Method and system for providing error information regarding a plurality of individual measurements

Abstract

The invention relates to a computer-implemented method for providing error information about a plurality of individual measurements. The method comprises the method step of providing (PROV-1) a plurality of individual measurements, each of which is assigned to an examination. The method further comprises the method step of receiving (REC) examination information about a population to be examined for the examination from a user via an interface. For each examination of the population, the method further comprises the method step of extracting (EXT) examination information based on the individual measurements belonging to the examination, determining (DET-1) based on the examination information whether one or more erroneous measurements were made in the examination, and determining (DET-2) examination error information based on the determination result. The method further comprises the method step of compiling (DET-3) error information based on the examination error information of the population to be examined. The method further comprises the method step of providing (PROV-2) the error information to the user via an interface.

Description

Technical Field

[0001] Medical examinations of patients using medical technology equipment are standard procedures in medicine. These equipment is typically an imaging system. Medical examinations of patients using imaging systems, such as radiography (X-ray imaging), computed tomography (CT), magnetic resonance tomography (MRT), positron emission tomography (PET), or single-photon emission computed tomography (SPECT), are standard procedures in medical diagnosis. At least one medical image of the patient's area to be examined is recorded or created. Laboratory diagnostic equipment is also often used as medical technology equipment for performing medical examinations. Background Art

[0002] To perform such medical examinations using medical technology equipment, an examination protocol is typically implemented. An examination protocol includes one or more individual measurements. In other words, the examination protocol provides for or defines one or more individual measurements. The individual measurements of an examination protocol may differ, for example, in the measurement settings associated with the individual measurements, such as the recorded body region and / or contrast agent enhancement and / or the measurement time and / or the reagent used. The measurement time may be coordinated, for example, with respect to the time of contrast agent administration, the electrocardiogram (EKG) signal, and / or the patient's movement state. Reagents may be used, in particular, in the laboratory to examine patient samples. Samples may be, in particular, blood, saliva, urine, or biopsy samples. The individual measurement settings in the examination protocol can be manually adjusted by an operator. The operator may be, in particular, a medical technical assistant (MTA), a medical technical radiology assistant (MTRA), a medical professional (MFA), a chemical technical assistant (CTA), or a doctor. The operator may adapt the individual measurement settings to the patient or sample being examined. For example, the operator may set the scanning area. The scanning region can be set by the operator according to the patient's body dimensions and describes the region that should be recorded for the medical examination of the patient by means of the imaging system. In addition, the operator can, for example, manually determine the measurement time.

[0003] If one of the individual measurements is performed incorrectly, the operator of the medical device can repeat the corresponding individual measurement. An incorrectly performed individual measurement can also be referred to as an erroneous measurement. Causes of an erroneous measurement can include, for example, a scan area that is too small and / or the recording of an incorrect body region and / or an incorrect measurement time and / or an exposure time that is too short (e.g., to observe a contrast agent bolus) and / or patient movement and / or an incorrect agent, etc. Therefore, when repeating an individual measurement that was performed incorrectly, at least one incorrectly selected measurement setting can be adjusted. Repeating an individual measurement can, in particular, lead to deviations from the examination protocol, since additional measurements must be performed.

[0004] Repeating individual measurements has a particularly negative impact on patients. Therefore, in medical examinations based on X-ray radiation, repeating individual measurements exposes the patient to additional X-ray exposure. Furthermore, repeating individual measurements results in an increased examination duration or repeated sample extraction from the patient. This can be uncomfortable for the patient and reduces the cost-effectiveness of medical equipment. Furthermore, repeating individual measurements causes increased wear and tear on medical equipment and / or increased consumption of laboratory materials.

[0005] Frequently, erroneous measurements are made due to operator error. For example, such errors can be incorrect patient support or positioning, an incorrectly selected scanning area, and / or other incorrectly selected measurement settings. Only a small fraction of erroneous measurements performed in daily clinical practice are caused by unavoidable events, such as patient movement.

[0006] Typically, after completing a medical examination, individual measurements are uploaded to a database, such as a Picture Archiving and Communication System (PACS). Operators often do not or only partially upload erroneous measurements to the database. Consequently, traceability of erroneous measurements is often difficult. Furthermore, if erroneous measurements are stored in the database, they are often not directly marked as such, making traceability difficult.

[0007] However, deviations from the examination protocol may also be expected. For example, the operator can use the examination protocol as a template and adapt it to given patient-specific and / or examination-specific conditions. Such expected deviations should not be counted as erroneous measurements. Summary of the Invention

[0008] It is therefore an object of the present invention to provide a method which provides error information about a plurality of individual measurements.

[0009] The object is achieved by a method for providing error information about a plurality of individual measurements, a method for providing a trained function, a system for providing error information about a plurality of individual measurements, a computer program product, and a computer-readable storage medium. Advantageous developments are listed in the following description.

[0010] Hereinafter, the solution according to the invention to the stated object will be described not only with respect to the claimed apparatus but also with respect to the claimed method. Features, advantages, or alternative embodiments mentioned herein can also be transferred to other claimed subject matter, and vice versa. In other words, substantial embodiments (e.g., those relating to the apparatus) can also be improved by means of features described or claimed in conjunction with the method. Corresponding functional features of the method are thus formed by corresponding substantial modules.

[0011] Furthermore, the solution according to the invention to the stated object is described not only with respect to a method and apparatus for providing error information regarding a plurality of individual measurements, but also with respect to a method and apparatus for providing a trained function. Features and alternative embodiments of the data structures and / or functions in the determination method and apparatus can be transferred to similar data structures and / or functions in the adjustment / optimization / training method and apparatus. Similar data structures can be characterized, in particular, by the use of the prefix "train." Furthermore, the trained function used in the method and apparatus for providing error information regarding a plurality of individual measurements can be trained, adjusted, and / or provided, in particular, by the method for providing a trained function.

[0012] The present invention relates to a computer-implemented method for providing error information regarding a plurality of individual measurements. The method comprises the method step of providing a plurality of individual measurements, each of which is associated with an examination. The method further comprises the method step of receiving examination information regarding a population to be examined from a user via an interface. For each examination of the population, the following steps are performed: extracting examination information based on the individual measurements belonging to the examination; determining, based on the examination information, whether one or more erroneous measurements were made during the examination; and determining examination error information based on the results of the determination. The method further comprises the method step of compiling error information based on the examination error information for the population to be examined. The method further comprises the method step of providing the error information to the user via an interface.

[0013] The medical examination may be performed, in particular, on or with the aid of a medical device. The medical device may include, in particular, an imaging system or a medical imaging system. During the medical examination, at least one medical image of a patient, a region of the patient, or a body region of the patient may be created, detected, or measured with the aid of the imaging system. The imaging system may include, in particular, an X-ray device, a flat-panel X-ray device, a computed tomography (CT) device, a magnetic resonance tomography (MRT) device, a positron emission tomography (PET) device, a single-photon emission computed tomography (SPECT) device, a mammography device, or the like, and / or combinations thereof. Alternatively, the medical device may be a laboratory diagnostic device. In particular, a patient sample may be analyzed with the aid of the laboratory diagnostic device during the medical examination. The patient sample may, in particular, be a blood sample, a saliva sample, a urine sample, a biopsy sample, or the like. The sample may be obtained from the patient, in particular, in preparation for the medical examination.

[0014] Performing a medical examination particularly includes creating, detecting, recording, or measuring one or more individual measurements. Here, an individual measurement describes a single measurement process performed using a medical device. In particular, individual measurements can be performed without interruption. Measurement settings can be constant or unchanging during the individual measurement. Individual measurements can particularly include measurement data and / or metadata related to the measurement data. Metadata is described further below. Measurement data can particularly include image data and / or data generated or determined during the measurement process. Individual measurements can differ particularly in their measurement settings. For example, measurement settings can include or describe, determine, or define the patient's body region to be examined, and / or information about the scan region and / or the time of measurement and / or the recording contrast and / or the sample type and / or the reagent and / or the exposure time and / or the exposure duration and / or the energy of the X-ray radiation in an X-ray radiation-based imaging system. Measurement settings can particularly be adapted to the patient and / or the medical examination. In particular, a medical examination can be performed on the examined body region to clarify a patient's condition or illness. When performing a tomography scan as an individual measurement, the scan region can particularly be the patient region recorded, detected, or displayed during the tomography scan. The measurement time can be, in particular, the time at which an individual measurement is recorded or performed according to an event. An event can be, for example, the administration of a contrast agent and / or an electrocardiogram (EKG) signal and / or the patient's movement state, etc. In particular, by recording a plurality of individual measurements at different measurement times after the administration of the contrast agent, it is possible to observe, for example, how a contrast agent bolus spreads. The recording contrast can be, in particular, a weight in an MRT recording. The sample type can, in particular, describe the type of sample, for example, a blood sample, a saliva sample, a urine sample, a biopsy sample, etc. The reagent can, in particular, be a substance used to analyze the sample.

