Object pose classification using joint position coordinates

Abstract

Disclosed herein is a medical instrument (100, 300). Execution of the machine executable instructions cause the processor (106) to receive (206) a set of joint position coordinates (128) for a subject (118) placed on a subject support (120), receive (207) a body orientation (132) in response to inputting the set of joint position coordinates into a predetermined logic module (130), calculate (208) a torso aspect ratio (134) from the set of joint position coordinates. If (210) the torso aspect ratio is greater than a predetermined threshold (136), then (212) the body posture of the subject is a lateral decubitus posture. Execution of the machine executable instructions further cause the processor to assign (220) the body posture as a supine posture if the torso aspect ratio is less than or equal to the predetermined threshold, if the subject is face up on the subject support, or as a prone posture if the subject is face down on the subject support. Execution of the machine executable instructions further cause the processor to generate (216) a subject posture label (142).

Description

TECHNICAL FIELD

[0001] The present invention relates to medical imaging and radiotherapy, in particular to positioning of a subject. BACKGROUND

[0002] In medical imaging modalities such as magnetic resonance imaging and in radiotherapy, it is of utmost importance that the subject is properly positioned before the procedure. Otherwise, the subject can be imaged improperly or the wrong part of the subject can be irradiated.

[0003] US patent application US2015 / 0092998 Al discloses a posture detection method and system. The posture detection method comprises: obtaining bone data of a target person; analyzing the bone data to obtain actual posture information of the target person; and recording the actual posture information of the target person. The posture information is recorded automatically, so that the doctor does not need to manually record the posture information. Therefore, scan omission or wrong scan direction caused by inconsistency between the real posture of the patient and the recorded posture information can be avoided, which ensures the reliability of medical diagnosis. SUMMARY

[0004] The invention provides in independent claims a medical instrument, a computer program product and a method. Embodiments are given in dependent claims.

[0005] Proper determination of the posture of a subject is important to ensure that the correct patient coordinate system is used during medical imaging or radiotherapy. Often, it is difficult to properly classify the posture of a subject. For example, a patient who is injured or ill can have a difficult time-specific posture. Other times, the operator of the medical instrument can incorrectly record the posture of the subject. Embodiments provide a means for automatically generating a subject posture label for a subject on a subject support. A set of joint position coordinates are received. These can be coordinates indicative of the position of the major joints of the skeletal system of the subject. The relative positions of the joint position coordinates enable a determination to be made as to whether the subject is feet first or head first on the subject support. A predetermined logic module can be programmed to make this determination.

[0006] In some examples, the determination as to whether the subject is feet first or head first can be relative to a selected coordinate system. In other examples, the direction of feet first or head first can be determined relative to a coordinate system of a radiotherapy system, a direction towards the radiotherapy system, or relative to a direction of motion of a subject support to transport the subject into or to the radiotherapy system.

[0007] Determining whether the subject is in a decubitus, supine or prone position can be a difficult task. In a decubitus position, the subject is on their side, and the decubitus position can sometimes be partially prone or supine. To make this determination, embodiments use the positions of the left hip joint, right hip joint, right shoulder joint and left shoulder joint.

[0008] A torso aspect ratio can be calculated. The positions of the left hip joint, right hip joint, right shoulder joint and left shoulder joint define a quadrilateral. The long side of the quadrilateral is defined by the left hip joint to the left shoulder joint and / or the right hip joint to the right shoulder joint. The short side of the quadrilateral is defined by the left hip joint to the right hip joint and / or the right shoulder joint to the left shoulder joint. The torso aspect ratio is the ratio of the length of the long side(s) of the quadrilateral to the length of the short side(s) of the quadrilateral. For the length of the long side, the length of the left hip joint to the left shoulder joint, the length of the right hip joint to the right shoulder joint or an average of the two can be used. For the length of the short side, the length of the left hip joint to the right hip joint, the length of the right shoulder joint to the left shoulder joint or an average of the two can be used.

[0009] If the torso aspect ratio is above a predetermined threshold, the subject is determined to be in a decubitus position. If either the hip or shoulder left and right partial position probabilities overlap and so the ratio starts to diverge, this will correspond to a strict decubitus position. If the subject is less than or equal to a predetermined threshold, the subject is in a supine or prone position. The predetermined logic module can be programmed to take as input the set of joint position coordinates and output whether the subject is face down or face up on the subject support.

[0010] In testing, using the golden ratio (or golden section) as the predetermined threshold was effective. In this case, the predetermined threshold would be approximately 1.6. In some examples, the predetermined threshold can be between 1.5 and 1.8.

[0011] In one aspect, the present application provides a medical instrument comprising: a memory storing machine executable instructions. The medical instrument further comprises a processor configured to control the medical instrument.

[0012] Depending on the configuration of the medical instrument, the processor can provide different types of control. In some examples, the medical instrument is a computer workstation or remote system available via the internet or as a cloud service. In this case, the control of the medical instrument is to provide computation, digital and / or image processing tasks. In other examples, the medical instrument can comprise additional components such as a camera system and / or a radiotherapy system. In this case, the control of the medical instrument by the processor comprises the control of these additional components.

[0013] The execution of the machine executable instructions causes the processor to receive a set of joint position coordinates for a subject placed on a subject support. A subject, such as a human, has a skeleton connected by joints. The joint position coordinates are indicative of the position of the joints of the skeletal system of the subject. Typically, only the major joints are indicated. The set of joint position coordinates comprises coordinates for a left hip joint, a right hip joint, a right shoulder joint and a left shoulder joint.

[0014] The execution of the machine executable instructions further causes the processor to identify a body orientation by inputting the set of joint positions to a predetermined logic module. The body orientation is head first or feet first. The identification of the body orientation with the predetermined logic module can be achieved by knowing that some joints are closer to the head or feet than other joints. For example, the right shoulder joint is closer to the head of the subject than the right hip joint. Knowing the position of the right hip joint and the right shoulder joint then indicates whether the body orientation is head first or feet first. This can be achieved, for example, by using different combinations of joint position coordinates.

[0015] As used herein, face up and face down refer to the orientation of the torso of the subject. When the subject is face up on the subject support, the back of the subject is resting on the subject support. When the subject is face down on the subject support, the chest of the subject is resting on the subject support.

