EP-4736115-A1 - EDGE DETECTION BRUSH

EP4736115A1EP 4736115 A1EP4736115 A1EP 4736115A1EP-4736115-A1

Abstract

A method for processing a medical image of a subject is provided. The method includes presenting on a display a slice through the medical image of the subject. The medical image includes voxels. The method further includes performing automatic edge detection in the slice of the medical image to obtain a segmented slice. The automatic edge detection is based on a user selected voxel in the slice of the medical image. The automatic edge detection is based on a user controllable edge detection brush. Not all of the voxels selected by the edge detection brush are designated as an edge.

Inventors

VARDI, MOR
ZIGELMAN, GIL
GHANTOUS, Samer

Assignees

Novocure GmbH

Dates

Publication Date: 20260506
Application Date: 20240628

Claims (15)

1. A computer-implemented method for processing a medical image of a subject, the method comprising: presenting on a display a slice through the medical image of the subject, wherein the medical image comprises voxels; and performing automatic edge detection in the slice of the medical image to obtain a segmented slice, wherein the automatic edge detection is based on a user selected voxel in the slice of the medical image, wherein the automatic edge detection is based on a user controllable edge detection brush, wherein not all of the voxels selected by the edge detection brush are designated as an edge.
2. The method of claim 1, wherein performing automatic edge detection comprises: determining a radius of the edge detection brush for use in the automatic edge detection; and determining an edge threshold of the edge detection brush, wherein image values of voxels selected by the edge detection brush are designated as an edge based on the edge threshold.
3. The method of claim 1, wherein performing automatic edge detection comprises: determining the user selected voxel in the slice of the medical image; determining an edge detection image value based on the user selected voxel in the slice of the medical image; determining a size of the edge detection brush for use in the automatic edge detection; determining an edge threshold of the edge detection brush; determining a range of image values to designate as an edge based on the edge detection image value and the edge threshold; receiving user selected voxels in the slice based on the edge detection brush interacting with the voxels in the slice and based on the size of the edge detection brush; and designating user selected voxels in the slice as edge voxels in the slice based on the range of image values to designate as an edge.
4. The method of claim 3, wherein determining the edge detection image value based on the user selected voxel in the slice of the medical image comprises: determining a center image value as an image value of the user selected voxel in the slice of the medical image, wherein the user selected voxel is in a center of a user selected location; determining adjacent image values as image values of voxels adjacent to the user selected voxel; calculating an average image value based on the center image value and the adjacent image values; and assigning the edge detection image value as the average image value.
5. The method of claim 3, wherein determining the edge detection image value based on the user selected voxel in the slice of the medical image comprises: determining a center image value as an image value of the user selected voxel in the slice of the medical image, wherein the user selected voxel is in a center of a user selected location; and
6. The method of claim 3, wherein the range of image values to designate as an edge comprises a number M of image values in the range, an upper limit UL of the range, and a lower limit LL of the range, wherein the number M of image values in the range is: M = N * (1 - P), where a range of image values for the medical image has a maximum number N of image values, and where the user selection for the edge threshold is a percentage P, wherein the upper limit UL of the range is: UL = IV + 0.5 * M, where image value IV is the edge detection image value, and wherein a lower limit LL of the range is: LL = IV - 0.5 * M.
7. The method of claim 3, wherein the range of image values to designate as an edge comprises a number M of image values in the range, an upper limit UL of the range, and a lower limit LL of the range, wherein the number M of image values in the range is: M = N * (I - (P * RL)), where a range of image values for the medical image has a maximum number N of image values, where the user selection for the edge threshold is a percentage P, and where a range limiter is a percentage RL, wherein the upper limit UL of the range is: UL = IV + 0.5 * M, where image value IV is the edge detection image value, and wherein a lower limit LL of the range is: LL = IV - 0.5 * M.
8. The method of claim 3, wherein designating user selected voxels in the slice as edge voxels in the slice comprises, for each user selected voxel in the slice: designating the user selected voxel as an edge if an image value of the user selected voxel is in the range of image values; and designating the user selected voxel as not an edge if an image value of the user selected voxel is not in the range of image values.
9. The method of claim 1, further comprising: performing automatic segmentation of a plurality of slices through the medical image to obtain automatically segmented slices through the medical image, the automatic segmentation based on the segmented slice.
10. The method of claim 9, wherein the automatically segmented slices are automatically segmented by interpolating between the segmented slice and a second segmented slice.
11. The method of claim 1, further comprising: generating a plurality of transducer layouts for application of tumor treating fields to the subject based on the segmented slice through the medical image.
12. The method of claim 1, wherein the segmented slice comprises at least one of a resection cavity or an edema of the subject having an edge detected by the automatic edge detection.
13. A computer-implemented method for processing a medical image of a subject, the method comprising: presenting on a display a first slice through a medical image of the subject, wherein the medical image comprises voxels; performing automatic edge detection in the first slice of the medical image to obtain a first segmented slice, wherein the edge detection is based on a user selected voxel in the first slice of the medical image; presenting on the display a second slice through the medical image of the subject, wherein the first slice and the second slice are in a same direction and are separated by a plurality of slices through the medical image; performing automatic edge detection in the second slice of the medical image to obtain a second segmented slice, wherein the edge detection is based on a user selected voxel in the second slice of the medical image; performing automatic segmentation of the plurality of slices between the first slice and the second slice based on the first segmented slice and the second segmented slice to obtain segmented slices, wherein a segmented medical image comprises the first segmented slice, the second segmented slice, and the segmented slices of the medical image.
14. The method of claim 13, further comprising: defining a region of interest (ROI) in the medical image or in the segmented medical image for application of tumor treating fields to the subject; creating a three-dimensional model of the subject based on the segmented medical image, the three-dimensional model of the subject including the region of interest; generating a plurality of transducer layouts for application of tumor treating fields to the subject based on the three-dimensional model of the subject; selecting at least two of the 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 the at least one selected recommended transducer layout.
15. A computer-implemented method for processing a medical image of a subject, the method comprising: presenting on a display a slice through a medical image of the subject, wherein the medical image comprises voxels; performing automatic edge detection in the slice of the medical image using a user- controllable edge detection brush to obtain a segmented slice, wherein not all of the voxels selected by the edge detection brush are designated as an edge; performing automatic segmentation of a plurality of slices through the medical image to obtain automatically segmented slices through the medical image and a segmented medical image, the automatic segmentation based on the segmented slice; and generating a plurality of transducer layouts for application of tumor treating fields to the subject based on the segmented medical image.