[0015] For example, during a single measurement, the operator can manually set or adjust the scan area and / or body area and / or other measurement settings. The operator can support or position the patient in, on, or in front of the imaging system. The operator can be, in particular, a Medical Technical Assistant (MTA), / or a Medical Technical Radiology Assistant (MTRA), / or a Chemical Technical Assistant (CTA), / or a female or male medical expert (MFA), / or a male or female physician, etc.

[0016] Information about individual measurements can, in particular, be included or stored in metadata about individual measurements and / or associated examinations. In other words, metadata can include information about individual measurements and / or associated examinations. Metadata can, for example, include measurement settings and / or a sequential number for each individual measurement and / or information about the medical equipment used and / or information about the operator and / or information about the patient. Information about the medical equipment used can, in particular, include the manufacturer and / or year of manufacture and / or maintenance status of the medical equipment. Information about the operator can, in particular, include the operator's name and / or the operator's coded name. Information about the patient can, for example, include the patient's name, weight, height, illness, reason for the examination, etc. Metadata can, in particular, include a DICOM header and / or a dose structured report. In particular, metadata can be queried, for example, using DICOM queries and retrieval.

[0017] In particular, each individual measurement may be an erroneous measurement. In other words, an individual measurement may be an erroneous measurement. Here, an erroneous measurement is an individual measurement that is performed erroneously. An erroneous measurement may in particular be performed or implemented with the aid of at least one erroneous measurement setting. In the case of an erroneous measurement, for example, an erroneous scanning area and / or an erroneous body area and / or an erroneous measurement time and / or an excessively short exposure time and / or an erroneous reagent may be selected or set. An erroneous measurement may also occur due to patient movement during an individual measurement. An erroneous measurement is indicated in particular by repeating the corresponding individual measurement, i.e., by performing the corresponding individual measurement twice. Thus, the corresponding individual measurement is performed once as an erroneous measurement and then again as a correct or successful individual measurement with the aid of the corrected measurement setting (and then again as a correct or successful individual measurement with the aid of the corrected measurement setting).

[0018] Individual measurements from a medical device can, in particular, be transmitted or loaded into a database. Individual measurements can, in particular, be archived in the database. Individual measurements transmitted to the database can, in particular, be referred to as archived individual measurements. The database can, in particular, be a central database. Individual measurements from multiple medical devices can, in particular, be stored in the central database. Individual measurements from multiple medical examinations can, in particular, be stored in the central database. The central database can, in particular, be stored or saved on a server or in a cloud system. The server can, in particular, be installed in the facility where the medical device is located, such as a hospital. Alternatively, the server can be an external server. The central database can, in particular, include a Picture Archiving and Communication System (PACS). Individual measurements can, in particular, be transmitted manually or automatically to the database by an operator. The operator can, in particular, select which individual measurements from the medical examination should be transmitted to the database. The operator can, in particular, transmit correctly performed or successful individual measurements. In particular, the operator can transmit only some or no erroneous measurements to the database.

[0019] In the method step of providing a plurality of individual measurements, in particular, individual measurements stored or saved in a database may be provided. In other words, individual measurements transferred to the database may be provided. The plurality of individual measurements may include not only correctly performed or successful individual measurements, but also erroneous measurements. Each of the individual measurements may be associated with a medical examination. In other words, each examination in the database may include one or more individual measurements. One or more examinations may be stored or saved in the database. Multiple medical examinations may be stored or saved in the database. The expression "a database, for example, includes a plurality of examinations" hereinafter may particularly mean that, for example, multiple examinations are stored or saved in the database.

[0020] In the method step of receiving test information, test information regarding a population to be tested, provided or input by a user, is received. The population may particularly include one or more medical tests. In particular, in the method step of providing multiple individual measurements, individual measurements associated with the population's tests are provided. The user may particularly be a person who wishes to analyze the occurrence of erroneous measurements. The interface may particularly be a user interface. The interface may, for example, be a keyboard and / or a graphical user interface (GUI) and / or a touch-sensitive screen. With the help of the test information, the user may define or determine the population to be tested. In other words, with the help of the test information, the tests to be associated with the population to be tested may be determined from a plurality of tests in a database. The population to be tested may particularly not include tests from the plurality of tests in the database, or may include all tests from the plurality of tests in the database. The operator may particularly manually select tests to be associated with the population or to be added to the population. Alternatively or additionally, the test information may, for example, specify that all tests performed using medical equipment from a particular manufacturer should be associated with the population. Alternatively or additionally, the test information can specify, for example, that all tests related to a disease and / or all tests performed by a specific operator are to be associated with a population. In particular, the test information can be used to filter multiple tests based on arbitrary information contained in individual measurements or in metadata included with the individual measurements. Based on these filters, tests can be associated with a population or not. In particular, by associating a test with a population, all individual measurements included in the database are associated with tests for the population. In alternative embodiments, the test information can be predefined or automatically determined.

[0021] In the method step of extracting examination information, examination information is extracted for each examination in the population based on the individual measurements belonging to the examination. In particular, the examination information can be extracted from metadata included by the individual measurements. The examination information may include a plurality of individual measurements in a database, which may be associated with the examination. Alternatively or additionally, the examination information may include a number for each individual measurement associated with the examination and included in the database. Alternatively or additionally, the examination information may include, for example, an average pixel value in the individual measurements and / or a result or diagnosis. The examination information may include, in particular, any information provided by the metadata about the examination or the individual measurements associated with the examination and included in the database.

[0022] In a specific method step, for each examination in the population, it is determined based on the examination information whether one or more erroneous measurements were made during the examination. In other words, it is determined based on the examination information whether one or more individual measurements of the individual measurements associated with the examination and included in the database are erroneous measurements. In particular, it is possible that no individual measurement is an erroneous measurement. In particular, half of the individual measurements associated with the examination are likely to be erroneous measurements. In this case, each individual measurement is performed once as an erroneous measurement and once as a corrected, successful individual measurement.

[0023] In the method step of determining examination error information, examination error information is determined based on each examination determined as a group. The examination error information may include, in particular, information about whether at least one erroneous measurement was performed during the examination. The examination error information may include, in particular, information about whether more than one erroneous measurement was performed during the examination. The examination error information may include, in particular, information about the number of erroneous measurements in the medical examination. The examination error information may include, in particular, information about which individual measurement of the examination was performed as an erroneous measurement. The examination error information may include, in particular, information about which measurement settings were corrected, changed, or adjusted in the erroneous measurement compared to the associated correctly performed or successful individual measurement. In other words, the examination error information may include information about why the erroneous measurement was erroneous.

[0024] In the method step of compiling error information, the error information can be compiled based on the examination error information of the examination of the population. In other words, the error information can be compiled for the entire population to be tested. The error information is particularly based on the examination error information of the examinations included in the population to be tested. In particular, the examination error information is determined for each examination of the population to be tested. The error information can be based in particular on a statistical evaluation and / or comparison of the examination error information. In particular, the examination information of the examination can be evaluated for compiling the error information. In other words, the error information can also be based on the examination information. For example, the error information can particularly indicate how frequently erroneous measurements occur in the examination of the population. The error information can particularly include information about which individual measurements are particularly frequently performed as erroneous measurements and / or what the causes of the erroneous measurements are. The error information can particularly include information about which operators and / or medical equipment are particularly frequently performing erroneous measurements.

[0025] In the method step of providing error information, the error information is provided to the user via an interface. In particular, the error information can be provided to the user via a display unit. The provision can be implemented in particular via a GUI. In other words, the interface can be configured as a display unit and / or a GUI. Alternatively, the error information can be stored in a memory so that the user can access the error information at any time. The memory can in particular be internal memory, such as a hard disk and / or temporary memory, such as a work memory. Alternatively, the memory can be external memory, such as a database, a cloud system, a USB memory device, an SD card, a CD, a DVD, an external hard disk, etc.

[0026] The inventors have recognized that it is possible to analyze erroneous measurements performed in a population with the aid of the described method. The inventors have also recognized that it is possible in the described manner to ascertain or identify the cause of the erroneous measurement. The inventors have recognized that knowledge of the causes can enable the execution of an optimized examination with respect to, for example, cost-effectiveness and / or patient X-ray exposure and / or patient comfort and / or examination duration. The inventors have recognized that, due to the large amount of data, a manual analysis of the examination of a population is not possible. The inventors have recognized that, with the aid of the described method steps, it is also possible to analyze large amounts of data in a population in a relatively short time and to provide error information to the user. It is possible by the method to analyze large and complex amounts of data for the purpose of compiling error information. The inventors have recognized that this can be achieved, in particular, by extracting examination information.

