Mammogram screening system

Abstract

A system for use in conducting a mammogram of a breast includes a breast support member, compression member and breast detector. The breast detector is spaced from the breast, when the breast is on the breast support member. The system identifying, using the breast detector, first and second pluralities of contact points between the breast and breast support member when the breast is not compressed and when the breast is compressed between the compression member and breast support member under an initial compression force. The area of the breast is calculated for both pluralities of points and, based on those areas, an imaging compression force is calculated, wherein calculating the imaging compression force may comprise estimating a density of the breast from the first breast area and second breast area. The imaging compression force can then be applied between the compression member and breast support member to facilitate reliable imaging. The breast detector may comprise a LiDAR system.

Description

[0001] Mammogram Screening System

[0002] Technical Field

[0003] The present invention relates, in general terms, to a system for mammogram screening. More particularly, the invention relates to, but is not limited to, a system for determining a force that should be applied to breast tissue during mammogram screening.

[0004] Background

[0005] Mammograms are the current gold standard for breast cancer screening. Despite this, many women do not attend regular mammogram screenings due to procedural pain.

[0006] An important procedural step in mammogram screening is the breast compression step, where a compression paddle is used to compress the breast. This compression reduces the radiation dose required for imaging, by reducing the thickness of the breast tissue to be imaged. Compression also enhances image quality by restraining the breast during imaging.

[0007] The force exerted on the breast during the compression step is determined by the radiographer or nurse conducting the screening procedure. Often times, too much force, beyond the optimal compression value is used to compress the breast resulting to significant patient discomfort. This is because it is generally understood that the force needs to be at least sufficient for clear and reliable imaging. So, some radiographers and nurses consider it better to apply slightly more force than necessary, which will not degrade images but will increase patient discomfort, rather than try to find the exact amount of force for ideal imaging and later finding the image quality is degraded due to the force being too low, thus warranting re-imaging and further patient discomfort. It is desirable, therefore, to provide a mechanism for reducing the variance in force applied to breast tissue during mammogram screening, and a system that determines an amount of force that should be applied to the breast for reliable screening without that force being so high that unnecessary discomfort results.

[0008] Summary

[0009] In view of the above problems, the technical solution described herein involves a system that determines an amount of force to apply to breast tissue during mammogram screening. In some embodiments, the amount of force is determined to be roughly equal to that which is required to evacuate blood from the breast tissue to improve imaging reliability over manual processes.

[0010] To that end, disclosed is a system for use in conducting a mammogram of a breast, comprising: memory; at least one processor; a breast support member; a compression member; and a breast detector that is spaced from the breast, when the breast is on the breast support member; wherein the memory stores instructions that, when executed by the at least one processor, cause the system to: identify, using the breast detector, a first plurality of contact points between the breast and breast support member when the breast is not compressed, and a second plurality of contact points between the breast and breast support member when the breast is compressed between the compression member and breast support member under an initial compression force; determine a first breast area from the first plurality of contact points and a second breast area from the second plurality of contact points; and calculate, based on the first breast area and second breast area, an imaging compression force to be applied between the compression member and breast support member.

[0011] Also disclosed is a method for determining an imaging compression force for conducting a mammogram of a breast, the method comprising : identifying, using a breast detector, a first plurality of contact points between the breast and breast support member when the breast is not compressed; applying an initial compression force to the breast using the compression member; identifying a second plurality of contact points between the breast and breast support member, under the initial compression force; determining a first breast area from the first plurality of contact points and a second breast area from the second plurality of contact points; and calculating, based on the first breast area and second breast area, an imaging compression force to be applied between the compression member and breast support member.

[0012] Advantageously, the system and method perform breast area determination. In some embodiments, Lidar is used to determine breast area before compression and during compression. Breast area data is directly used for estimating optimal breast compression and estimate breast density.

[0013] Advantageously, embodiments of the invention calculate optimal compression force for mammogram compression. Radiographers need not rely on individual judgement for estimating the amount of compression that is sufficient for reliably taking a clear mammogram. This helps reduce the pain for patients since radiographers can now apply exactly the force required for reliable imaging, rather than selecting a force that is on average greater than needed. This also helps reduce the odds of mammogram retakes and greater radiation exposure due to insufficient force being exerted. Rather than subjective measures for determining compression force, the present system and method make an objective determination of optimal mammogram compression force.

