second, third, or fourth transducer may include a polymer film adapted to generate an alternating electric field.

[0040] Figure 4A is a schematic diagram showing an exemplary design of a transducer for applying an alternating electric field. The transducer array 401 includes 20 electrode elements 402 arranged on a substrate 403, the electrode elements 402 being electrically and physically connected to one another by conductive wiring 404. In some embodiments, the electrode elements 402 may include ceramic disks.

[0041] Figure 4B is a schematic diagram showing an exemplary design of a transducer for applying an alternating electric field. The transducer 405 may include one or more substantially flat electrode elements 406. In some embodiments, the electrode elements 406 and 407 are non-ceramic dielectric materials arranged across a plurality of flat conductors. Examples of non-ceramic dielectric materials arranged on flat conductors include polymer films arranged on pads on a printed circuit board or on substantially flat metal pieces. In some embodiments, such polymer films have a high dielectric constant, for example, a dielectric constant greater than 10. In some embodiments, the electrode elements 406 may have a variety of shapes. For example, the electrode elements may be triangular, rectangular, circular, oval, ovaloid, ovoid, or elliptical in shape, or substantially triangular, substantially rectangular, substantially circular, substantially oval, substantially ovaloid, substantially ovoid, or substantially elliptical in shape. In some embodiments, each of the electrode elements 406 may have the same shape, a similar shape, and / or a different shape.

[0042] Figure 6 shows an example of a computer device 600 for use in embodiments of the present invention. In one example, the device 600 may be a computer for performing certain ingenious techniques disclosed herein, such as determining the conductivity of a medical image based on measured resistance values, generating at least one transducer position for delivering a tumor treatment field to a subject, obtaining a trained machine learning model for identifying conductivity in a medical image, and / or selecting a transducer position for delivering a tumor treatment magnetic field to a subject. For example, blocks 102-110 in Figure 1 and / or blocks 202-210 in Figure 2 may be performed by a computer such as the device 600. In one example, the device 600 may be a controller device for applying an alternating electric field (e.g., a TT field) having a modulated electric field according to the embodiments herein. The device 600 may be used as the controller 310 in Figure 3. The device 600 may include one or more processors 602, memory 603, one or more input devices, and one or more output devices 605.

[0043] In one example, based on input 601, one or more processors 602 may generate control signals to control a voltage generator in order to implement an embodiment of the present disclosure. In one example, input 601 is a user input. In another example, input 601 may be from another computer communicating with the device 600. Input 601 may be received in conjunction with one or more input devices (not shown) of the device 600.

[0044] Memory 603 is accessible by one or more processors 602 (for example, via link 604), so that one or more processors 602 can read information from and write information to memory 603. When executed by one or more processors 602, memory 603 may store instructions that implement one or more embodiments of the present disclosure.

[0045] One or more output devices 605 may provide information about the operation of the present invention, such as the selection of the transducer array, the voltages generated, and other operational information. Output devices 605 may provide visualization data according to a particular embodiment of the present invention.

[0046] The apparatus 600 may be an apparatus for generating at least one transducer position for delivering a tumor treatment field to a subject, and / or for obtaining a trained machine learning model for identifying conductivity in a medical image, and / or for selecting a transducer position for delivering a tumor treatment field to a subject, the apparatus including one or more processors (such as one or more processors 602), and a memory accessible by one or more processors (such as memory 603) which, when executed by one or more processors, stores instructions causing the apparatus to perform one or more of the methods described herein.

[0047] Memory 603 may be a non-temporary processor-readable medium containing a set of instructions for generating at least one transducer position for delivering a tumor treatment field to a subject, and / or obtaining a trained machine learning model for identifying conductivity in a medical image, and / or selecting a transducer position for delivering a tumor treatment field to a subject, the instructions, when executed by a processor (such as processor 602), cause the processor to perform one or more of the methods described herein.