[0027] According to one aspect of the invention, the error information comprises a statistical evaluation of examination error information of a group, in particular a statistical evaluation of the frequency of erroneous measurements at medical-technical equipment and / or the frequency of erroneous measurements caused by an operator and / or the frequency of erroneous measurements in examinations of specific diseases.

[0028] The error information may include, in particular, information about the frequency of erroneous measurements related to the manufacturer of the medical device. Alternatively or additionally, the error information may include information about the frequency of erroneous measurements related to the type of medical device. The type may include, in particular, a flat-panel X-ray device, a CT device, an MRT device, a PET device, a SPECT device, specific laboratory diagnostic devices, etc. The error information may include, in particular, information about the frequency of erroneous measurements related to the maintenance status of the medical device.

[0029] Alternatively or additionally, the error information may include information about the frequency of operator-dependent erroneous measurements, etc. In other words, the error information may include information about which operator particularly often performs erroneous measurements.

[0030] Alternatively or additionally, the error information can include information about the frequency of erroneous measurements related to the specific disease of the examined patient and / or multiple examined patients. In this context, the frequency of erroneous measurements related to the medical examinations typically performed in the disease can also be included in particular by the error information.

[0031] In particular, in addition to the examination error information, the examination information of the examination of the population can also be evaluated for statistical evaluation.

[0032] The inventors have therefore recognized that it is possible to assess which medical devices or operators are most frequently performing erroneous measurements. In particular, the inventors have recognized that, based on error information, specific operators can be retrained in a targeted manner, or that medical devices from specific manufacturers can be preferred when repurchasing. Furthermore, the inventors have recognized that, based on the knowledge gained from error information, it is possible to optimize examinations in a targeted manner, for example, with respect to specific diseases.

[0033] According to another aspect of the present invention, the check includes a first set of individual measurements and a second set of individual measurements. The first set includes successful individual measurements. The second set includes erroneous measurements. The check information includes at least one piece of information about the individual measurements in the first set and at least one piece of information about a subset of the individual measurements in the second set.

[0034] In particular, the operator and / or the medical device transfers correctly performed or successful individual measurements of the examination to a database. Successful individual measurements of the examination are, in particular, part of a first set of individual measurements. Individual measurements or erroneous measurements performed during the examination are, in particular, not transferred at all, or only partially, partially, or completely, manually or automatically by the operator to the database. In particular, erroneous measurements of the examination are part of a second set of individual measurements. A subset of the second set includes, in particular, only the erroneous measurements of the examination that have been transferred to the database. In particular, the subset may not include any erroneous measurements of the second set, or may include one, several, or all erroneous measurements of the second set. If the subset includes all erroneous measurements of the second set, then the subset includes, in particular, all erroneous measurements performed during the examination. Therefore, the examination information of the examination includes at least one information item regarding all successful individual measurements. The examination information of the medical examination also includes at least one information item regarding erroneous measurements of the examination that are part of a subset of the second set. The information included in the examination information can be constructed as described above.

[0035] The archived individual measurements include, in particular, all successful individual measurements of the first set of individual measurements and erroneous measurements of a subset of the second set of individual measurements. The individual measurements of the first and second sets of individual measurements can collectively be referred to as performed individual measurements. Thus, performed individual measurements include, in particular, all individual measurements performed during the examination.

[0036] The inventors have recognized that the inspection information includes at least one item of information about a single measurement of the first set and a single measurement of a subset of the second set. The inventors have recognized that it must be taken into account that the inspection information may not include at least one item of information for each single measurement inspected, but only for the single measurements transmitted to the database. The inventors have recognized that the non-uniform transmission of the single measurements to the database makes manual compilation of error information unfeasible.

[0037] According to another aspect of the present invention, a first set of individual measurements is predefined in an inspection protocol. The method further comprises the step of providing an inspection protocol. Based on the inspection protocol, a step of determining whether one or more erroneous measurements were made during the inspection is performed.

[0038] The examination protocol defines the individual measurements that should be successfully performed in a medical examination. In particular, the examination protocol can define, determine, or predefine the order of the individual measurements. The examination protocol can in particular include measurement settings for performing, implementing, or detecting the individual measurements. Therefore, erroneous measurements are not included in the examination protocol. In particular, the examination protocol can be predefined. In particular, the examination protocol can be adjusted by the operator. In particular, the operator can create the examination protocol. In particular, the operator can adjust the measurement settings via the examination protocol. If an erroneous measurement is performed, the operator can correct the measurement settings via the examination protocol. In particular, the number of individual measurements included in the examination protocol does not change due to the correction. When correcting, in particular, only the measurement settings of the individual measurements to be corrected are adjusted. The examination protocol can in particular be specially designed for examining a specific disease or condition of a patient. The examination protocol can in particular be designed specifically for a specific medical device.

[0039] In the provided method step, an examination protocol is provided. In particular, an examination protocol is provided for each examination of a population to be examined. In particular, the examination protocol can be provided via a database.

[0040] The method step of determining whether one or more erroneous measurements were made during the examination can then be carried out based on the examination protocol.

[0041] The inventors have recognized that an inspection protocol includes information about how the inspection should ideally be performed. In other words, the inventors have recognized that the inspection protocol describes how the inspection should be performed without erroneous measurements. In particular, the inventors have recognized that the inspection protocol includes information about successful individual measurements.

[0042] According to another aspect of the invention, the examination information comprises at least one of the following information about the individual measurements of the first set and the individual measurements of the subset of the second set: the name of the individual measurement, the type of the individual measurement, the body region measured in the individual measurement, the dose-length product of the individual measurement, the naming of the examination protocol, the reason for the examination.

[0043] The examination information specifically includes at least one of the above-listed information for all individual measurements transferred to the database for the examination. In other words, the examination information includes at least one of the above-listed information for the individual measurements of the first set and the individual measurements of the subset of the second set. The name of the individual measurement can specifically be the name of the individual measurement. The name can specifically be descriptive. For example, the name can include the following: "Skull CT." The name can specifically be specific to an individual measurement in the first set. The type of the individual measurement can specifically describe how the individual measurement was performed. The type can include, for example, "Helical acquisition," "Constant angle acquisition," or "T1-weighted recording." The measured body region specifies which body region was examined or recorded using the individual measurement. The dose-length product specifies the X-ray dose applied to the patient. The dose-length product describes the X-ray exposure to the patient as a result of X-ray imaging using the CT system. The X-ray voltage of the individual measurement specifies the voltage applied to the X-ray source that emits the X-ray radiation. Therefore, the X-ray voltage specifically includes information about the X-ray spectrum used to record the medical image. The naming of the examination protocol can be particularly one-to-one. The naming of the examination protocol can be particularly descriptive. In other words, the naming of the examination protocol can be descriptive. The naming of the examination protocol can particularly describe the disease for which the examination protocol is implemented. Alternatively or additionally, the naming of the examination protocol can include terms that describe the entire examination, such as "chest CT". The reason for the examination can particularly indicate why the examination is performed. The reason for the examination can particularly indicate which disease the examination is performed for the purpose of diagnosing or examining. In other words, the reason for the examination can indicate the context in which the examination is performed. The disease can particularly be the reason for the examination.

[0044] The inventors have realised that it may be possible to infer or determine from the information listed above that may be included in the inspection information of the inspection whether one or more erroneous measurements were made.

[0045] According to another aspect of the present invention, the method step of determining whether one or more erroneous measurements have been made includes the method step of determining a difference between a number of protocols for a single measurement and a number of checks for the single measurement. The number of protocols for a single measurement includes the number of single measurements in a first set of single measurements. The number of checks for a single measurement includes the number of single measurements in a single measurement in the first set and in a subset of single measurements in a second set. Checking for error information is based on the difference between the number of protocols and the number of checks.

[0046] The number of agreements for individual measurements includes, in particular, the number of individual measurements that are ideally performed during an examination. In other words, the number of agreements includes the minimum number of individual measurements that are performed during an examination if no erroneous measurements are performed. The number of agreements for individual measurements includes, in particular, the number of successful individual measurements in the first set of individual measurements. The number of agreements can be determined, in particular, based on the examination protocol. In other words, the number of agreements for individual measurements includes the number of individual measurements defined in the examination protocol.