[0014] Advantageously, breast density is estimated for mammogram compression. Breast density affects how much force is required during breast compression. A breast of lower density usually requires less force to sufficiently compress while a breast of higher density requires more, assuming that both are of the same size. Knowing the breast density can help further optimise the optimal compression force required for good breast tissue imaging.

[0015] Advantageously, embodiments perform breast volume estimation. Breast volume estimation provides another vector to characterise the breast during mammogram screenings. Since breast density here is estimated based on the rate of deformation of the breast, the change in volume in the breast is correlated to the rate of deformation required to be observed to predict a breast density value.

[0016] Brief description of the drawings

[0017] Embodiments of the present invention will now be described, by way of nonlimiting example, with reference to the drawings in which :

[0018] Figure 1 shows a system for determining a force to apply during a mammogram, including the mammogram machine itself.

[0019] Figure 2 illustrate an embodiment of the steps performed during mammogram imaging in accordance with present teachings.

[0020] Figure 3 shows graphs of the breast area and compression force or pressure, and the contact area calculation.

[0021] Figure 4 is an illustration of a system for use in conducting a mammogram, in accordance with present teachings. Figure 5 shows the system of Figure 4, without the front cover over the breast detector (presently Lidar).

[0022] Figure 6 shows volume estimation.

[0023] Figure 7 shows the breast density prediction algorithm implemented by the present system.

[0024] Detailed description

[0025] The system described below implements a method to analyse characteristics of the breast, for estimating an optimal compression force for application by a radiographer to a breast. This removes the need for a radiographer to judge the compression force to apply, thereby removing force variations between radiographers, without jeopardising either imaging clarity or patient comfort. Since radiographers typically apply "at least sufficient" force to facilitate reliable imaging, present methods pave the way for improved patient comfort by objectively reducing the compression force applied on the breast by optimising the force.

[0026] A system 100, for implementing such a method, is shown in Figure 1. The system 100 is used during a mammogram of a breast. The system 100 comprises memory 102 and at least one processor 104. The memory 102 and processor(s) 104 are shown in broken lines as they are internal to the system 100, and may be located in a position other than that shown. The system 100 comprises a breast support member 106, a compression member 108, and a breast detector 110. During use, a breast is positioned in the breast support member 108. Compression member 108 and breast support member 106 move towards each other to close the distance therebetween and thus compress the breast. After compression, the breast is imaged using an X-ray machine 112, or other imaging machine, in the usual way. In some embodiments, the system is a complete mammogram system. In these embodiments, the system comprises an X-ray tube or X-ray imaging device, face guard for the patient, indicators and controls. In other embodiments, the system 100 can be fitted to a mammogram system, and adopt particular components of that mammogram system, such as the compression member and breast support member.

[0027] Regarding relative movement between the breast support member 106 and compression member 108, the breast support member 106 will generally remain stationary and the compression member 108 will approach the breast support member 106. Otherwise, the patient would need to move along with the breast support member 106 during compression.

[0028] The system 100 may further comprise an actuator 118. The actuator 118 may be a pedal controlled actuator or other standard mammogram actuator for controlling relative movement between the compression paddles - breast support member 106 and compression member 108. The actuator 118 therefore controls that relative movement, to apply the initial compression force and, thereafter, the imaging compression force, to the breast. The actuator 118 may also be automatic, forming part of an automatic mammogram machine - e.g., after capturing the first set of contact points, the actuator 118 automatically moves the compression member 108 towards the breast support member 106, under a predetermined initial force. The system 100 can then calculated the optimum force, and the actuator 188 can automatically actuate to apply that force.

[0029] The breast support member 106 and compression member 108 may be any kind of compression paddles. The breast support member 106, may comprise one or more indicia - e.g., visible markings on an upward-facing surface (i.e., the surface that contacts the breast). The indicia can be used by the breast detector to determine the contact points between the breast and the breast support member 106. The contact points may form a line of contact between the breast and breast support member 106.

[0030] The breast detector 110 is positioned to be spaced from the breast, when in use. For example, the breast detector 110 is positioned in Figure 1, back against the vertical housing 114 of the mammogram system. By distancing the breast from the breast detector 110, the breast detector 110 avoids leaving confounding artifacts on the X-ray images. The system therefore does not use any sensors embedded in the compression member 108 itself, or in the mammogram surface - breast support member 106. The inclusion of sensors within the x-ray imaging field (e.g., in the compression paddle (compression member) and / or mammogram surface) introduces noise and increases radiation required for the mammogram imaging. Lidar is an indirect form of sensing, by which the image quality is not impaired and the required dose of X-ray is unaffected.