[0016] In one example, if the subject is viewed from above, a point between the left shoulder joint, the right shoulder joint, the left hip joint and the right hip joint can be selected. If a person rotates about this point, the right shoulder joint will always be in a clockwise direction (as viewed from above) from the left shoulder joint. Likewise, the left shoulder joint will always be in a clockwise direction (as viewed from above) from the right shoulder joint.

[0017] Various other logical conditions can also be constructed that are similar to these two. In fact, by selecting any three of the left shoulder joint, the right shoulder joint, the left hip joint, and the right hip joint and applying logic, an equivalent predetermined logic can be used to determine whether the object is facing up or down. It can be easier in some cases to use three of the predetermined logic modules, as three points automatically define a plane in which the three selected joints can rotate.

[0018] As a specific example, the left shoulder joint, the right shoulder joint, and the right hip joint are used. These three points define a triangle on a plane. If a center point such as the centroid of the triangle is taken, when the object is facing down, as viewed from above, the right shoulder joint is always adjacent to it and in a clockwise direction when coming from the left shoulder joint, and the right hip joint is always adjacent to it and in a clockwise direction when coming from the right shoulder joint. The left shoulder joint is always adjacent to it and in a clockwise direction when coming from the right hip joint. It is clear from this example that there are many variations to this that can be used to provide logical conditions for the predetermined logic module.

[0019] The execution of the machine executable instructions further causes the processor to calculate a torso aspect ratio from the positions of the left hip joint, the right hip joint, the right shoulder joint, and the left shoulder joint. These four joint position coordinates define a quadrilateral. The torso aspect ratio can be calculated, for example, as described above.

[0020] The execution of the machine executable instructions further causes the processor to determine whether the torso aspect ratio is greater than a predetermined threshold. If this is the case, the body posture of the object is identified as a side lying posture. This is the posture of the object on the right or left side. The execution of the machine executable instructions further causes the processor to determine whether the aspect ratio is less than or equal to the predetermined threshold, and then identify whether the body posture of the object is facing up or down. If the orientation of the object is facing down, the body orientation is identified as a prone posture. If the orientation of the object is facing up, the body orientation is defined as a supine posture.

[0021] The determination of whether the subject is face up or face down can be determined again using the predetermined logic module. Knowing whether the subject is feet first or head first, the position of the individual joint position coordinates can be used to determine whether the subject is face down or face up. This then defines whether the subject is in a prone or supine position. Execution of the machine executable instructions further cause the processor to generate a subject posture label based at least on the body orientation and the body posture. The subject posture label can be useful, for example, in control systems for controlling imaging and radiotherapy applications and medical instruments that mark medical images acquired with medical imaging systems. Using the torso aspect ratio to determine whether the subject is in a lateral position can be useful because often subjects are not completely supine or lateral. This provides an effective means of distinguishing between various postures.

[0022] In another embodiment, the set of joint position coordinates includes a left knee joint, a right knee joint, and a neck joint. The left knee joint coordinate includes a left knee joint angle having a first restricted range of motion; the right knee joint includes a right knee joint having a second restricted range of motion. The neck joint includes a neck joint angle having a third restricted range of motion. These three joints are only able to move within a limited range of angles. This knowledge of the restricted ranges of motion can be used to identify the position of the subject. By inputting the first restricted range of motion, the second restricted range of motion, and the third restricted range of motion into the predetermined logic module, the lateral position is classified as a left lateral position or a right lateral position. Having knowledge of the possible angles that the joints can take for a subject in a left or right lateral position enables the knowledge of these restricted ranges of motion to determine whether the subject is in a left lateral position or a right lateral position.

[0023] In another embodiment, execution of the machine executable instructions further cause the processor to calculate a lateral position measure based on a difference between the predetermined threshold and the torso aspect ratio. The subject posture label further includes a lateral position measure. For example, the difference between the predetermined threshold can be input into an algorithm that returns a number that indicates how close the posture of the subject is to a lateral position. This can be useful to know whether the subject is having difficulty assuming a lateral position.

[0024] In another embodiment, the memory further contains a joint localizer module configured to construct the set of joint position coordinates in response to receiving an image of a subject on the subject support. Execution of the machine executable instructions further cause the processor to: receive an image of a subject on the subject support; and receive the set of joint position coordinates in response to inputting the image into the joint localizer module.

[0025] In another embodiment, the memory further includes a neural network configured to output, in response to receiving an image of an object on the object support, a separate joint position probability map for each joint position coordinate of a set of joint position coordinates. Execution of the machine executable instructions causes the processor to receive an image of an object on the object support. Execution of the machine executable instructions further causes the processor to receive, by inputting the image into the neural network, a separate joint position probability map for each of the set of joint positions. Execution of the machine executable instructions further causes the processor to calculate the set of joint position coordinates from the separate joint position probability map for each of the set of joint positions.

[0026] Determining the position of an individual joint can be extremely difficult when an object is on an object support. The neural network is not used to provide a particular coordinate. Rather, each of the set of joint position coordinates has its own probability map output. The probability map can then be used to determine the joint position. For example, the voxel or pixel with the highest value can be selected as the joint position. In other examples, the overall distribution can be examined and instead the centroid or mean or average position can be selected. This can provide a more robust means of identifying the position of a joint of an object.

[0027] In another embodiment, the probability map for each of the set of joint positions has the same pixel scale as the image of the object on the object support. This can be beneficial in identifying the position of a joint of an object in the image.

[0028] In another embodiment, the medical instrument includes a camera. Execution of the machine executable instructions further causes the camera to capture the image of the object on the object support.

[0029] In another embodiment, execution of the machine executable instructions further causes the processor to train the neural network with a set of labeled images. The labeled images may, for example, include images of objects with the positions of various joints indicated in the image. The set of labeled images includes images of: an object placed on the object support, a front view of the object, a back view of the object, an empty object support, an object support including medical equipment, and a partially occluded image of the object. This, in combination with generating a separate joint position probability map for each joint, can result in a system that better identifies the set of joint position coordinates than existing systems.

[0030] In another embodiment, the medical instrument further comprises a medical imaging system configured for acquiring medical imaging data from an imaging zone. The subject support is configured for at least partially supporting a subject within the imaging zone. A feet first label can indicate that the feet enter the imaging zone first when the subject moves into the imaging zone. Likewise, a head first label indicates that a head region of the subject moves into the imaging zone first when the subject moves into the imaging zone.

[0031] In another embodiment, the medical imaging system is a magnetic resonance imaging system.