[0047] The number of checks specifically includes the number of successful individual measurements of the first set plus the number of erroneous measurements of the subset of the second set. If the subset is equal to the second set, the number of checks includes the number of individual measurements actually performed during the check. The number of checks of individual measurements then specifically includes the number of successful individual measurements plus the number of erroneous measurements. If the subset does not include all erroneous measurements of the individual measurements of the second set, the number of checks of individual measurements includes the number of successful individual measurements plus the erroneous measurements that have been transferred to the database. In other words, the number of checks includes the number of archived individual measurements.

[0048] In the method step of determining the difference, the number of measured agreements is subtracted from the number of measured checks. Alternatively, the number of measured checks can be subtracted from the number of measured agreements. In particular, the number of erroneous measurements in a subset of the individual measurements of the second set can be determined in this manner. In particular, if the subset includes all individual measurements of the second set, the difference can describe the total number of erroneous measurements in the check.

[0049] In particular, the examination error information can be based on a difference value. The examination error information can include information about whether the difference value is equal to or not equal to zero. In particular, the examination error information can include information about the difference value. The examination information can include information about the number of erroneous measurements provided in connection with the medical examination. The examination error information can include information about the number of erroneous measurements transferred to the database during the examination. Thus, if all erroneous measurements have been transferred to the database, if all erroneous measurements have been provided, or if the subset is equal to the second set, the examination error information can include information about the total number of erroneous measurements performed during the examination.

[0050] The inventors have recognized that, in this manner, the total number of erroneous measurements can be determined during an examination, particularly when all individual measurements are transferred to a database. Furthermore, the inventors have recognized that it is then also possible to determine whether at least one erroneous measurement was actually performed. The inventors have recognized that this information can be included in the examination error information. Furthermore, the inventors have recognized that this information can be determined for each examination of the population to be examined.

[0051] According to another aspect of the present invention, each individual measurement in the first set and the second set is assigned a consecutive number. The check information includes the numbers of the individual measurements in the first set and the numbers of the individual measurements in the subset of the second set. The method step of determining whether one or more erroneous measurements have been made includes the method step of verifying whether the numbers of the individual measurements in the check information are consecutive. The check error information includes the result of this verification.

[0052] During an examination, each individual measurement can in particular be assigned a consecutive number. Thus, each individual measurement of the first set and each individual measurement of the second set can in particular be assigned a one-to-one number. The numbering for each examination can in particular be one-to-one. Additionally, the numbering for the medical device can be one-to-one. The numbering can in particular describe the order of the individual measurements. The examination information can in particular include the consecutive numbers of the individual measurements provided. The examination information can in particular include the consecutive numbers of the individual measurements transmitted to the database. In other words, the examination information can in particular include the consecutive numbers of the individual examinations of the first set and the consecutive numbers of the individual examinations of the subset of the second set.

[0053] In the verification method step, it is checked whether the numbers in the examination information are continuous. If the numbers in the examination information are not continuous, it can be inferred that there is at least one erroneous measurement that has not been provided. In particular, it can then be inferred that there is at least one erroneous measurement that has not been transmitted to the database. In particular, it can be inferred that there is at least one erroneous measurement in the second set that is not part of the subset. In particular, the number of "gaps" in the consecutive numbers of the continuous sequence can be determined during the verification, so that the number of erroneous measurements that have not been provided or transmitted can be inferred. In particular, if no erroneous measurements are provided or transmitted to the database in the verification method step, it can be inferred that there is a total number of erroneous measurements in the medical examination. In other words, if the subset does not include an erroneous measurement from the second set, the total number of erroneous measurements can be inferred.

[0054] The result of the method step of checking can include in particular information on whether at least one erroneous measurement is present. Alternatively or additionally, the result can include information on the number of erroneous measurements that were not provided or not transferred to the database. The check error information can in particular include the result of the check.

[0055] The inventors have recognized that if no erroneous measurements are provided or not transmitted to the database, it is possible to identify or verify whether an erroneous measurement has been performed based on the consecutive numbering of the individual measurements. The inventors have also recognized that the method is particularly interesting if no erroneous measurements are transmitted to the database, or only a small number of erroneous measurements are transmitted to the database. In particular, the inventors have recognized that, using the combination of the above-described method steps for determining the check error information, the total number of checked erroneous measurements can be determined, regardless of whether the erroneous measurements have been transmitted to the database. In other words, the combination of the method steps of determining the difference between the number of protocols for an individual measurement and the number of checks for the individual measurements and verifying whether the numbering of the individual measurements in the check information is consecutive makes it possible to determine the total number of erroneous measurements, regardless of whether all, none, or only a portion of the erroneous measurements have been transmitted to the database.

[0056] According to another aspect of the present invention, an erroneous measurement in the individual measurements of the second set corresponds to an individual measurement in the individual measurements of the first set. Here, checking the error information includes information indicating which individual measurement in the individual measurements of the first set has a corresponding erroneous measurement in the subset of the second set.

[0057] In particular, as described above, each erroneous measurement is an erroneous single measurement or an erroneously performed single measurement. In particular, a single measurement that should be successfully performed in order to perform the inspection can be specified in the inspection protocol. Since this single measurement should be successfully or correctly performed for a complete inspection, this erroneous single measurement is performed again or repeated after correcting or adjusting the measurement parameters or settings. This single measurement is then, in particular, successfully performed. This successful single measurement is in particular part of the first set of single measurements. This successful single measurement then corresponds in particular to the erroneous measurement. In other words, this erroneous measurement corresponds to this successful single measurement. In particular, most of the measurement settings may be identical in the erroneous measurement and the corresponding successful single measurement. In particular, only the manually settable measurement settings of the erroneous measurement and the corresponding successful single measurement may differ from each other. In particular, the reason or purpose for performing the single measurement remains unchanged or the same in the erroneous measurement and the corresponding successful single measurement. The names of the erroneous measurement and the corresponding successful single measurement are in particular the same or unchanged.

[0058] In particular, the check information for each individual measurement of the first set and each individual measurement of the subset of the second set includes at least one piece of information that is equal, unchanged, or identical in the faulty measurement and the corresponding successful individual measurement. The at least one piece of information can be associated one-to-one with the faulty measurement and the corresponding successful individual measurement. The at least one piece of information for the individual measurement can be included in the check protocol. In particular, the check information for each individual measurement of the first set and each individual measurement of the subset of the second set can include a combination of pieces of information that are equal, unchanged, or identical in the faulty measurement and the corresponding successful individual measurement. This combination of pieces of information can be associated one-to-one with the faulty measurement and the corresponding successful individual measurement. Based on the at least one piece of information or the combination of pieces of information, a successful individual measurement in the first set can be associated with a faulty measurement in the subset of the second set. In this manner, corresponding faulty measurements and successful individual measurements can be associated with each other. If the subset corresponds to the second set, a successful individual measurement in the first set can be associated with each faulty measurement performed.

[0059] The inventors have recognized that it is possible to determine which individual measurements in an inspection protocol are particularly prone to errors in the described manner. In other words, it is possible to determine in the described manner which individual measurements provided for in the inspection protocol are performed as erroneous measurements. This allows, in particular, to infer which individual measurements provided for in the inspection protocol are performed particularly frequently with errors, i.e., as erroneous measurements. In other words, this allows to deduce which individual measurements are repeated particularly frequently.

[0060] According to another aspect of the present invention, a name is assigned to each individual measurement of the first set and the second set. The names of the individual measurements of the first set are one-to-one. The names of erroneous measurements of the second set correspond to the names of the corresponding individual measurements of the first set. The check information includes the names of the individual measurements of the first set and the individual measurements of the subset of the second set. The method step of determining whether one or more erroneous measurements has occurred includes the method step of determining identical names in the check information. The check error information includes information based on identical names.

[0061] In particular, a test error message is determined for each test of the group, which includes information based on the same name in the test information of the corresponding test.

[0062] In particular, when a successful single measurement is performed with a corrected measurement setting or corrected measurement parameters, the name of the corresponding single measurement remains unchanged compared to the name of the corresponding erroneous measurement. In particular, names in the inspection protocol can be defined or preset for the single measurements. The single measurements preset in the inspection protocol can be implemented as successful single measurements and / or erroneous measurements. The name can in particular be the naming of the single measurement. The name can in particular include the reason or target of the single measurement. The name within the single measurements of the first set can in particular be associated one-to-one with the single measurement. The name within the single measurements of the second set can in particular be associated one-to-one with the erroneous measurement. The corresponding successful single measurements of the first set can in particular be associated with the erroneous measurement based on the name. The inspection information in particular includes the name of the successful single measurement in the first set and the name of the erroneous measurement in the subset of the second set.