[0048] Exemplary Embodiments The present invention includes other exemplary embodiments ("Embodiments") as follows:

[0049] Embodiment 1: A computer implementation method for a treatment plan for delivering a tumor treatment field to a subject, the method comprising: acquiring a medical image of the subject, the medical image having a plurality of voxels, the medical image representing a plurality of tissue types of the subject, and at least one voxel associated with each tissue type; determining the conductivity of the subject's tissue types in the medical image using a trained machine learning model and the medical image of the subject, the trained machine learning model being trained using medical images of a plurality of other subjects and resistance values ​​obtained by applying a TT field to the other subjects; identifying the location of a tumor in the medical image of the subject; and generating at least one transducer location for delivering the tumor treatment field to the subject based on the conductivity of the subject's tissue types in the medical image and the location of the tumor in the medical image.

[0050] Embodiment 2: The method according to Embodiment 1, wherein determining the conductivity of the subject for the tissue type generates a conductivity mapping of the medical image of the subject.

[0051] Embodiment 2A: The method according to Embodiment 1, wherein determining the conductivity of the subject to the tissue type depends on one or more frequencies used to deliver the tumor treatment field to the subject.

[0052] Embodiment 3: Identifying the location of the tumor in the medical image of the subject is based on user input, as described in Embodiment 1.

[0053] Embodiment 4: The method according to Embodiment 1, wherein identifying the location of the tumor in the medical image of the subject is based on segmenting the tumor tissue from other tissues in the medical image.

[0054] Embodiment 5: The method according to Embodiment 1, wherein the plurality of tissue types include one or more of skin, bone, skull, organs, or brain.

[0055] Embodiment 6: The method according to Embodiment 1, wherein the resistance value obtained from the application to the tumor treatment field is obtained for multiple voltages at one or more frequencies of the tumor treatment field.

[0056] Embodiment 7: The method according to Embodiment 1, wherein the resistance value obtained from the application of the tumor treatment field is based on the current measured when the tumor treatment field is applied to the other subject at the relevant voltage.

[0057] Embodiment 8: The method according to Embodiment 1, wherein the trained machine learning model is trained over a certain voltage range and a certain frequency range to apply the tumor treatment field to the other subject.

[0058] Embodiment 9: The method according to Embodiment 1, wherein the trained machine learning model is for determining a specified tumor treatment field frequency and conductivity for a specified cancer.

[0059] Embodiment 9A: The method according to Embodiment 1, wherein the trained machine learning model is for determining conductivity for a specified frequency range for delivering the tumor treatment field.

[0060] Embodiment 9B: The method according to Embodiment 1, wherein the trained machine learning model is for determining conductivity for a frequency of about 250 kHz for delivering the tumor treatment field.

[0061] Embodiment 9C: The method according to Embodiment 1, wherein the trained machine learning model is for determining conductivity for a frequency of about 500 kHz for delivering the tumor treatment field.

[0062] Embodiment 10: The method according to Embodiment 1, further comprising training a machine learning model to obtain the trained machine learning model, wherein the machine learning model is trained using the resistance values ​​obtained by applying the medical images and tumor treatment fields of the other subject to the other subject, and at least one of the medical images of the other subject includes a tumor.

[0063] Embodiment 11: The method according to Embodiment 10, wherein training the machine learning model to obtain the trained machine learning model includes calculating the resistance value based on a current measured when the tumor treatment field is applied to the other subject at a relevant voltage, the current being obtained from memory.

[0064] Embodiment 11A: A method for delivering a tumor treatment field to a subject, the method comprising: arranging a plurality of transducer arrays on the subject, wherein at least one transducer array is positioned on the subject based on at least one transducer position generated by a machine learning model trained to identify a specified frequency and conductivity in a medical image for a specified cancer; and delivering the tumor treatment field to the subject using the transducer array positioned on the subject.

[0065] Embodiment 12: A computer implementation method for obtaining a trained machine learning model for determining conductivity in medical images, the method comprising: acquiring a plurality of medical images of a plurality of subjects, each medical image having a plurality of voxels, each medical image comprising a plurality of tissue types of the subject, and at least one voxel associated with each tissue type; acquiring measured resistance values ​​of each subject by applying a tumor treatment field to each subject; and training a machine learning model to determine conductivity in medical images, the machine learning model being trained using the plurality of medical images of the plurality of subjects and the measured resistance values ​​of each subject obtained by applying the tumor treatment field to each subject.

