The present invention concerns a monitoring method for an examination room in which a magnetic resonance system having an examination region, in particular an examination tunnel, is situated. The present invention also concerns a monitoring system and a non-transitory, computer-readable data storage medium that implements such a method.
The present invention also concerns an evaluation computer that implements such a monitoring.
The present invention also concerns an examination room, having a magnetic resonance (MR) system sutured therein, the MR system having an examination region, in particular an examination tunnel and wherein a video camera system is associated with the examination room, with which images are acquired from an acquisition region, wherein the acquisition region encompasses a region upstream of the examination region, and wherein the video camera system is connected to an evaluation computer.
An examination room of the above general type is known from DE 10 2015 211 148 A1 and from the corresponding US 2016 0 367 169 A1. With this examination room, images are acquired from the region upstream of the examination region during the course of preparation of an examination. A person is localized using the acquired images. Information can be projected onto particular body parts of the person as a function of the acquired localization.
Magnetic resonance systems often have a superconducting basic field magnet, which generates a high static basic magnetic field of for example 1.5 T or 3 T. The strong magnetic field attracts ferromagnetic objects with a correspondingly strong force. The magnetic field can impair the function of a cardiac pacemaker or a different implant. It can also result in ferromagnetic objects, which are accidentally brought into the region of the force field, being attracted by the force field. As a result, the ferromagnetic object can cause damage if it strikes objects or a person.
The owner/operator of the magnetic resonance system is obliged to train operating personnel of the magnetic resonance system appropriately so as to avoid such incidents, as well as to also post appropriate warning notices at the accesses to the (closed) examination room in which the magnetic resonance system is located. Nevertheless, accidents continue to occur due to untrained personnel or patients or relatives of patients, who enter the examination room and bring magnetizable objects with them, for example an oxygen cylinder, a ventilator, a wheelchair or, in the case of cleaning personnel, cleaning equipment or a floor polisher. The appropriate warning signs are either not seen or not observed. In some cases, the danger is underestimated by persons entering the room, despite cognition of the warning signs. In rare cases, accidents of this kind occur even with personnel who have been trained appropriately, if the training occurred a long time ago.
An object of the present invention is to provide a monitoring method and system which dangerous situations of this kind and accidents can be avoided as much as possible.
According to the invention, a monitoring method of the general type mentioned in the introduction makes use of an evaluation processor that, independently of operation of the magnetic resonance system, receives images of an acquisition region from a video camera system, wherein the acquisition region encompasses accesses to the examination room and/or a region upstream of the examination region. Whenever, in a standby mode, the evaluation processor detects a person in the received images, it checks whether an alarm condition different from just the detection of a person is met. Whenever the alarm condition is met, the evaluation processor changes into an alarm state in which, at least once, it emits at least one acoustic signal that is audible in the examination room, and/or emit an optical signal that is visible in the examination room, and otherwise maintains the standby mode. In the alarm state, the evaluation processor checks whether a termination command has been specified to it via a man-machine interface. Upon receipt of the termination command, the evaluation processor changes into an off-state in which it no longer emits the acoustic signal and/or the optical signal. In the alarm state and in the off-state, the evaluation processor checks whether it still detects the person in the images. Whenever it still detects the person in the alarm state and in the off-state, the evaluation processor maintains the current state and otherwise passes into the standby mode.
The video camera system and the evaluation processor are therefore inventively used not just during the course of operation of the magnetic resonance system (in other words, during the course of examinations and in the preliminary stage of such examinations), but operate more or less continuously. The evaluation processor firstly checks whether it detects a person at all. Detection of a person as such (per se) still does not trigger an alarm, however. Instead, an alarm is triggered only if an alarm condition is also met. The alarm condition is met if, using additional criteria aside from the “mere” detection of a person, it is detected that a dangerous situation could exist. In this case an acoustic signal and/or an optical signal is emitted as an output, in other words an appropriate warning.
In some cases the warning will be a false alarm. This is not critical, however, since in this case the false alarm can be ended by specifying the termination command.
The alarm condition can be configured in various ways.