[0032] In another embodiment, the medical imaging system is a positron emission tomography system.

[0033] In another embodiment, the medical imaging system is a single photon emission tomography system.

[0034] In another embodiment, the medical imaging system is a digital X-ray system.

[0035] In another embodiment, the medical imaging system is a computed tomography system or CT system.

[0036] In another embodiment, the memory further comprises a medical imaging protocol. The medical imaging protocol may, for example, contain instructions or commands that can be translated into instructions that can be used to control the medical imaging system to acquire medical imaging data. The medical imaging protocol comprises a selected posture label. Execution of the machine executable instructions further causes the processor to, in case the subject posture label is different from the selected posture label, perform any of the following: provide a warning signal; select a different medical imaging protocol match comprising the subject posture label, provide an instruction to reposition the subject; and a combination thereof. This embodiment can be beneficial because it can prevent the subject from being incorrectly imaged in the medical imaging system.

[0037] In another embodiment, execution of the machine executable instructions further causes the processor to control the medical imaging system to acquire medical imaging data. This can be done, for example, using commands or instructions contained in the medical imaging protocol. Execution of the machine executable instructions further causes the processor to create an image file comprising the medical imaging data and the subject posture label. For example, the image file can be a DICOM image. Execution of the machine executable instructions further causes the processor to store the image file in a picture archiving and communication system. This embodiment can be beneficial because it ensures that the position of the subject is stored together with the acquired medical imaging data.

[0038] In another embodiment, the medical instrument further comprises a radiotherapy system configured for irradiating a target within a radiation zone. The subject support is configured to at least partially support the subject within the radiation zone. The addition of such a radiation zone can be beneficial because the subject posture of the subject during radiotherapy is of utmost importance for proper treatment.

[0039] In another embodiment, the memory further comprises a radiotherapy protocol. The radiotherapy protocol comprises a selected posture label. Execution of the machine executable instructions further causes the processor to, in the event that the subject posture label is different from the selected posture label, perform any of the following: provide a warning signal; stop execution of the radiotherapy protocol; provide instructions for repositioning the subject; and combinations thereof. This embodiment can be beneficial because it can prevent the subject from being incorrectly irradiated.

[0040] In another embodiment, the set of joint position coordinates further comprises: a left elbow joint coordinate and a right elbow joint coordinate. The left elbow joint coordinate comprises a left elbow joint angle. The right elbow joint comprises a right elbow joint angle. The left shoulder joint coordinate comprises a left shoulder joint angle. The right shoulder joint coordinate comprises a right shoulder joint angle.

[0041] Execution of the machine executable instructions further causes the processor to determine an arm position classification by inputting the right shoulder joint angle, the left shoulder joint angle, the right elbow joint angle, and the left elbow joint angle into the predetermined logic module. The arms of a subject can only bend in certain directions as defined by the elbows and shoulders. Thus, the predetermined logic module can be programmed to use this information to determine an arm position classification. Execution of the machine executable instructions further causes the processor to append the arm position classification to the subject posture label. This can be beneficial because it can provide more detail about the position of the subject during radiotherapy or imaging.

[0042] In another aspect, the present invention provides a method of medical imaging. The method comprises receiving a set of joint position coordinates from a subject placed on a subject support. The set of joint position coordinates comprises coordinates for: a left hip joint, a right hip joint, a right shoulder joint, and a left shoulder joint. The method further comprises identifying a body orientation by inputting the set of joint positions into a predetermined logic module. The body orientation is head first or feet first. The method further comprises calculating a torso aspect ratio from the positions of the left hip joint, the right hip joint, the right shoulder joint, and the left shoulder joint. The method further comprises identifying a body posture of the subject as a prone posture if the torso aspect ratio is below a predetermined threshold.

[0043] The method further comprises identifying a body posture of the subject as a prone position according to the predetermined logic module if the torso aspect ratio is higher than the predetermined threshold and the body orientation is the subject positioned face up on the subject support. In this case, the predetermined logic module will take the set of joint position coordinates as input.

[0044] The method further comprises determining a body posture of the subject is a supine position if the torso aspect ratio is higher than the predetermined threshold and the predetermined logic module determines the subject is face down. This can be achieved, for example, by inputting the set of joint position coordinates into the predetermined logic module. The method further comprises generating a subject posture label at least according to the body orientation and the body posture.

[0045] In another aspect, the application provides a computer program product comprising machine executable instructions for execution by a processor controlling a medical system. Execution of the machine executable instructions further causes the processor to receive a set of joint position coordinates for a subject positioned on a subject support. The set of joint position coordinates comprises coordinates for a left hip joint, a right hip joint, a right shoulder joint and a left shoulder joint. Execution of the machine executable instructions further causes the processor to identify a body orientation by inputting the set of joint positions into a predetermined logic module. The body orientation is either head first or feet first.

[0046] Execution of the machine executable instructions further causes the processor to calculate a torso aspect ratio from the positions of the left hip joint, the right hip joint, the right shoulder joint and the left shoulder joint. Execution of the machine executable instructions further causes the processor to determine a body posture of the subject is a lateral position if the torso aspect ratio is greater than a predetermined threshold.

[0047] Execution of the machine executable instructions further causes the processor to determine a body posture of the subject is a prone position if the aspect ratio is less than or equal to the predetermined threshold and the predetermined logic module determines the subject is face down. The set of joint position coordinates can be input into the predetermined logic module to determine this.

[0048] Execution of the machine executable instructions further causes the processor to determine a body posture of the subject is a supine position if the aspect ratio is less than or equal to the predetermined threshold and the predetermined logic module determines the subject is face up. The set of joint position coordinates can be input into the predetermined logic module to determine this. Execution of the machine executable instructions further causes the processor to generate a subject posture label at least according to the body orientation and the body posture.

[0049] In another aspect, the method is implemented as machine executable instructions or code for execution by a processor that controls a medical instrument.

[0050] It is to be understood that one or more of the above-described embodiments of the present application can be combined, as long as the combined embodiments are not mutually exclusive.

[0051] As will be appreciated by one of skill in the art, aspects of the present application can be embodied as a device, a method or a computer program product. Accordingly, aspects of the present application can take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects (in general all of which can be referred to herein as a "circuit", "module" or "system"). Furthermore, aspects of the present application can take the form of a computer program product on one or more computer readable medium(s) having computer executable code embodied thereon.