[0066] Embodiment 13: The method according to Embodiment 12, wherein two or more of the medical images are associated with similar regions of one of the multiple subjects.

[0067] Embodiment 13A: The method according to Embodiment 12, wherein the plurality of medical images are acquired from memory and the measured resistance values ​​are acquired from memory.

[0068] Embodiment 14: The method according to Embodiment 12, wherein the measured resistance value is associated with the voltage range and frequency range when the tumor treatment field is applied to the plurality of subjects.

[0069] Embodiment 15: The method according to Embodiment 12, further comprising calculating the measured resistance value using the current measured when the tumor treatment field is applied to the plurality of subjects at the relevant voltage.

[0070] Embodiment 15A: The method according to Embodiment 15, wherein the current used to calculate the measured resistance value is in the range of approximately 0.1 amperes to 2.0 amperes.

[0071] Embodiment 15B: The method according to Embodiment 15, wherein the current used to calculate the measured resistance value is in the range of approximately 0.5 amperes to 1.0 ampere.

[0072] Embodiment 16: The method of Embodiment 12, further comprising applying a tumor treatment field to the plurality of subjects with a relevant voltage to obtain a measured current, and using the measured current and the relevant voltage to calculate the measured resistance value.

[0073] Embodiment 17: The method according to claim 16, wherein the measured current is obtained from the position on the subject receiving the tumor treatment field.

[0074] Embodiment 18: The method according to Embodiment 16, wherein the associated voltage is generated by a voltage generator used to generate the tumor treatment field.

[0075] Embodiment 19: The method according to claim 11, wherein the trained machine learning model can determine the conductivity of voxels in a medical image related to the subject's tissue.

[0076] Embodiment 20: Apparatus for selecting transducer positions for delivering a tumor treatment 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: determine the conductivity of a tissue type of the subject in a medical image using a trained machine learning model and a medical image of the subject, the trained machine learning model being trained using medical images of several other subjects and resistance values ​​obtained by applying a tumor treatment field to the other subjects; identify the location of a tumor in the medical image of the subject; and generate at least one transducer position for delivering the tumor treatment field to the subject based on the conductivity of the tissue type of the subject in the medical image and the location of the tumor in the medical image.

[0077] Embodiment 20A: A non-temporary processor-readable medium comprising a series of instructions recorded thereon, the series of instructions, when executed by a processor, causing the processor to determine, using a trained machine learning model and a medical image of the subject, the conductivity of a tissue type of the subject in the medical image, the trained machine learning model being trained using medical images of several other subjects and resistance values ​​obtained by applying a tumor treatment field to the other subjects, the location of a tumor in the medical image of the subject, and the generation of at least one transducer position for delivering the tumor treatment field to the subject, based on the conductivity of the tissue type of the subject in the medical image and the location of the tumor in the medical image.

[0078] Optionally, in each embodiment described herein, the voltage generating component supplies a transducer with an AC waveform suitable for delivering TT field therapy to the subject's body at a frequency in the range of about 50 kHz to about 1 MHz.

[0079] Embodiments described in any heading or portion of this disclosure may be combined with embodiments described in the same heading or other portions of this disclosure, unless otherwise stated herein or unless the context clearly contradicts the description. For example, an embodiment described in dependent claim form to a given embodiment (e.g., a given embodiment described in independent claim form) may be combined with other embodiments (described in independent claim form or dependent claim form).