[0063] In the method step of determining identical names in the examination information, consistent names are determined in the examination information. In other words, in this method step, the names of individual measurements that appear multiple times in the examination information are determined. In particular, names that appear twice are determined. Because names can be associated one-to-one with individual measurements predefined in the examination protocol, names can also be associated one-to-one with individual measurements in the first set. In particular, it follows that in the case of repeated or identical names in the examination information, the corresponding individual measurements are contained not only in the individual measurements of the first set, but also in the individual measurements of the subset of the second set. In other words, a name can only appear multiple times in the examination information if the individual measurement corresponding to the name appears not only in the first set but also in the subset of the second set.

[0064] The check error information can be based on the same name. The check error information can indicate which individual measurement of the first set of individual measurements was initially incorrectly performed as an incorrect measurement. The check error information can include the same name. Alternatively or additionally, the check error information can include multiple identical names in the check information.

[0065] The inventors have recognized that known information, such as the names of individual measurements, can be used to provide information about which individual measurements were performed incorrectly. In particular, the inventors have recognized that this makes it possible to determine which individual measurements were performed particularly frequently and incorrectly. In other words, this makes it possible to determine which individual measurements were repeated.

[0066] According to another aspect of the invention, the method step of determining whether one or more erroneous measurements have been made comprises the method step of applying a trained function to the inspection information, thereby generating detection error information.

[0067] The examination information includes in particular measurement settings of the individual measurements in the first set and in a subset of the second set.

[0068] In the method step of using the trained function, test error information is generated based on the test information using the trained function. In particular, the test error information is ascertained for each test of the population using the trained function.

[0069] Typically, the trained function mimics the cognitive capabilities that connect humans to human intelligence. In particular, through training based on training data, the trained function can adapt to new situations and recognize patterns and perform extrapolation.

[0070] The parameters of the trained function can generally be adjusted by means of training. For this purpose, supervised training, semi-supervised training, unsupervised training, reinforcement learning, and / or active learning can be used, in particular. Furthermore, representation learning (an alternative term is “feature learning”) can be used. In particular, the parameters of the trained function can be iteratively adjusted over multiple training steps.

[0071] In particular, the trained function may include a neural network, a support vector machine, a random tree or decision tree, and / or a Bayesian network, and / or the trained function may be based on k-means clustering, Q-learning, a genetic algorithm, and / or association rules. In particular, the trained function may include a combination of multiple unrelated decision trees or an ensemble of decision trees (random forest). In particular, the trained function may be determined using eXtreme Gradient Boosting (XGBoosting). In particular, the neural network may be a deep neural network, a convolutional neural network, or a convolutional deep neural network. Furthermore, the neural network may be an adversarial network, a deep adversarial network, and / or a generative adversarial network. In particular, the neural network may be a recurrent neural network. In particular, the recurrent neural network can be a network with long short-term memory (LSTM), in particular a gated recurrent unit (GRU). The trained function can in particular include a combination of the described approaches. The approaches described here for the trained function are particularly referred to as the network architecture of the trained function.

[0072] The inventors have recognized that inspection error information can be determined based on inspection information by means of a trained function.

[0073] According to another aspect of the present invention, a trained function includes a trained first sub-function and a trained second sub-function. The trained first sub-function includes an unsupervised learning algorithm. The trained second sub-function includes a classification algorithm. The results of the trained first sub-function are input into the trained second sub-function.

[0074] The inspection information of the group inspection is used in particular as input data for the trained first sub-function. The output data of the trained first sub-function is used in particular as input data for the trained second sub-function. The trained second sub-function outputs inspection error information of the inspection as output data.

[0075] The examination information may include in particular the information described above. The examination information includes in particular the information described above for successful single measurements in the first set and erroneous measurements in a subset of the second set. The classification data in the examination information may in particular be converted into binary data by means of one-hot encoding (English: One-Hot-Encoding). In particular, the classification data may be information in the examination information that can only present two states, such as on / off; yes / no, etc. For example, information about the administration of contrast agents may be classification information. The information may in particular only present the expressions "yes" (contrast agent has been taken) and "no" (contrast agent has not been taken).

[0076] The trained first sub-function can be configured in particular for identifying outliers. The trained first sub-function can be configured in particular for identifying outliers in examination information. Outliers can in particular be unusual values ​​of measurement settings in the examination information. The values ​​of the outliers can in particular deviate significantly from corresponding values ​​in other examinations. In particular, an outlier can be, for example, an unusually large scan area in a single measurement. Outliers in the measurement settings of a single measurement can in particular indicate that the corresponding single measurement is an erroneous measurement. The trained first sub-function can in particular be based on k-means clustering and / or a Gaussian mixture model (Gaussian mixture model or Gaussian mixture algorithm).

[0077] The trained second sub-function can, in particular, include a classification algorithm. The trained second sub-function can, in particular, be based on an ensemble of decision trees (random forests) and / or a support vector machine (SVM). The trained second sub-function is designed to filter outliers that were identified using the trained first sub-function but were intentionally or deliberately generated or measured by the operator. Intentionally generated outliers by the operator are not, in particular, an indication of an erroneous measurement. For example, the operator can intentionally select a large scanning area to provide greater safety during the examination. This, in particular, is not an erroneous measurement.

[0078] The inventors have recognized that the combination of the trained first sub-function and the trained second sub-function ensures that the inspection error information does not include or includes a small number of outliers that are mistakenly identified as erroneous measurements. The inventors have recognized that only the annotated second training output data is required for training the trained second sub-function. The inventors have also recognized that by applying an unsupervised learning algorithm, the annotated first training output data is not required for training the trained first sub-function. The inventors have recognized that based on the output data of the trained first sub-function, the annotated second training output data can be generated more simply and more time-savingly than based on the inspection information. In other words, the inventors have recognized that by combining the first sub-function and the second sub-function, the training of the entire trained function can be optimized, accelerated or simplified.

[0079] The present invention also relates to a computer-implemented method for providing a trained function. The method comprises the steps of providing training input data, wherein the training input data comprises at least one check information regarding a check. The method further comprises the steps of providing training output data, wherein the training output data comprises at least one check error information regarding the check. The method further comprises the steps of training the trained function based on the training input data and the training output data. The method further comprises the steps of providing the trained function.

[0080] In particular, the training output data can be created manually. In particular, the training input data can be annotated, labeled, or marked for this purpose by an expert. In other words, an expert can identify erroneous measurements in the training input data for use in determining the training output data. Alternatively, successful individual measurements can be identified in the training input data in order to determine or generate the training output data. In particular, the training output data can be determined or generated automatically.

[0081] The training can be performed in two phases. In particular, if the trained function comprises a trained first sub-function and a trained second sub-function as described above, the training is performed in two phases.

[0082] In particular, a first trained sub-function can be trained in a first step or phase. The trained first sub-function is particularly trained based on training input data. Sub-training output data is generated or determined. The first sub-training output data particularly includes information about outliers in the measurement settings included in the examination information for individual measurements. The first sub-training output data can particularly be determined or ascertained automatically using unsupervised learning. Alternatively, the first sub-training output data can be manually annotated.

[0083] The trained second sub-function can then be trained, in particular, in a second step or second phase. The trained second sub-function can be trained, in particular, based on the first sub-training output data and the training output data as the second sub-training input data. The training output data can be determined, in particular, by an expert using manual annotation from the second sub-training input data or the first sub-training output data. The trained second sub-function can be trained, in particular, using supervised training.

[0084] The inventors have recognized that a two-stage training is an effective method for training a trained function for automatically determining inspection error information for an inspection.

[0085] According to one aspect of the invention, the trained function is continuously further trained by means of feedback. Here, feedback is provided by the user checking the error message.

[0086] A user who checks the error information is also, in particular, a user of the error information. The user can indicate, in particular via the user interface, in the form of user information, whether the user is satisfied with the provided results regarding the error information. The user can provide corrected error information as user information via the user interface. The user information can be used, in particular, as training output data for training the trained function.

[0087] The inventors have realized that in this way, the trained function can be continuously further trained and optimized in the application.

[0088] The present invention also relates to a system for providing error information regarding a plurality of individual measurements, the system comprising a computing unit and an interface. The computing unit and / or the interface are configured to provide a plurality of individual measurements, each of which is associated with an examination. The interface is further configured to receive examination information about a population to be examined from a user. The computing unit is further configured to extract examination information for each examination in the population based on the individual measurements associated with the examination. The computing unit is further configured to determine, based on the examination information, whether one or more erroneous measurements were made during the examination for each examination in the population. The computing unit is further configured to determine, based on the determined results and the examination information, examination error information for each examination in the population. The computing unit is further configured to generate error information based on the examination error information for the population to be examined. The computing unit and / or the interface are further configured to provide the error information to the user.