[0052] Any combination of one or more computer readable medium can be utilized. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. "Computer readable storage medium" as used herein encompasses any tangible storage medium which can store instructions which are executable by a processor of a computing device. The computer readable storage medium can be referred to as a computer readable non-transitory storage medium. The computer readable storage medium can also be referred to as a computer readable tangible medium. In some embodiments, a computer readable storage medium can also be able to store data which is able to be accessed by a processor of a computing device. Examples of computer readable storage media include, but are not limited to: a floppy disk, a magnetic hard disk drive, a solid state hard disk, flash memory, a USB thumb drive, Random Access Memory (RAM), Read Only Memory (ROM), an optical disk, a magneto-optical disk, and the register file of the processor. Examples of optical disks include Compact Disks (CDs) and Digital Versatile Disks (DVDs), such as CD-ROM, CD-RW, CD-R, DVD-ROM, DVD-RW, or DVD-R disks. The term computer readable storage medium also refers to various types of recording media capable of being accessed by a computer device. For example a recording medium of a computer program can be a modulated carrier signal on a network or a local wired network. Computer executable code embodied on one or more computer readable medium can be sent over a network using a transmission medium or a carrier wave. Such delivery can be performed using a transmission medium or carrier wave including the Internet, wireline, optical, radio frequency (RF), etc. or any suitable combination thereof.

[0053] A computer readable signal medium can include a propagated data signal with computer executable code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal can take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium can be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate or transport program for use by or in connection with an instruction execution system, apparatus, or device.

[0054] “Computer memory” or “memory” is an example of a computer readable storage medium. Computer memory is any memory accessible by a processor. “Computer storage” or “storage” is another example of a computer readable storage medium. Computer storage is any non-volatile computer readable storage medium. In some embodiments, computer storage can also be computer memory, or vice versa.

[0055] “Processor” as used herein encompasses an electronic component which acts to interpret and execute instructions. Reference to a computing device including “processor” should be interpreted as possibly containing more than one processor or processing core. The processor can for example be a multi-core processor. Processor can also refer to a collection of processors within a single computer system or distributed across multiple computer systems. The term computing device should also be interpreted to possibly refer to a collection or network of computing devices each including one or more processors. Computer executable code can be executed by a single processor or distributed across multiple processors which can be within the same computing device or even distributed across multiple computing devices.

[0056] Computer executable code can comprise machine executable instructions or a program which causes a processor to perform an aspect of the present application. Computer executable code for carrying out operations for aspects of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer executable code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (“LAN”) or a wide area network (“WAN”), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0057] The computer executable code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0058] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0059] These computer program instructions can also be stored in a computer readable medium that can direct a computer, other programmable data processing apparatus, or other devices to function in a particular manner, such that the instructions stored in the computer readable medium produce an article of manufacture including instructions which implement the function / act specified in the flowchart and / or block diagram block or blocks.

[0060] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0061] A "user interface" as used herein is an interface that allows a user or operator to interact with a computer or computer system. A "user interface" can also be referred to as a "human interface device." A user interface can provide information or data to the operator and / or receive information or data from the operator. A user interface can enable input from an operator to be received by the computer and can provide output to the operator from the computer. In other words, the user interface can allow an operator to control or manipulate a computer, and the interface can allow the computer to indicate the effects of the operator's control or manipulation. Display of data or information on a display or graphical user interface is an example of providing information to an operator. Receiving data through a keyboard, mouse, trackball, touchpad, pointing stick, graphics tablet, joystick, gamepad, webcam, microphone, pedal, wired glove, remote control, and accelerometer are all examples of user interface components that enable receipt of information or data from an operator.

[0062] A "hardware interface" as used herein encompasses an interface that enables a processor of a computer system to interact with and / or control an external computing device and / or apparatus. A hardware interface can allow the processor to send control signals or instructions to the external computing device and / or apparatus. A hardware interface can also enable the processor to exchange data with the external computing device and / or apparatus. Examples of hardware interfaces include, but are not limited to: a universal serial bus, an IEEE 1394 port, a parallel port, an IEEE 1284 port, a serial port, an RS-232 port, an IEEE-488 port, a Bluetooth connection, a wireless local area network connection, a TCP / IP connection, an Ethernet connection, a control voltage interface, a MIDI interface, an analog input interface, and a digital input interface.

[0063] A "display" or "display device" as used herein encompasses an output device or user interface suitable for displaying images or data. A display can output visual, audio, and / or tactile data. Examples of displays include, but are not limited to: a computer monitor, a television screen, a touchscreen, a tactile electronic display, a Braille screen, a cathode ray tube (CRT), a storage tube, a bistable display, e-paper, a vector display, a flat-panel display, a vacuum fluorescent display (VF), a light-emitting diode (LED) display, an electroluminescent display (ELD), a plasma display panel (PDP), a liquid crystal display (LCD), an organic light-emitting diode display (OLED), a projector, and a head-mounted display.

[0064] Magnetic resonance (MR) data is defined herein as the measurements of radio frequency signals emitted by atomic spins recorded using the antenna of a magnetic resonance apparatus during a magnetic resonance imaging scan. Magnetic resonance data is an example of medical image data. A magnetic resonance imaging (MRI) image or MR image is defined herein as a reconstructed two- or three-dimensional visualization of anatomical data contained within magnetic resonance imaging data. This visualization can be performed using a computer. BRIEF DESCRIPTION OF DRAWINGS

[0065] A preferred embodiment of the present application will now be described, by way of example only, and with reference to the accompanying drawings in which:

[0066] Figure 1 An example of a medical instrument is illustrated;

[0067] Figure 2 A flowchart of a method of a medical instrument is shown illustrating operation Figure 1 ;

[0068] Figure 3 Another example of a medical instrument is illustrated;

[0069] Figure 4 An example of a neural network is illustrated;

[0070] Figure 5 An example of an image of an object overlaid with a set of joint position coordinates is shown; and

[0071] Figure 6 Another example of an image of an object overlaid with a set of joint position coordinates is shown.