For example, the alarm condition may be met only if, in addition to the person, the evaluation processor detects in the received images an object on the detected person that is different from the person and his or her clothing. Detection of the object is a necessary condition, but not imperatively an adequate one for meeting the alarm condition. It is possible for the detection of the object to already be adequate for changing to the alarm state, but it is not obligatory. For example, it is possible that the alarm condition is only met if the detected object adequately matches at least one predetermined object type. The object types can be specified as required to minimize the aforementioned risks. For example, the object types can be a cylinder (for example an oxygen cylinder), a chair (in particular a wheelchair), a watch and/or glasses. Other object types, such as a ventilator, a bed or a patient bed, a mop, a broom or motor-driven cleaning equipment are also conceivable.
Alternatively or additionally, the alarm condition may be met only if the detected person is in a predetermined section of the acquisition region. For example, the region of the magnetic resonance system which should be regarded as a “danger zone” can be determined in advance. In this case, a safety zone is also defined around the danger zone and this safety zone is also defined as a corresponding section of the acquisition region. In this case, the alarm condition is therefore only met if the detected person moves into the safety zone *thereby allowing the alarm to be emitted before the person is in the danger zone).
Alternatively or additionally, the alarm condition is met only if the detected person is a person different from at least one predetermined person. In this case, triggering of the acoustic signal can be limited to cases in which a person is detected who is not “known” to the evaluation processor as being authorized.
The evaluation processor can perform a 2D evaluation of the images acquired by the video camera system, but the evaluation processor preferably performs a 3D evaluation of the images acquired by the video camera system. Such a 3D evaluation also provides an item of depth information. As a result it is often easier to evaluate the region in which a person is situated and/or whether and possibly which, further object is detected on the person.
The alarm condition can be statically specified to the evaluation processor. In a preferred embodiment, the evaluation processor is designed as a self-learning system. In this case it is possible for the evaluation processor, when the termination command has been specified to it in the alarm state, to modify the alarm condition such that a situation, for which the alarm condition was previously regarded as having been met, to be removed from the alarm condition definition, or at least a lower weighting is associated with that situation.
The present invention also encompasses a non-transitory, computer-readable data storage medium encoded with programming instructions that, when loaded into a computer of a monitoring system, and possibly distributively loaded into other components of the monitoring system, cause the computer to operate the monitoring system in order to implement any or all embodiments of the method according to the invention as, described above.
The object is also achieved by a monitoring system of the type mentioned in the introduction wherein the acquisition region, alternatively or additionally to the region upstream of the examination region, encompasses accesses to the examination room, and the monitoring system has an evaluation processor according to the invention, as described above.
According to
A magnetic resonance scanner 6 of a magnetic resonance system is situated in the examination room 1. The magnetic resonance scanner 6 has an examination region 7, for example an examination tunnel. The examination region 7 of the magnetic resonance scanner 6 is the region in which a temporally static, locally essentially homogeneous magnetic field (in practice usually called a B0 field) is generated by a basic field magnet 8 of the magnetic resonance scanner 6. The B0 field has a high magnetic field strength, for example 1.5 T or more.
A video camera system 9 is associated with the examination room 1. The video camera system 9 can be arranged inside the examination room 1. Images B are acquired from an acquisition region by means of the video camera system 9. The acquisition region comprises the accesses 5 to the examination room 1 and/or a region 10 upstream of the examination region 7, for example the region in which an examination table 11 is located before introduction of the examination table 11 into the examination region 7.
According to
First, a state Z of the evaluation processor 12 according to
In a step S3 the evaluation processor 12 performs an evaluation of the received group of images B. It is possible that the evaluation processor 12 performs a 2D evaluation of the group of images B. The evaluation processor 12 preferably performs a 3D evaluation of the group of images, however. For example, an item of depth information can be determined from the respective image B for at least one of the acquired images B on the basis of a corresponding projection of a known pattern in the acquisition region. Alternatively, it is possible, by correlation of a number of images B, to determine a corresponding three-dimensional item of information. The relevant modes of procedure are generally known to persons skilled in the art.
In a step S4 the evaluation processor 12 checks whether it detects a person 15 during the course of the evaluation. The detection of step S4 should not be understood as meaning identification of the actual person 15. It is therefore not a matter of whether a particular person 15 is detected (“that is Mr. Müller”), but whether a person 15 is detected at all (“there is someone here”). If the evaluation processor 12 detects a person 15, the evaluation processor 12 moves to step S5. Otherwise it returns to step S2.