[0089] Such a system for providing error information about a plurality of individual measurements can be designed, in particular, for implementing the previously described method for providing error information about a plurality of individual measurements of at least one medical examination and aspects thereof. The system for providing error information about a plurality of individual measurements is designed for implementing the method and aspects thereof in that the interface and the computing unit are designed for implementing the corresponding method steps.

[0090] The present invention also relates to a computer program product and a computer-readable medium having a computer program. A largely software-based implementation has the advantage that previously used systems for providing error information about a plurality of individual measurements can also be easily retrofitted by means of a software update to operate in the described manner. In addition to the computer program, such a computer program product may optionally include additional components, such as, for example, documentation and / or additional components, as well as hardware components, such as, for example, a hardware key (dongle, etc.) for using the software.

[0091] In particular, the present invention also relates to a computer program product having a computer program that can be directly loaded into the memory of a system for providing error information about a plurality of individual measurements, the computer program product having a program segment so that when the program segment is executed by the system for providing error information about a plurality of individual measurements, all steps of the method for providing error information about a plurality of individual measurements of at least one medical examination and aspects of the method described above are performed.

[0092] In particular, the present invention relates to a computer-readable storage medium on which a program segment is stored that is readable and executable by a determination system and / or a training system so that when the program segment is executed by a system for providing error information about multiple individual measurements, all steps of the method for providing error information about multiple individual measurements of at least one medical examination and aspects of the method described above are performed.

[0093] The above characteristics, features and advantages of the present invention will become more clear and easier to understand with reference to the following drawings and descriptions. The drawings and descriptions should not limit the present invention and its embodiments in any way. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] In the different figures, identical components are provided with corresponding reference numerals. The figures are generally not true to scale.

[0095] The accompanying drawings show:

[0096] Figure 1 A first exemplary embodiment of a method for providing error information about a plurality of individual measurements is shown.

[0097] Figure 2 A second exemplary embodiment of a method for providing error information about a plurality of individual measurements is shown.

[0098] Figure 3 A third exemplary embodiment of a method for providing error information about a plurality of individual measurements is shown.

[0099] Figure 4 A fourth exemplary embodiment of a method for providing error information about a plurality of individual measurements is shown.

[0100] Figure 5 A fifth exemplary embodiment of a method for providing error information about a plurality of individual measurements is shown.

[0101] Figure 6 A sixth exemplary embodiment of a method for providing error information about a plurality of individual measurements is shown.

[0102] Figure 7 A first embodiment for providing a trained function is shown,

[0103] Figure 8 A second embodiment for providing a trained function is shown,

[0104] Figure 9 A system for providing error information about a plurality of individual measurements is shown.

[0105] Figure 10 A training system for providing a trained function is shown. DETAILED DESCRIPTION

[0106] Figure 1 A first exemplary embodiment of a method for providing error information regarding a plurality of individual measurements is shown.

[0107] In the method step of providing PROV-01 multiple individual measurements, the multiple individual measurements are provided, for example, via a database. Here, the individual measurements can each be associated with an examination. In other words, the examination includes one or more individual measurements. The examination can be implemented or performed with the aid of a medical device. Here, the individual measurements associated with the examination are detected, recorded, or measured with the aid of the medical device. The individual measurements can in particular be transferred to the database automatically or manually. Alternatively, providing PROV-01 can be performed directly via the medical device. The medical device can in particular be an imaging system, such as, for example, a flat-panel X-ray device, a CT device, an MRT device, a PET device, a SPECT device, or a mammography device. Alternatively, the medical device can be a laboratory diagnostic device. In particular, multiple individual measurements can be recorded with the aid of more than one medical device.

[0108] An examination is performed, in particular, on a patient. The examination includes one or more individual measurements. The individual measurements of the examination may differ, for example, in measurement settings, such as the patient's examined body region and / or scan region and / or measurement time and / or reagent used and / or exposure time and / or exposure duration and / or recording contrast and / or, in the case of an X-ray radiation-based imaging system, the energy of the X-ray radiation. The measurement settings may be adapted, in particular, to the patient and / or the medical examination. An individual measurement may be a successful individual measurement or an erroneous measurement. An erroneous measurement is an erroneous individual measurement. In particular, in the case of an erroneous measurement, one or more measurement settings may be incorrectly set or selected. In particular, the erroneous individual measurement or the erroneous measurement is repeated using corrected measurement settings. In particular, the plurality of individual measurements may include multiple successful individual measurements and may include one or more erroneous measurements. In particular, for each erroneous measurement, the plurality of individual measurements may include a corresponding successful individual measurement. The corresponding successful individual measurement is a repeat of the erroneous measurement using corrected measurement settings. In particular, the examination includes a first set of individual measurements and a second set of individual measurements. The first set of individual measurements includes, in particular, all successful or correctly executed individual measurements associated with the check. The second set of individual measurements includes, in particular, error measurements associated with the check. Each error measurement in the second set can, in particular, be associated with an individual measurement from the first set. The second set can, in particular, be empty. Alternatively, the second set can include as many individual measurements as the first set.

[0109] The individual measurements may include, in particular, measurement data and / or metadata regarding the individual measurements.

[0110] The measurement data may in particular be image data. The measurement data may in particular be data which are detected or measured or determined during a single measurement.

[0111] In this case, metadata includes information about an examination or a single measurement associated with the examination. Metadata can be associated with the examination in particular by means of the single measurement that includes the metadata. In particular, metadata can include, for example, a DICOM header and / or a dose structured report (English: Dose Structured Report). In particular, metadata can be queried, for example, by means of DICOM queries and retrievals. Alternatively, metadata can be, for example, a laboratory protocol, etc. In particular, metadata can include information about one or more measurement settings and / or about the medical technology equipment used for the examination and / or about the operator performing the examination and / or about the patient on whom the examination was performed, etc.

[0112] The database can be an internal database in particular. Alternatively, the database can be an external database. In particular, the database can be a central database. The central database can, for example, be saved or stored on a server and / or in a cloud system. For image data, the central database can include, for example, a picture archiving and communication system (PACS). Individual measurements of multiple medical technology devices can be loaded or transferred into the central database. In particular, all successful individual measurements included in the individual measurements of the first set are transferred to the database. In particular, individual measurements of a subset of the second set are transferred to the database. In other words, a subset of the erroneous measurements performed during the examination is transferred to the database. The subset can be empty in particular. Alternatively, the subset can correspond to the second set. The operator can, in particular, determine which erroneous measurements should be transferred to the database. The subset then includes, in particular, the erroneous measurements transferred to the database. In particular, all erroneous measurements or none of the erroneous measurements can be transferred to the database automatically.

[0113] In the method step of receiving REC test information, test information about the examined population is received from the user via an interface. In particular, in the method step of providing PROV-1 multiple individual measurements, individual measurements associated with the examination included in the population are provided. The user can be, in particular, a person who analyzes the occurrence of erroneous measurements. The interface can be, in particular, configured as a keyboard and / or a GUI and / or a touch-sensitive screen. The user can specify, via the test information, criteria or test criteria according to which the population is to be associated with the examination. For example, the criteria can be the manufacturer of the medical-technical equipment used to perform the examination, the operator during the examination, the symptoms or causes of the examination, etc. The test information can, in particular, include one or more of these criteria.

[0114] In the method step of extracting EXT examination information, examination information is extracted based on the individual measurements belonging to the examination. In particular, the examination information can be extracted from metadata included in the individual measurements. The examination information in particular includes information about the individual measurements included in the examination. Alternatively or additionally, the examination information includes information about the examination. The examination information in particular may include information about the manufacturer of the medical technology equipment used to perform the examination. The examination information in particular may include information about the operator performing the examination. The examination information in particular may include at least one of the following information: the name of the individual measurement, the type of the individual measurement, the number of examinations of the individual measurement, the number of the individual measurement, the body area measured in the individual measurement, the dose-length product of the individual measurement, the X-ray voltage of the individual measurement, the name of the examination protocol, and the reason for the examination.

[0115] In the method step of determining DET-1, it is determined based on the examination information for each examination of the population whether one or more erroneous measurements were made. In particular, the method step is performed for the examination for which the determination is made to check whether one or more erroneous measurements were made during the examination. In other words, the method step determines whether one or more individual measurements of the examination were performed erroneously.