[0072] LIST OF REFERENCE NUMBERS

[0073] 100 medical instrument

[0074] 102 computer

[0075] 104 hardware or network interface

[0076] 106 processor

[0077] 108 user interface

[0078] 110 memory

[0079] 120 machine executable instructions

[0080] 122 image of an object (optional)

[0081] 124 neural network

[0082] 126 joint position probability map

[0083] 128 set of joint position coordinates

[0084] 130 predetermined logic module

[0085] 132 body orientation (head or feet first)

[0086] 134 torso aspect ratio

[0087] 136 predetermined threshold value

[0088] 138 body posture

[0089] 140 arm position classification (optional)

[0090] 142 object posture label

[0091] 200 receive an image of an object on an object support

[0092] 202 receive, in response to inputting the image into a neural network, a separate joint position probability map

[0093] 204 compute, from the separate joint position probability map for each of the set of joint positions, the set of joint position coordinates

[0094] 206 receive a set of joint position coordinates for an object placed on an object support

[0095] 207 receive, in response to inputting the set of joint position coordinates into a predetermined logic module, a body orientation

[0096] 208 compute, from the positions of the left hip joint, the right hip joint, the right shoulder joint, and the left shoulder joint, a torso aspect ratio 210 is the torso aspect ratio higher than a predetermined threshold value?

[0097] 212 assign a side lying posture to the body posture

[0098] 214 determine an arm position

[0099] 216 generate an object posture label

[0100] 218 is the object on the object support face up or face down?

[0101] 220 the body posture is a supine posture

[0102] 222 the body posture is a prone posture

[0103] 300 medical system

[0104] 302 magnetic resonance imaging system

[0105] 304 magnet

[0106] 306 bore of the magnet

[0107] 308 imaging zone

[0108] 309 field of view

[0109] 310 magnetic field gradient coil

[0110] 314 radio frequency coil

[0111] 316 transceiver

[0112] 318 subject

[0113] 320 subject support

[0114] 322 camera

[0115] 330 pulse sequence command

[0116] 332 magnetic resonance data

[0117] 334 magnetic resonance image

[0118] 336 dicom image

[0119] 500 right ankle joint coordinate

[0120] 502 left ankle joint coordinate

[0121] 504 right knee joint coordinate

[0122] 504 left knee joint coordinate

[0123] 508 right hip joint coordinate

[0124] 508 left hip joint coordinate

[0125] 512 right shoulder joint coordinate

[0126] 514 left shoulder joint coordinate

[0127] 516 right elbow joint coordinate

[0128] 518 left elbow joint coordinate

[0129] 520 right wrist joint coordinate

[0130] 522 left wrist joint coordinate

[0131] 524 neck joint coordinate DETAILED DESCRIPTION

[0132] Like numbered elements or components in these drawings have the same or equivalent functions. If the function is equivalent, elements previously discussed will not be repeated in the subsequent drawings.

[0133] Figure 1 An example of a medical instrument 100 is illustrated. Figure 1 The medical instrument 100 in the figure comprises a computer 102. The computer 102 comprises a hardware or network interface 104 for connecting to other components of the medical instrument and / or for networking with other computer systems. The computer 102 is further shown to contain a processor 106. The processor is also connected to the hardware or network interface 104. Furthermore, there is a memory 110 and an optional user interface 108 that are also connected to the processor.

[0134] The memory 110 is shown to contain machine executable instructions 120 that enable the processor 106 to control the operation and functionality of the medical instrument 100 and to perform various data and imaging processing tasks. The memory 110 is further shown to contain an image of an object 122. In some examples, the image of an object 122 can be optional. The memory 110 is further shown to contain a neural network 124 that is configured for receiving the image of an object 122 and outputting a joint position probability map 126 for each member of a set of joint position coordinates 128. In some examples, the neural network 124 and the key position probability map 126 can be optional. The joint position probability map 126 is an individual probability map that each contains probabilities of a particular joint of the object.

[0135] The joint position probability map 126 can be processed to derive or generate a set of joint position coordinates 128. The memory 110 is further shown to contain a predetermined logic module 130. The set of joint position coordinates 128 is a defined position of the object. By using the predetermined logic, the information contained in the set of joint position coordinates 128 can be used to derive properties of the object; for example, the position of the head relative to the feet. Furthermore, the information can be used to determine whether the object is facing up or down on the object support. Various joints of the object have a limited range of motion. This limited range of motion can also be useful for inferring the position or orientation of the object. For example, a knee joint and an elbow joint only bend in a certain direction within a predetermined range for the object. For example, a determination of the angle of a particular knee joint or elbow joint can be used to determine the way the object is lying or even the arm position.

[0136] The memory 110 is also shown to contain a body orientation 132, which indicates whether the subject is head first or feet first on the subject support. This can be determined by inputting the set of joint position coordinates 128 into a predetermined logic module 130. The memory 110 is also shown to contain a trunk aspect ratio. The trunk aspect ratio was previously described to be determined from the positions of the hip joint and the shoulder joint. The trunk aspect ratio 134 can be compared to a predetermined threshold 136. In some examples, this can for example be equal to the golden section. If the trunk aspect ratio 134 is below this predetermined threshold 136, then in some examples the body posture 138 can be inferred to be a lateral lying posture. If the trunk aspect ratio 134 is above the predetermined threshold 136, then the body posture is a supine or prone posture.

[0137] The predetermined logic module 130 can be used to determine whether the subject is facing upwards or downwards. The body posture can then be determined to be in a supine or prone posture using the predetermined logic module 130. The angle of the knee joint of a healthy or normal subject can only be bent within a certain angular range. Therefore, the angle of the knee joint can be useful for example in determining whether the subject is in a left or right lateral lying posture. The shoulder joint and elbow joint are also bent within a predetermined range. The predetermined logic module 130 can also be used to determine a hand arm position classification 140. In some examples, this can be optional. The memory is then also shown to contain a subject posture label 142 determined from the body posture 138, the optional hand arm classification 140 and the body orientation 132.

[0138] Figure 2 A flowchart of a method of a medical instrument 100 is shown, which illustrates the operation Figure 1 of the medical instrument 100. First, in step 200, an image 122 of a subject on a subject support is received. Next, in step 202, a joint position probability map 126 is received by inputting the image 122 into a neural network 124. Then, in step 204, a set of joint position coordinates 128 is calculated from the joint position probability map 126. As mentioned previously, different statistical analyses can be applied on the distribution shown in the joint position probability map 126 to calculate the set of joint position coordinates 128. Then, in step 206, the set of joint position coordinates 126 is received. Next, in step 207, a body orientation 132 is identified by inputting the set of joint position coordinates 128 into a predetermined logic module 130. In step 208, a trunk aspect ratio 134 is calculated from the positions of the left hip joint, the right hip joint, the right shoulder joint and the left shoulder joint.