[0031] In use, instructions stored in memory 102, are executed by the processor(s) 104, to cause the breast detector 110 to detect the breast. It performs detection, twice. In the first instance, a first plurality of contact points (between breast and breast support member 106) are identified. This can be done using a breast detector 110 that includes range finding capabilities, such as Lidar, laser, radar, sonar and others. Present embodiments employ Lidar. The first set or plurality of points is collected when the breast is not compressed. A second plurality of contact points is then identified or collected, when the breast is compressed between the compression member 108 and breast support member 106, under an initial compression force. The initial compression force may be a predetermined force, that is standard for all patients, or may be another force specified by a clinician.

[0032] The area of the breast can then be calculated from the first and second pluralities of contact points. The imaging force can then be determined from the different breast areas - i.e., the area when under no compression and the area when compressed under the initial compression force. From this calculation, the imaging compression force, to be applied between the compression member and breast support member, can be calculated.

[0033] In the embodiments below, the breast detector comprises Lidar-based sensors. Lidar can be replaced with an alternative sensing device and software that segments or identifies the portion of the breast on the breast support member from the surrounding image or equipment, and calculates the area of the breast support member 106 that is obstructed from the view of the image capture device by the breast.

[0034] The Lidar sensor may be a linear lidar sensor array. In some embodiments, 11 sensors are used. Greater or fewer sensors may be used as needed, to identify a sufficiently accurate set of contact points between breast and breast support member 106, for an optimised compression force to be calculated therefrom. The readings from the breast detector 110 are used to create a 2D-dot projection map to determine the area of the breast that is on the mammogram detection surface. From there, the processor(s) compute the breast area with its appropriate compression force by using a predetermined compression pressure target. In some embodiments, a compression pressure target of lOkPa is used, that target being roughly sufficient to overcome the venous blood pressure in the breast, and prevent or restrict blood flow in the breast - i.e., sufficient to expel venous blood in a standard way for all breast sizes and firmness. Beyond the pressure required to expel venous blood, applying more force will yield minimal change to volume reduction, due to nonlinear tissue elasticity. Breast tissue spread is the main reason for bruising and pain experienced during mammography. In some embodiments, the compression force is calculated to result in a predetermined pressure equal to the blood pressure in the breast, being applied across the breast. This is roughly the minimum force required to ensure reliable imaging. By obtaining data on the compression paddle height, it is also possible to predict breast volume and density in theoretical formulas suggested below. For example, the area of the breast and the difference in height between the compression paddles (breast support member 106 and compression member 108) can be multiplied, to approximate the breast volume.

[0035] The breast detector 110 may capture the contact points as a 2D point cloud array. The current system 100 uses 11 VL53L0X sensors spaced 3.4cm apart to create the 2D point cloud array. The breast area can then be determined directly from the point cloud array. In experiments, the breast area calculation was preformed using processing 4.0 software, with visualisations generated with Arduino IDE firmware on the teensy 4.0 development board, or another microprocessor - e.g., a microprocessor powered by standard USB that also handles data transfer between the breast detector 106 and processor(s) 104. In this embodiment, the teensy 4.0 is a micro-controller, responsible for managing the sensors of the breast detector 110 and controlling the visuals on screen. The processing 4.0 software is the software that processes the sensor data from the breast detector 110 through an algorithm that generates a response (e.g., after measuring the breast area before compression - since paddle height differential is not relevant with the compression member is not applying any force - the response may be to apply the initial force, unless that initial force is applied manually, or the response may be the optimal force calculated after the initial force has been applied) that is then displayed to the operator.

[0036] The Lidar module may have a sampling (refresh) rate of 100 ms for each distance reading, to focus on accuracy. The data received, i.e., contact point locations, is also denoised by a moving average algorithm to get a stable force prediction estimate. Sensors can be spaced at any appropriate distance apart, such as 3.4 cm. Closer spacing results in increased resolution. The accuracy of Lidar sensors drops if the breast is too close - e.g., 50 mm from the breast detector. So, the breast detection 106 is spaced at least far enough away from the breast that the quality of the image captured by the breast detector 106 is not sacrificed.