In step S5 the evaluation processor 12 checks whether an alarm condition is met. The alarm condition is a condition different from detection of a person 15 per se. Detection of a person 15 per se, therefore, does not always or necessarily trigger an alarm, but an alarm is triggered only if in addition the alarm condition is met. Possible embodiments of the alarm condition will be illustrated in more detail below in connection with the further figures. If the alarm condition is met, the evaluation processor 12 moves to a step S6. Otherwise, it returns to step S2.
In step S6, the state Z of the evaluation processor 12 is set to an alarm state Z2. The evaluation processor 12 therefore moves into the alarm state Z2. In a subsequent step S7 the evaluation processor 12 checks whether it is in the alarm state Z2. If this is the case, the evaluation processor 12 carries out step S8. Otherwise, step S8 is skipped. In step S8 the evaluation processor 12 emits an alarm signal A. The alarm signal A usually is an acoustic signal. Examples of suitable acoustic signals are a conventional alarm sound (sound of a horn, siren sound and the like) or the emitting of an appropriate spoken message, such as “Caution! There is a very strong magnetic field here. Go back immediately”. The acoustic signals can be emitted via a loudspeaker 16 (see
Alternatively or additionally, the evaluation processor 12 emits an optical signal in step S8. For example, the evaluation processor 12 can switch on a yellow or red flashing light or, via a display device 17, (see
Step S9 can follow step S8 in which the evaluation processor 12 takes further measures, for example transmits appropriate messages to devices arranged remotely, so that emergency measures can be initiated there.
In step S10 the evaluation processor 12 checks whether a termination command T has been specified to it via a human-machine interface 18. If the termination command T is specified, the evaluation processor 12 moves to a step S11. In step S11 the state Z of the evaluation processor 12 is set into an off-state Z3. Otherwise, the evaluation processor 12 skips step S11.
The evaluation processor 12 then receives a group of images B from the video camera system 9 in a step S12 and evaluates the received group of images B in a step S13. Steps S12 and S13 correspond in terms of content to steps S2 and S3.
In a step S14 the evaluation processor 12 checks whether it does not detect a person during the course of the evaluation, in other words, no longer detects the person 15. If the evaluation processor 12 continues to detect the person 15, the evaluation processor 12 returns to step S7. Otherwise, it returns to step S1.
Due to the checking in step S7, the evaluation processor 12 therefore no longer emits the acoustic signal and/or the optical signal if it is in the off-state Z3.
It is possible that step S7 as such is not present. In this case the evaluation processor 12 passes from step S14 either to step S9 (if present) or to step S10. In this case the alarm signal A is only emitted once during the transition from standby mode Z1 into the alarm state Z2.
The approach of
The alarm condition can, as already mentioned, be configured in different ways. Some of the possible embodiments will be illustrated in more detail below in connection with
According to
Furthermore, according to the flowchart in
It is also possible to perform the check of
Furthermore, according to the flowchart in
The identification of the person can be configured as required. In particular, appropriate methods of facial recognition are generally known to persons skilled in the art. Alternatively or additionally, the stature of the person 15 can also be evaluated.
In the case of simultaneous detection of a number of individuals 15, the embodiment of
As a rule, the alarm condition is static from the perspective of the evaluation processor 12. It is therefore specified to the processor 12 from outside and is not modified by the evaluation processor 12. It is, however, possible for the evaluation processor 12 to be self-learning insofar as it can independently learn “non-critical” situations. This will be illustrated in more detail below in connection with
The present invention has many advantages. In particular, it is possible to almost completely prevent accidents, which can be attributed to accidental disregard of the safety regulations. The operational safety can be significantly increased.
Although modifications and changes may be suggested by those skilled in the art, it is the intention of the Applicant to embody within the patent warranted hereon all changes and modifications as reasonably and properly come within the scope of the Applicant's contribution to the art.
Number | Date | Country | Kind |
---|---|---|---|
17179594.1 | Jul 2017 | EP | regional |