[0139] The method then proceeds to block 210. In 210 the question is whether the torso aspect ratio is above a predetermined threshold. If the answer is yes, the subject is in a side lying position and the method proceeds to step 212. In step 212 it is determined whether the subject is in a left or right side lying position. This can be achieved by inputting the set of joint position coordinates 128 into a predetermined logic module 130. Next the method proceeds to step 214. In some examples step 214 is optional. In step 214 a hand position classification 140 is determined by inputting the set of joint position coordinates 128 into a predetermined logic module 130. After step 214 is performed the method proceeds to step 216. In step 216 a subject posture label 142 is generated from the body orientation 132 and at least the body posture 138. In some examples the hand position classification 140 is also used in the generation of the subject posture label 142.

[0140] Returning to step 210, if the answer to the question is no, the method proceeds to step 218. Step 218 is a different decision block and the question is whether the subject is face up or face down. As described previously, the answer to this question can be obtained by inputting the set of joint position coordinates 128 into a predetermined logic module 130. If the answer is face up, the method proceeds to step 220 and the body posture is identified as a supine posture. If the answer is face down, the method proceeds to step 222 and the body posture is identified as a prone posture. After step 220 or 222 is performed, the method again proceeds to step 214 and then to step 216, as described previously.

[0141] Figure 3 Another example of a medical system 300 is illustrated. In addition to the medical system comprising a magnetic resonance imaging system 302, Figure 3 The medical system 300 in Figure 1 The medical apparatus 100 in

[0142] The magnetic resonance imaging system 302 comprises a magnet 304. The magnet 304 is a superconducting cylindrical type magnet with a bore 306 through it. It is also possible to use different types of magnets: for example, it is also possible to use both split cylindrical magnets and so-called open magnets. A split cylindrical magnet is similar to a standard cylindrical magnet, except that the cryostat has been split into two sections to allow access to the isoplanar plane of the magnet, such a magnet can for example be used in conjunction with charged particle beam therapy. An open magnet has two magnet sections, one on top of the other, with a large enough space in between to receive a subject: the arrangement of the two sections is similar to the arrangement of Helmholtz coils. Open magnets are common because the subject is less constrained. Inside the cryostat of the cylindrical magnet there is a series of superconducting coils.

[0143] Within the bore 306 of the cylindrical magnet 304, there is an imaging zone 308 in which the magnetic field is strong and uniform enough to perform magnetic resonance imaging. The magnetic resonance data is typically acquired for a field of view. Within the view of the camera 322, a subject 318 is shown as being supported by a subject support 320.

[0144] The camera 322 is shown in a position such that the subject 318 can be imaged when placed on the subject support 320. In this example, the subject 318 is positioned head first. When the subject support 320 is used to move the subject 318 into the imaging zone 308, the head of the subject will enter the bore 306 of the magnet 304 first.

[0145] Within the bore 306 of the magnet, there is also a set of magnetic field gradient coils 310 that are used to acquire preliminary magnetic resonance data to spatially encode magnetic spins within the imaging zone 308 of the magnet 304. The magnetic field gradient coils are connected to a magnetic field gradient coil power supply. The magnetic field gradient coils 310 are meant to be representative. Typically, the magnetic field gradient coils include three separate sets of coils to spatially encode in three orthogonal spatial directions. The magnetic field gradient coil power supply supplies current to the magnetic field gradient coils 310. The current supplied to the magnetic field gradient coils 310 is controlled as a function of time and can be ramped or pulsed.

[0146] Adjacent to the imaging zone 308 is a radio frequency coil 314 that is used to manipulate the orientation of magnetic spins within the imaging zone 308 and to receive radio transmissions from spins that are also within the imaging zone. The radio frequency antenna can include multiple coil elements. The radio frequency antenna can also be referred to as a channel or an antenna. The radio frequency coil 314 is connected to a radio frequency transceiver 316. The radio frequency coil 314 and the radio frequency transceiver 316 can be replaced by separate transmit and receive coils and separate transmitters and receivers. It should be understood that the radio frequency coil 314 and the radio frequency transceiver 316 are representative. The radio frequency coil 314 is also meant to represent a dedicated transmit antenna and a dedicated receive antenna. Likewise, the transceiver 316 can also represent separate transmitters and receivers. The radio frequency coil 314 can also have multiple receive / transmit elements and the radio frequency transceiver 316 can have multiple receive / transmit channels. For example, if parallel imaging techniques such as SENSE are performed, the radio frequency coil 314 will have multiple coil elements.

[0147] The transceiver 316 and the gradient controller 312 are shown connected to the hardware interface 104 of the computer system 102.

[0148] The memory 110 is also shown to contain pulse sequence commands 330. The pulse sequence commands can for example contain tags that can be compared with the subject pose tags 142. This can be used as a quality control check. In this example, the pulse sequence commands 330 can be considered as a protocol. The memory 110 is also shown to contain magnetic resonance data 332, which is acquired by controlling the magnetic resonance imaging system 302 with the pulse sequence commands 330. The memory 110 is also shown to contain magnetic resonance images 334 reconstructed from the magnetic resonance data 332. The memory 110 is also shown to contain DICOM images 336 constructed from the magnetic resonance images 334 and also stores the subject pose tags 142.

[0149] Figure 4 The operation of the neural network 124 is illustrated. The image 122 of the subject is shown to be input into the neural network 124. In response, the neural network 124 outputs a number of joint position probability maps 126. The number of joint position probability maps 126 will match the number of elements in the set of joint position coordinates 128.

[0150] For MR or CT examinations, the appropriate patient position and orientation is preferably entered by the technician to ensure the correct patient coordinate system for diagnosis, for example to ensure the correct lateralization of anatomical structures for diagnosis. Setting the patient is a dynamic process, which depends on the level of compliance of the patient. Therefore, deviations from the prescribed patient position and orientation are very common and require the attention of the technician not to forget the corresponding changes in the examination protocol. Failure to do so can have serious consequences for the diagnosis and the responsible technician. Therefore, the automatic detection of the actual patient position and orientation is of particular interest. For this purpose, camera images can be used. Due to the necessary amount of marker data, the classification of the pose from such images using a neural network is challenging. For intermediate poses, the definition of the pose class is sometimes difficult. The extension of the pose class is also challenging.