[0080] Numerous modifications, alterations, and changes are possible to the embodiments described without departing from the scope of the invention as defined in the claims. The invention is not limited to the embodiments described and is intended to have the entire scope as defined by the following claims and their equivalents. [Explanation of Symbols]

[0081] 308 AC Voltage Generator 310 Controller 312 processors 314 memory 316 sensors (multiple sensors possible) 600 equipment 602 One or more processors 603 memory 605 One or more output devices

Claims

1. A computer implementation method for generating at least one transducer position for delivering a tumor treatment site to a subject, wherein the method is Acquiring medical images of a subject, wherein the medical images have multiple voxels, represent multiple tissue types of the subject, and at least one voxel is associated with each tissue type. The determination of conductivity for the tissue type of the subject in the medical image, using a trained machine learning model and the medical image of the subject, wherein the trained machine learning model is trained using medical images of multiple other subjects and resistance values ​​obtained by applying a tumor treatment field to the other subjects. To identify the location of the tumor in the medical image of the subject, A method comprising generating at least one transducer position for delivering the tumor treatment field to the subject based on the conductivity of the tissue type of the subject in the medical image and the position of the tumor in the medical image.

2. The method according to claim 1, wherein determining the conductivity of the subject for the tissue type generates a conductivity mapping of the medical image of the subject.

3. The method according to claim 1, wherein the location of the tumor in the medical image of the subject is determined based on user input.

4. The method according to claim 1, wherein the resistance value obtained from the application of the tumor treatment field is obtained for a plurality of voltages at one or more frequencies of the tumor treatment field.

5. The method according to claim 1, wherein the trained machine learning model is trained over a certain voltage range and a certain frequency range for the application of the tumor treatment field to the other subject.

6. The method according to claim 1, wherein the trained machine learning model is for determining a specified tumor treatment field frequency and conductivity for a specified cancer.

7. The process further includes training a machine learning model to obtain the trained machine learning model, The machine learning model is trained using the medical images of the other subject and the resistance values ​​obtained by applying the tumor treatment area to the other subject. The method according to claim 1, wherein at least one medical image of the other subject includes a tumor.

8. To train the aforementioned machine learning model and obtain the aforementioned trained machine learning model is to The method of claim 7, comprising calculating the resistance value based on a current measured when the tumor treatment field is applied to the other subject with a relevant voltage, wherein the current is obtained from memory.

9. A computer implementation method for obtaining a trained machine learning model for identifying conductivity in medical images, wherein the method is: Acquiring multiple medical images of multiple subjects, wherein each medical image has multiple voxels, each medical image includes multiple tissue types of the subject, and at least one voxel is associated with each tissue type. By applying a tumor treatment area to each subject, the measured resistance value of each subject is obtained, A method comprising training a machine learning model to determine the conductivity in the medical images, wherein the machine learning model is trained using the plurality of medical images of the plurality of subjects and the measured resistance values ​​of each subject obtained by applying the tumor treatment field to each subject.

10. The method according to claim 9, wherein the measured resistance value is associated with the voltage range and frequency range when the tumor treatment field is applied to the plurality of subjects.

11. The method according to claim 9, further comprising calculating the measured resistance value using the current measured when the tumor treatment field is applied to the plurality of subjects at the relevant voltage.

12. The tumor treatment field is applied to the multiple subjects with the relevant voltage to obtain the measured current, The method according to claim 9, further comprising calculating the measured resistance value using the measured current and the associated voltage.

13. The method according to claim 12, wherein the measured current is obtained from the position on the subject receiving the tumor treatment field.

14. The method according to claim 9, wherein the trained machine learning model can determine the conductivity of voxels in a medical image related to the tissue of a subject.

15. An apparatus for selecting a transducer position for delivering a tumor treatment site to a subject, the apparatus comprising one or more processors and a memory accessible by the one or more processors, the memory, when executed by the one or more processors, The method involves determining the conductivity of a tissue type in a medical image of a subject using a trained machine learning model and the subject's medical image, wherein the trained machine learning model is trained using medical images of multiple other subjects and resistance values ​​obtained by applying a tumor treatment field to the other subjects. To identify the location of the tumor in the medical image of the subject, A device comprising: a device that stores instructions for generating at least one transducer position for delivering the tumor treatment field to the subject based on the conductivity of the tissue type of the subject in the medical image and the position of the tumor in the medical image; and a memory for storing instructions for performing these actions.

Images