[0116] In the method step of determining DET-2 test error information, test error information is determined for each test in the population based on the result of determining DET-1. The test error information for a test includes, in particular, information about whether at least one erroneous measurement was performed during the test. The test error information for a test may include, in particular, information about how many erroneous measurements were performed during the test. The test error information may include, in particular, information about which individual measurements of the test were performed erroneously.

[0117] The method steps of extracting EXT examination information, determining whether DET-1 has made one or more erroneous measurements during the examination, and determining DET-2 examination error information are carried out or executed in particular for each examination in the group examination. Figure 1 In the diagrams in FIG, large boxes are shown around corresponding method steps, which boxes include the corresponding method steps.

[0118] In the method step of compiling DET-3 error information, error information is compiled based on examination error information from a population to be tested. In particular, the examination error information from the population can be compared and / or statistically evaluated to compile the DET-3 error information. Furthermore, the error information can be based on the examination information from the population. For example, the error information can indicate which individual measurements are most frequently performed incorrectly, and / or in which medical devices, incorrect measurements are most frequently performed, and / or by which operators, and / or for which symptoms or causes of examination, or for which diseases or conditions, incorrect measurements are most frequently performed, and / or the causes of the incorrect measurements. With respect to medical devices, the error information can include, among other things, information regarding the occurrence of incorrect measurements based on the manufacturer of the medical device, and / or the maintenance status of the medical device, and / or the year of manufacture of the medical device, and / or the series and / or type of the medical device. The cause of an incorrect measurement can be determined, in particular, by comparing the incorrect measurement with a corresponding successful individual measurement. Measurement settings that are altered in the successful individual measurement compared to the incorrect measurement can indicate, in particular, that the measurement settings are the cause of the incorrect measurement.

[0119] In the method step of providing PROV-2 error information, the error information is provided to the user via an interface. In particular, the error information can be provided to the user via a display unit. The display unit can be, in particular, a screen or a computer screen. The error information can be displayed on the display unit in a GUI. Alternatively, the error information can be stored on a storage medium from which the user can retrieve the error information. The storage medium can be, in particular, a hard disk, an SD card, a CD, a DVD, a USB memory stick, etc. The error information can be stored in a database from which the user can retrieve the error information. Alternatively, the error information can be sent to the user via e-mail and / or SMS (Short Message Service).

[0120] Figure 2 A second exemplary embodiment of a method for providing error information regarding a plurality of individual measurements is shown.

[0121] The method steps of providing PROV-1 multiple individual measurements, receiving REC verification information, extracting EXT check information, determining whether DET-1 made one or more erroneous measurements, obtaining DET-2 check error information, compiling DET-3 error information, and providing PROV-2 error information are similar to those described in connection with Figure 1 Description to implement.

[0122] In the method step of providing a PROV-3 examination protocol, an examination protocol is provided for each examination of the population. In particular, the successful individual measurements included in the examination are defined or preset in the examination protocol. In other words, the individual measurements of the first set of examinations are preset or defined in the examination protocol. In particular, corresponding measurement settings are defined in the examination protocol for each individual measurement of the examination. In particular, the examination protocol can be designed for the examination of a specific disease. The examination protocol particularly defines the ideal process of the examination. In particular, the examination protocol defines the minimum number of individual measurements required to perform the examination. In particular, the operator can adjust one or more measurement settings in the examination protocol. In particular, the operator can correct the corresponding measurement settings in the examination protocol to correct erroneous measurements. In particular, the number of individual measurements preset in the examination protocol is not changed.

[0123] In particular, the method step of determining whether DET-1 has made one or more erroneous measurements during the examination can additionally be based on the examination protocol.

[0124] In an alternative embodiment, a plurality of examination protocols can be provided in the method step of providing a plurality of individual measurements in PROV-1. An examination can then be associated with each examination protocol. In other words, an examination protocol can be associated with each examination.

[0125] Figure 3 A third exemplary embodiment of a method for providing error information regarding a plurality of individual measurements is shown.

[0126] The method steps of providing PROV-1 multiple individual measurements, receiving REC verification information, extracting EXT check information, determining whether DET-1 made one or more erroneous measurements, obtaining DET-2 check error information, compiling DET-3 error information, and providing PROV-2 error information are similar to those described in connection with Figure 1 The method steps of the PROV-3 inspection protocol are implemented according to the Figure 2 Description to achieve.

[0127] The method step of determining whether DET-1 performed one or more erroneous measurements includes the method step of determining a DET-11 difference. In the method step of determining the DET-11 difference, the difference between the number of protocols for the individual measurements and the number of checks for the individual measurements is determined. The number of protocols for the individual measurements corresponds in particular to the number of individual measurements predefined in the test protocol. The number of checks for the individual measurements checked corresponds to the number of individual measurements in the first set of individual measurements plus the number of individual measurements in the subset of the second set. The number of checks for the individual measurements corresponds in particular to the number of individual measurements provided for the test in the method step of providing PROV-1 multiple individual measurements. The number of checks corresponds in particular to the number of individual measurements transferred to the database for the test. The difference corresponds in particular to the number of erroneous measurements in the individual measurements of the subset of the second set. In particular, if the subset of the second set is equal to the second set, the difference corresponds to the total number of erroneous measurements checked. In other words, if all erroneous measurements are transferred to the database, the difference corresponds to the total number of erroneous measurements. In particular, if all erroneous measurements for the test are provided in the step of providing PROV-1 multiple individual measurements, the difference corresponds to the total number of erroneous measurements.

[0128] The check error information can then be based in particular on the difference value. The check error information can in particular comprise the difference value.

[0129] Figure 4 A fourth exemplary embodiment of a method for providing error information regarding a plurality of individual measurements is shown.

[0130] The method steps of providing PROV-1 multiple individual measurements, receiving REC verification information, extracting EXT check information, determining whether DET-1 made one or more erroneous measurements, obtaining DET-2 check error information, compiling DET-3 error information, and providing PROV-2 error information are similar to those described in connection with Figure 1 Description to implement.

[0131] The method step of determining whether DET-1 has made one or more erroneous measurements includes the method step of checking CHECK-12. In the method step of checking CHECK-12, it is checked whether the numbers of the individual measurements in the check information are consecutive. In particular, each individual measurement of the check is assigned a consecutive number. In other words, each individual measurement of the first set and the second set includes a one-to-one number. The check information includes in particular the numbers of the individual measurements provided in the step of providing PROV-1 with a plurality of individual measurements of the check. In other words, the check information includes the numbers of the individual measurements of a subset of the first set and the second set. In particular, if the numbers of the erroneous measurements that were not provided or not transmitted are not included in the check information, the erroneous measurements that were not provided or not transmitted can be identified or determined. In particular, the erroneous measurements that were not provided can be identified or determined by "gaps" in the numbers in the check information. In particular, when the subset of the second set is empty, all erroneous measurements can be identified. In particular, the number of erroneous measurements can be determined in this manner based on the missing numbers.

[0132] The check error information can then be based in particular on the number of faulty measurements. The check error information can in particular comprise the number of faulty measurements.

[0133] This embodiment can in particular be combined with the third embodiment.The total number of all erroneous measurements made during the examination can then in particular be determined, irrespective of whether this total number is part of a single measurement of a subset of the second set.

[0134] Figure 5 A fifth exemplary embodiment of a method for providing error information regarding a plurality of individual measurements is shown.

[0135] The method steps of providing PROV-1 multiple individual measurements, receiving REC verification information, extracting EXT check information, determining whether DET-1 made one or more erroneous measurements, obtaining DET-2 check error information, compiling DET-3 error information, and providing PROV-2 error information are similar to those described in connection with Figure 1 Description to implement.

[0136] The method step of determining whether DET-1 performed one or more erroneous measurements includes the method step of determining identical names for DET-13. In the method step of determining identical names for DET-13, identical names for individual measurements are determined in the test information. In other words, names that appear two or more times are determined in the test information. The test information particularly includes the name for each individual measurement provided in the method step of providing PROV-1 multiple individual measurements. The names are one-to-one for the corresponding individual tests. The names of the corresponding individual measurements are identical. The name of the erroneous measurement is particularly identical to the corresponding successful individual measurement. The corresponding individual measurement is particularly a repetition of the erroneous measurement using corrected measurement settings. The test information particularly includes the names of the individual measurements of the first set and the individual measurements of a subset of the second set. Ideally, each individual measurement with the one-to-one name is performed only once during the test. In particular, two identical names in the test information can be used to infer that one of the two names represents an erroneous measurement. In particular, it can also be inferred which individual measurement was performed once as an erroneous measurement.