[0151] The example can perform the classification of the pose using the position of the body joint positions (joint position coordinates 128). Body joints represent the degrees of freedom of movement of the human body. Such body joint positions allow the calculation of joint angles and apparent projection body proportions. The example can use any of the following predetermined logic:

[0152] - determine the relative position of the upper body joints with respect to the lower body joints

[0153] o derive a patient head first vs. feet first probability

[0154] - calculate the apparent torso proportions and determine deviations from predetermined thresholds 136. The golden section can be used as a predetermined threshold.

[0155] o the degree of deviation from the predetermined threshold, i.e. the golden section, gives the degree of lateral decubitus pose

[0156] o If side lying position: Use knee, hip and neck angles to determine side (angle with natural movement restrictions)

[0157] o If not side lying: Decide supine vs. prone based on shoulder side and hip side position

[0158] - Determine arm up or down position based on shoulder angle

[0159] All thresholds can be adjusted and it is possible to determine positions including head first supine / prone, left / right side lying, left / right arm up / down.

[0160] The algorithm works with ordinary camera images, i.e. no 3D information is required. If 3D information is available, it can be used for cross-checking. However, it was found that the combination of joint angles and body proportions provides a more robust signature to derive the patient position. Since all important joints can be located, the position categories can be extended and can be adapted to the needs of the modality. Transitional categories are also possible: for example, the algorithm allows for a category 65% prone / 35% left side lying, which can be useful for a more precise positioning of the patient, e.g. for radiotherapy applications.

[0161] Figure 5 An example of an image 122 of an object 318 on an object support 320 is illustrated. The object is in a feet first position. The feet will enter the magnetic resonance imaging system first. In addition, the object is in a supine position and facing upwards. The right arm of the object is folded across the chest and the left arm is raised over the head. Joint angle coordinates are also marked. Right ankle joint coordinate 500, left ankle joint coordinate 502, right knee joint coordinate 504, left knee joint coordinate 506, right hip joint coordinate 508, left hip joint coordinate 510, right shoulder joint coordinate 512, left shoulder joint coordinate 514, right elbow joint coordinate 516, left elbow joint coordinate 518, right wrist joint coordinate 520, left wrist joint coordinate 522, and neck joint coordinate 524 are all visible. The label for this position is marked as 142. The labels are: feet first (FF), supine (S), left arm up (LAU) and right arm down (RAD). Figure 5 A typical supine position, feet first, is shown. Here, the apparent torso proportions are very close to the golden ratio.

[0162] Figure 6 Another example of an image 122 of an object on an object support is illustrated. In this example, the object is in a side lying position. It can be seen how the angles around the knee joints 504, 506 can be used to determine in which direction the object is turned. The label 142 is: side lying position (D) right (R), left arm down (LAD), and right arm up (RAU). Figure 6A typical knee, hip and neck angle for a side lying position is illustrated. The position is successfully classified as head first, left side lying, left arm down, right arm up. Note the apparent torso proportions, i.e. the ratio between shoulder-hip-distance / shoulder-span (and / or hip-span).

[0163] While the application has been illustrated and described in detail in the drawings and foregoing description, such illustration and description is to be considered illustrative or exemplary and not restrictive; the application is not limited to the disclosed embodiments.

[0164] Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed application, from a study of the drawings, the disclosure, and the claims. In the claims, the word "comprising" does not exclude other elements or steps, and the words "a" or "an" do not exclude a plurality. A single processor or other unit can fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measured cannot be used to advantage. A computer program can be stored and / or distributed on a suitable medium, such as an optical storage medium or a solid state medium supplied together with or as part of other hardware, but can also be distributed in other forms, such as via the internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

Claims

1. A medical instrument (100, 300), comprising: - a memory (110) storing machine executable instructions (120) and a predetermined logic module (130); and - a processor (106) configured to control the medical instrument, wherein execution of the machine executable instructions causes the processor to: - receive (206) a set of joint position coordinates (128) for a subject (118) placed on a subject support (320); wherein the set of joint position coordinates comprises coordinates for a left hip joint (510), a right hip joint (508), a right shoulder joint (512), and a left shoulder joint (514); - receive (207) a body orientation (132) in response to inputting the set of joint position coordinates into the predetermined logic module, wherein the body orientation is head first or feet first with respect to a selected coordinate system; - calculate (208) a torso aspect ratio (134) from the positions of the left hip joint, the right hip joint, the right shoulder joint, and the left shoulder joint; - wherein, if (210) the torso aspect ratio is greater than a predetermined threshold (136), assign (212) a body posture of the subject as a lateral decubitus posture; - wherein, if (210) the torso aspect ratio is less than or equal to the predetermined threshold, receive a determination of whether the subject is face up or face down on the subject support in response to inputting the set of joint position coordinates into the predetermined logic module, wherein face up and face down are orientations of a torso of the subject, wherein if the subject is face up, a back of the subject is resting on the subject support, wherein if the subject is face down, a chest of the subject is resting on the subject support; - if the subject is face up on the subject support, assign (220) the body posture as a supine posture; - if the subject is face down on the subject support, assign (222) the body posture as a prone posture; - generate (216) a subject posture label (142) comprising at least the body orientation and the body posture. the set of joint position coordinates comprises a left knee joint (506) and a right knee joint (504) and a neck joint (524); wherein the left knee joint coordinates comprise a left knee joint angle having a first limited range of motion; wherein the right knee joint comprises a right knee joint angle having a second limited range of motion; wherein the neck joint comprises a neck joint angle having a third limited range of motion; wherein, in response to inputting the first limited range of motion, the second limited range of motion, and the third limited range of motion into the predetermined logic module, the lateral decubitus posture is classified (212) as a left lateral decubitus posture or a right lateral decubitus posture.

2. The medical instrument of claim 1, wherein, execution of the machine executable instructions further causes the processor to calculate a lateral decubitus posture deviation from a difference between the predetermined threshold and the torso aspect ratio, and wherein the subject posture label further comprises a lateral decubitus posture deviation.

3. The medical instrument of claim 1 or 2, wherein, ​ 4. The medical instrument of claim 1, 2, or 3, wherein, The memory further includes a joint localizer module (124) configured to construct the set of joint position coordinates in response to receiving the image (122) of the object on the object support, wherein execution of the machine executable instructions further cause the processor to: - receive (200) the image of the object on the object support; - receive (206) the set of joint position coordinates in response to inputting the image into the joint localizer module.