[0037] Thus, the sensors may refresh, or otherwise periodically capture sets of contact points. The first plurality of contact points may be selected based on the contact points captured such periodic capture. The breast area used for determining an optimum force, may be the average breast area for all sets of contact points, the smallest breast area (indicating systolic blood pressure) or largest breast area (indicating diastolic blood pressure). This ensures the area of the breast is calculated using consistent factors - e.g., point during a pulse cycle - as the basis for calculation.

[0038] Such a system is shown with the a breast detector 400, with cover 402 on in Figure 4, and with the cover removed in Figure 5, showing the series of sensors 404.

[0039] There are two methods to determine breast area, disclosed herein. The first involves drawing a best fit half ellipse through the contact points to match the organic curvature of a breast and then determining the area from the size of the half ellipse thereby produced. The second method is directly using the area created by the individual sensor points, away from an imaginary line that represents the edge of the mammogram detection area.

[0040] The main method of interaction between user and the system may be via an interface, such as a virtual or physical button that is pressed to send instructions to the system 100 to display the predicted compression force and / or proceed with a subsequent step in the procedure. To that end, the system 100 includes display 120 for displaying the initial compression force and the imaging compression force. The imaging compression force is the force to be applied, during imaging. As shown in Figure 2, during setup (image A) the breast is positioned on the breast support member or mammogram surface. Once the breast is in position, the system can automatically commence detection of the breast (e.g., when the breast itself is detected on the mammogram surface, but no movement of the breast has been detected for a predetermined period of time, such as 5 seconds) or a user may push a button to manually commence breast detection and area calculation (image B). The initial compression force is then determined and applied to the breast (image C) - the initial force may be a predetermined force applied to all patients (e.g., a force that is less than the usual compression force applied during a mammogram), may be a force that depends on age, visually assessed breast size, bra size or other measure. The area of the breast is then calculated again while the breast is subject to the initial compression force (image D). The imaging compression force is then determined (image E) based on the areas calculated while the breast is at rest and while it is under the initial compression force.

[0041] Notably, in the embodiments described herein, the area is detected when no compression force is applied to the breast and when a predetermined compression force is applied to the breast. The same methodology could be applied between area (or volume) measurements, based on a first compression force and a second compression force, where those two forces are unequal. Similarly, the system may calculate volume directly (e.g., from a 3-dimensional point cloud) rather than calculating the area of the breast on the mammogram surface. In calculating the volume, the area will necessarily also be calculated. For example, the system 100 may calculate the volume of the breast for the first plurality of contact points and, separately, for the second plurality of contact points. From that calculation, the imaging compression force is calculated based on the first breast volume and second breast volume.

[0042] Volume may be calculated according to the formula height x area, where the height is the vertical distance between the compression paddles (breast support member 106 and compression member 108) and the area is the area between the line of contact between the breast and breast support member 106, and the edge of the breast support member 106 (mammogram plate) closest to the patient. The area may, in some instances, be approximated by a semicircle. As reflected in Figure 7, alternative approximations may be given for area and volume - e.g., in cubic centimetres, such as n x depth H x height C x width at edge of mammogram plate (breast support member 106) / 4.

[0043] Figure 3 illustrates two ways of determining breast contact area. As the system 100 can have low resolution in some embodiments, a predictive area method is implemented by drawing a half ellipse to form the predicted shape of the breast and to facilitate data smoothing. This can be achieved by considering the coordinates of each contact point in the point cloud map, and drawing an elliptical regression line through the contact points (or through the contact points that deviate from a baseline 300) - e.g., an elliptical least squares fit.

[0044] The main method for determining area is to calculate the area drawn by the sensors (i.e., the elliptical line) to an imaginary or predetermined line marked at the far (distal, with respect to housing 114) edge of the detection plate - e.g., at 280 mm. Clearly, differently sized plates will use a different predetermined or imaginary line distance.

[0045] Breast density can be calculated using any known formula, provided the system 100 can gather the requisite inputs. For example, a formula for breast density estimation may be: where p is the density, a is an adjustment factor, y is the rate of change in area based on the initial force used, AA is the change in breast area between no compression and compression under initial compression force, Acis the compressed breast area (area of breast on application of initial force) and0is the initial, pre-compressed breast area (area of breast with no compression force applied).

[0046] Breast density influences the deformation of the breast during compression. For a breast of specific density, when a known force is applied to it, the rate of change in area is influenced by the force (y) correlated to its density. A final adjustment factor (a) is required for the final calculation. This number is determined empirically, and may be a function - e.g., of age, breast size or dimensions, and other factors. Alternatively, a may comprise multiple values stored in a lookup table or other data structure, the value of a depending again of age, breast size or dimensions and / or other factors.