Description

EDGE DETECTION BRUSH CROSS-REFERENCE TO RELATED APPLICATIONS [0001] This application claims priority to U.S. Provisional Application No. 63/609,246 filed December 12, 2023, U.S. Provisional Application No. 63/524,470 filed June 30, 2023, and U.S. Provisional Application No. 63/524,387 filed June 30, 2023, the contents of each of which are incorporated by reference herein in their entirety. This application is related to U.S. Patent Application No. 18/750,582 filed June 21, 2024 and U.S. Patent Application No. 18/750,190 filed June 21, 2024, the contents of each of which are incorporated by reference herein in their entirety. BACKGROUND [0002] Tumor treating fields (TTFields) are low intensity alternating electric fields within the intermediate frequency range (for example, 50 kHz to 1 MHz), which may be used to treat tumors as described in U.S. Patent No. 7,565,205. In current commercial systems, TTFields are induced non-invasively into a region of interest by electrode assemblies (e.g., arrays of capacitively coupled electrodes, also called electrode arrays, transducer arrays or simply “transducers”) placed on the patient’s body and applying alternating current (AC) voltages between the transducers. Conventionally, a first pair of transducers and a second pair of transducers are placed on the subject’s body. AC voltage is applied between the first pair of transducers for a first interval of time to generate an electric field with field lines generally running in the front-back direction. Then, AC voltage is applied at the same frequency between the second pair of transducers for a second interval of time to generate an electric field with field lines generally running in the right-left direction. The system then repeats this two-step sequence throughout the treatment. BRIEF DESCRIPTION OF THE DRAWINGS [0003] FIG. 1 is a flowchart depicting an example computer-implemented method for processing a medical image of a subject to generate transducer layouts for application of tumor treating fields to the subject according to an embodiment. [0004] FIG. 2 is a flowchart depicting an example computer-implemented method for automatic edge detection in a medical image of a subject according to an embodiment. [0005] FIG. 3 is a flowchart depicting an example computer-implemented method for generating transducer layouts for application of tumor treating fields to a subject according to an embodiment. [0006] FIG. 4 is an example interface of an application for automatic edge detection according to an embodiment. [0007] FIG. 5 is an example interface of an application for generating at least one transducer layout for delivering TTFields to a subject according to an embodiment. [0008] FIG. 6 is an example interface of an application for generating at least one transducer layout for delivering TTFields to a subject according to an embodiment. [0009] FIG. 7 depicts an example system to apply alternating electric fields to a subject. [0010] FIG. 8 depicts an example placement of transducers on a subject’s head. [0011] FIG. 9 depicts an example computer apparatus according to one or more embodiments described herein. [0012] Various embodiments are described in detail below with reference to the accompanying drawings, wherein like reference numerals represent like elements. DESCRIPTION OF EMBODIMENTS [0013] When generating transducer layouts for application of tumor treating fields (TTFields) for a subject, medical images of the subject are often used. Examples of such medical images are magnetic resonance imaging (MRI) scans, computed tomography (CT) medical image, positron emission tomography (PET) medical images, and/or the like including combinations and/or multiples thereof. Segmentation is often performed on medical images to divide a medical image into two or more segments with each segment representing a different object of interest. A medical image is often segmented to extract or isolate an object of interest (e.g., abnormal tissue) in the medical image from other structures (e.g., healthy tissue). For example, a medical image can be segmented to isolate a tumor from heathy tissue. Segmentation can be performed manually, semi-automatically, and/or automatically. [0014] As discovered by the inventors, when generating transducer layouts for application of TTFields for a subject, it can be difficult to perform segmentation on medical images, which is a technical problem. For example, separating tissue associated with a tumor or other similar object (e.g., abnormal tissue) from non-tumorous tissue (e.g., healthy tissues) is difficult because of the irregular shape of the tissues, the resolution of the medical images, noise in the medical images, and other similar factors. Further, segmentation is performed on each slice of a medical image. When performed manually, segmentation is a time consuming process because segmentation is performed on each slice of the medical image to isolate abnormal tissues from healthy tiss