5. The medical instrument of claim 4, wherein, The joint localizer module includes a neural network configured to output, in response to receiving the image of the object on the object support, a separate joint position probability map for each joint position coordinate in the set of joint position coordinates, wherein execution of the machine executable instructions further cause the processor to: - receive (202) the separate joint position probability maps (126) in response to inputting the image into the neural network; and - compute (204) the set of joint position coordinates from the separate joint position probability maps for each joint position in the set of joint positions.

6. The medical instrument of claim 5, wherein, The medical instrument includes a camera (322), wherein execution of the machine executable instructions further cause the camera to acquire the image of the object on the object support.

7. The medical instrument of claim 5 or 6, wherein, Execution of the machine executable instructions further cause the processor to train the neural network with a set of labeled images, wherein the set of labeled images includes images of: an object placed on an object support, an anterior view of an object, a posterior view of an object, an empty object support, an object support containing medical equipment, and a partially occluded image of an object.

8. The medical instrument of any one of the preceding claims, wherein, The medical instrument further includes a medical imaging system (302) configured for acquiring medical imaging data (332) from an imaging zone, wherein the object support is configured for at least partially supporting the object within the imaging zone.

9. The medical instrument of claim 8, wherein, The medical imaging system is any one of: a magnetic resonance imaging system (332), a positron emission tomography system, a single photon emission tomography system, a digital X-ray system, and a computed tomography system.

10. The medical instrument of claim 8 or 9, wherein, The memory further includes a medical imaging protocol, wherein the medical imaging protocol includes a selected posture label, wherein execution of the machine executable instructions further cause the processor to, in case the object posture label is different from the selected posture label, perform any one of: - provide a warning signal; and - select a different medical imaging protocol including the object posture label; - provide an instruction to reposition the object; and - a combination thereof.

11. The medical instrument of any one of claims 8, 9, or 10, wherein, Execution of the machine executable instructions further cause the processor to: - control the medical imaging system to acquire medical imaging data; - create an image file (336) including the medical imaging data and the object posture label; and - store the image file in a picture archiving and communication system.

12. The medical instrument of any one of the preceding claims, wherein, The medical instrument further comprises a radiotherapy system configured for irradiating a target within an irradiation zone, wherein the subject support is configured for at least partially supporting the subject within the irradiation zone, wherein the memory further comprises a radiotherapy protocol, wherein the radiotherapy protocol comprises a radiotherapy posture label, wherein execution of the machine executable instructions further causes the processor to perform any of the following operations in case the subject posture label is different from the radiotherapy posture label: - provide a warning signal; and - stop execution of the radiotherapy protocol; - provide instructions for repositioning the subject; and - a combination thereof.

13. The medical instrument of any one of the preceding claims, wherein, The set of joint position coordinates further comprises: a left elbow joint coordinate (518) and a right elbow joint coordinate (516), wherein the left elbow joint comprises a left elbow joint angle, wherein the right elbow joint comprises a right elbow joint angle, wherein the left shoulder joint coordinate comprises a left shoulder joint angle, wherein the right shoulder joint coordinate comprises a right shoulder joint angle, wherein execution of the machine executable instructions further causes the processor to: receive a hand position classification (140) in response to inputting the right shoulder joint angle, the left shoulder joint angle, the right shoulder joint angle, and the left shoulder joint into the predetermined logic module, wherein the subject posture label further comprises the hand position classification.

14. A method of medical imaging, wherein, The method comprises: - receiving a set of joint position coordinates (128) for a subject (118) placed on a subject support (320), wherein the set of joint position coordinates comprises coordinates for: a left hip joint (510), a right hip joint (508), a right shoulder joint (512), and a left shoulder joint (514); - receiving (206) a body orientation (132) in response to inputting the set of joint position coordinates into a predetermined logic module, wherein the body orientation is head first or feet first relative to a selected coordinate system; - calculating (208) a torso aspect ratio (134) from the positions of the left hip joint, the right hip joint, the right shoulder joint, and the left shoulder joint; - wherein if the torso aspect ratio is greater than a predetermined threshold (136), assigning (212) a body posture of the subject as a lateral decubitus posture; - wherein if (210) the torso aspect ratio is less than or equal to the predetermined threshold, receiving a determination of whether the subject is face up or face down on the subject support in response to inputting the set of joint position coordinates into the predetermined logic module; - if the subject is face up on the subject support, assigning (220) the body posture as a supine posture, wherein face up and face down are orientations of a torso of the subject, wherein face up and face down are orientations of a torso of the subject, wherein if the subject is face up, a back of the subject is resting on the subject support, wherein if the subject is face down, a chest of the subject is resting on the subject support; - if the subject is face up on the subject support, assigning (220) the body posture as a supine posture; - generating (216) a subject pose label (142) comprising at least the body orientation and the body posture.

15. A computer program product comprising machine executable instructions (120) for execution by a processor (106) controlling a medical instrument (100, 300), wherein, Execution of the machine executable instructions causes the processor to: - receive (206) a set of joint position coordinates (128) for a subject placed on a subject support; wherein the set of joint position coordinates comprises coordinates for: a left hip joint (510), a right hip joint (508), a right shoulder joint (512), and a left shoulder joint (514); - receive (207) a body orientation (132) in response to inputting the set of joint position coordinates into a predetermined logic module (130), wherein the body orientation is head first or feet first relative to a selected coordinate system; - calculate (208) a torso aspect ratio (134) from the positions of the left hip joint, the right hip joint, the right shoulder joint, and the left shoulder joint; - wherein if (210) the torso aspect ratio is greater than a predetermined threshold, assigning (212) a body posture of the subject as a lateral posture; - wherein if (210) the torso aspect ratio is less than or equal to the predetermined threshold, receiving a determination of whether the subject is face up or face down on the subject support in response to inputting the set of joint position coordinates into the predetermined logic module, wherein face up and face down are orientations of a torso of the subject, wherein face up and face down are orientations of a torso of the subject, wherein if the subject is face up, a back of the subject is resting on the subject support, wherein if the subject is face down, a chest of the subject is resting on the subject support; - if the subject is face up on the subject support, assigning (220) the body posture as a supine posture; - if the subject is face down on the subject support, assigning (222) the body posture as a prone posture; and - generating (216) a subject pose label (142) comprising at least the body orientation and the body posture.

Images