[0047] The force to be applied can then be determined based on [compression force (N)] = [Pressure (Pa) x Area (m2)], with the pressure set at 10 kPa and the area being determined by the Lidar array.

[0048] The present system, and method implemented by that system, provide an objective compression force for radiographers to use during mammogram imaging. This avoids the need to rely on personal judgement to decide on compression force. This thereby bridges the gap between the experience levels of senior and junior radiographers and introducing more consistency to the procedure of mammogram imaging.

[0049] It will be appreciated that many further modifications and permutations of various aspects of the described embodiments are possible. Accordingly, the described aspects are intended to embrace all such alterations, modifications, and variations that fall within the spirit and scope of the appended claims.

[0050] Throughout this specification and the claims which follow, unless the context requires otherwise, the word "comprise", and variations such as "comprises" and "comprising", will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

[0051] The reference in this specification to any prior publication (or information derived from it), or to any matter which is known, is not, and should not be taken as an acknowledgment or admission or any form of suggestion that that prior publication (or information derived from it) or known matter forms part of the common general knowledge in the field of endeavour to which this specification relates.

Claims

Claims1. A system for use in conducting a mammogram of a breast, comprising: memory; at least one processor; a breast support member; a compression member; and a breast detector that is spaced from the breast, when the breast is on the breast support member; wherein the memory stores instructions that, when executed by the at least one processor, cause the system to: identify, using the breast detector, a first plurality of contact points between the breast and breast support member when the breast is not compressed, and a second plurality of contact points between the breast and breast support member when the breast is compressed between the compression member and breast support member under an initial compression force; determine a first breast area from the first plurality of contact points and a second breast area from the second plurality of contact points; and calculate, based on the first breast area and second breast area, an imaging compression force to be applied between the compression member and breast support member.

2. The system of claim 1, wherein the imaging compression force is calculated to result in a predetermined pressure being applied across the breast, equal to a blood pressure in the breast.

3. The system of claim 1 or 2, wherein calculating the imaging compression force comprises estimating a density of the breast from the first breast area and second breast area.

4. The system of any one of claims 1 to 3, wherein the breast detector comprises a Lidar system.

5. The system of any one of claims 1 to 4, further comprising an actuator, the at least one processor operating the actuator to perform at least one of: move the compression member towards the breast support member, to apply the initial compression force to the breast; and move the compression member towards the breast support member, to apply the imaging compression force to the breast.

6. The system of any one of claims 1 to 5, wherein the at least one processor calculates a first breast volume based on the first breast area and second breast volume based on the second breast area, and the imaging compression force is calculated based on the first breast volume and second breast volume.

7. The system of any one of claims 1 to 6, further comprising a display for displaying the initial compression force and the imaging compression force.

8. A method for determining an imaging compression force for conducting a mammogram of a breast, the method comprising: identifying, using a breast detector, a first plurality of contact points between the breast and breast support member when the breast is not compressed; applying an initial compression force to the breast using the compression member; identifying a second plurality of contact points between the breast and breast support member, under the initial compression force; determining a first breast area from the first plurality of contact points and a second breast area from the second plurality of contact points; andcalculating, based on the first breast area and second breast area, an imaging compression force to be applied between the compression member and breast support member.

9. The method of claim 8, wherein calculating the imaging compression force comprises determining the imaging compression force as a force required to apply a predetermined pressure across the breast, equal to a blood pressure in the breast.

10. The method of claim 8 or 9, wherein calculating the imaging compression force comprises estimating a density of the breast from the first breast area and second breast area.

11. The method of any one of claims 8 to 10, wherein the breast detector comprises a Lidar system.

12. The method of any one of claims 8 to 11, further comprising operating an actuator to perform at least one of: move the compression member towards the breast support member, to apply the initial compression force to the breast; and move the compression member towards the breast support member, to apply the imaging compression force to the breast.

13. The method of any one of claims 8 to 12, wherein calculating the imaging compression force comprises calculating a first breast volume based on the first breast area and second breast volume based on the second breast area, and calculating the imaging compression force based on the first breast volume and second breast volume.

14. The method of any one of claims 8 to 13, further comprising displaying the initial compression force and the imaging compression force on a display.

Images