[0137] The error check information may include, in particular, information about identical names. The error check information may include, in particular, information about how many identical names can be determined in the error check information. The error check information may then include, in particular, information about the number of erroneous measurements in the subset of the second set. Alternatively or additionally, the error check information may include information about which individual measurements were performed as erroneous measurements. The error check information may include, in particular, identical names or names that occur twice or more.

[0138] This exemplary embodiment can be combined in particular with the third exemplary embodiment and / or the fourth exemplary embodiment.

[0139] Figure 6 A sixth exemplary embodiment of a method for providing error information regarding a plurality of individual measurements is shown.

[0140] The method steps of providing PROV-1 multiple individual measurements, receiving REC verification information, extracting EXT check information, determining whether DET-1 made one or more erroneous measurements, obtaining DET-2 check error information, compiling DET-3 error information, and providing PROV-2 error information are similar to those described in connection with Figure 1 Description to implement.

[0141] The method step of determining whether DET-1 has made one or more erroneous measurements includes the method step of applying a trained function APP-14 to the examination information. This generates examination error information. The input data of the trained function includes, in particular, the examination information. The output data of the trained information includes, in particular, the examination error information. The trained function may include, in particular, a trained first sub-function and a trained second sub-function. The trained first sub-function may include, in particular, an unsupervised or unsupervised learning algorithm. The trained first sub-function may be based, in particular, on a k-means clustering algorithm and / or a Gaussian mixture model. The trained second sub-function may include, in particular, a classification algorithm. The trained second sub-function may be based, in particular, on an ensemble of decision trees (random forests) and / or a support vector machine. The output data of the trained first sub-function may be used, in particular, as input data for the trained second sub-function. In particular, the trained first sub-function may be used to identify outliers in the examination information. In other words, using the trained first sub-function, it is possible to identify individual measurements in the examination information that have at least one measurement setting that significantly deviates from the measurement setting used according to the standard. Such deviations can, in particular, indicate an erroneous measurement. In particular, such deviations may be intentional by the operator and do not indicate an erroneous measurement. The trained second sub-function can, in particular, be designed to identify whether this is an intended or intentional deviation or an erroneous measurement.

[0142] The error check information includes, in particular, the results of the trained function. The error check information may include, in particular, information about the number of erroneous measurements and / or about which individual measurements were performed as erroneous measurements and / or which measurement settings caused the erroneous measurements, etc. In particular, it is possible to infer from the deviation which measurement settings caused the erroneous measurements.

[0143] Figure 7 A first embodiment for providing a trained function is shown.

[0144] In the method step of providing TPROV-1 training input data, training input data for training the trained function is provided. The training input data includes at least one inspection information about the inspection. The inspection information can be constructed as described above.

[0145] In the method step of providing TPROV-2 training output data, training output data is provided for training the function to be trained. The training output data includes at least one test error information of the test.

[0146] The training output data may be based on the training input data. In other words, the training output data may be correlated with the training input data. The training output data may be determined manually and / or automatically from the training input data. In the case of manual determination, an expert may determine the training output data based on the training input data. The expert may determine the training output data by annotating the training input data.

[0147] In the method step of training a trained function (TRAIN), the trained function is trained based on training input data and training output data. In particular, output data is determined from the training input data using the trained function. In other words, the output data is determined, generated, or ascertained by applying the trained function to the training input data. In particular, the output data is compared with the training output data. The trained function is particularly adjusted so that the output data corresponds as closely as possible to the training output data. This is particularly referred to as "supervised learning."

[0148] In the method step of providing a TPROV-2 trained function, the trained function is provided for application to unknown input data with unknown output data.

[0149] Figure 8 A second embodiment for providing a trained function is shown.

[0150] The method steps of providing TPROV-1 training input data, providing TPROV-2 training output data and providing TPROV-3 trained function are similar to those of Figure 8 Description implementation.

[0151] The method step of training the trained function of TRAIN comprises the method steps of training a first trained sub-function of SUB-TRAIN-1 and training a second trained sub-function of SUB-TRAIN-2.

[0152] The trained first sub-function and the trained second sub-function can be Figure 7 In the method step of training the first sub-function trained by SUB-TRAIN-1, the first sub-function trained is trained in particular by means of unsupervised learning. In the method step of training the second sub-function trained by SUB-TRAIN-2, the second sub-function trained is trained in particular by means of supervised learning. Figure 7 The training input data described in the embodiment of the present invention are used in particular as training input data of the first sub-function to be trained. The sub-training output data of the first sub-function to be trained are used in particular as sub-training input data of the second sub-function to be trained. Figure 7The training output data described in the embodiment of the present invention are used in particular as training output data for the second sub-function to be trained. The training output data can be determined manually and / or automatically from the sub-training input data of the second function to be trained. Supervised learning can in particular be similar to Figure 7 Description to achieve.

[0153] Figure 9 A system SYS for providing error information about a plurality of individual measurements is shown. Figure 10 A training system TSYS for providing trained functions is shown.

[0154] The illustrated system SYS for providing error information about a plurality of individual measurements is configured to carry out the method according to the present invention for providing error information about a plurality of individual measurements. The illustrated training system TSYS is configured to carry out the method according to the present invention for providing a trained function. The system SYS comprises an interface SYS.IF, a calculation unit SYS.CU, and a memory unit SYS.MU. The training system TSYS comprises a training interface TSYS.IF, a training calculation unit TSYS.CU, and a training memory unit TSYS.MU.

[0155] The system SYS and / or the training system TSYS can be, in particular, a computer, a microcontroller, or an integrated circuit (IC). Alternatively, the system SYS and / or the training system TSYS can be a real or virtual computer network (the technical term for a real computer network is "cluster," and the technical term for a virtual computer network is "cloud"). The system SYS and / or the training system TSYS can be configured as a virtual system (the technical term is "virtualization") running on a computer or a real computer network or a virtual computer network.

[0156] The interface SYS.IF and / or the training interface TSYS.IF can be a hardware or software interface (e.g., a PCI bus, USB, or FireWire). The computing unit SYS.CU and / or the training computing unit TSYS.CU can include hardware and / or software components, such as a microprocessor or a so-called FPGA (Field Programmable Gate Array). The memory unit SYS.MU and / or the training memory unit TSYS.MU can be designed as a non-persistent working memory (Random Access Memory, RAM) or as a persistent mass storage device (hard disk, USB memory stick, SD card, solid-state drive (SSD)).

[0157] The interface SYS.IF and / or the training interface TSYS.IF can in particular comprise a plurality of sub-interfaces, which carry out different method steps of the respective method according to the invention. In other words, the interface SYS.IF and / or the training interface TSYS.IF can be configured as a plurality of interfaces SYS.IF and / or training interfaces TSYS.IF. The computing unit SYS.CU and / or the training computing unit TSYS.CU can in particular comprise a plurality of sub-computing units, which carry out different method steps of the respective method according to the invention. In other words, the computing unit SYS.CU and / or the training computing unit TSYS.CU can be configured as a plurality of computing units SYS.CU and / or training computing units TSYS.CU.

[0158] Even if not explicitly stated, individual embodiments, individual aspects or features of an embodiment may be combined or exchanged with one another, if this is meaningful and within the meaning of the present invention, without departing from the scope of the present invention. The advantages of the present invention described with reference to an embodiment also apply to the other embodiments, if applicable, unless explicitly stated otherwise.

Claims

1. A computer-implemented method for providing error information about a plurality of individual measurements, the method comprising the following method steps: - providing (PROV-1) a plurality of individual measurements, each of which is associated with an examination, wherein each examination comprises one or more individual measurements; - receiving (REC) from the user via the interface test information about the population to be tested for the examination; For each check of the group: - extracting (EXT) examination information based on said single measurement belonging to said examination; - determining (DET-1) based on the inspection information whether one or more erroneous measurements were made during the inspection; as well as - determining (DET-2) a check error message based on the result of the determination (DET-1); - compiling (DET-3) the error information based on the examination error information of the population to be tested; as well as - providing (PROV-2) said error information to the user via said interface, wherein the examination comprises a first set of individual measurements and a second set of individual measurements, wherein said first set comprises a single successful measurement, wherein said second set comprises error measurements made, wherein the inspection information comprises at least one information about a single measurement of the first set, and wherein the inspection information comprises at least one information about a single measurement of a subset of the second set, wherein individual measurements of the first set are predefined in the examination protocol, The method further comprises the following steps: - provide (PROV-3) the inspection protocol, and The step of determining (DET-1) whether one or more erroneous measurements were made during the examination based on the examination protocol.

2. The method according to claim 1, wherein the error information comprises a statistical evaluation of the error information of the population.

3. The method according to claim 1, The examination information includes at least one of the following information about the individual measurements of the first set and the individual measurements of a subset of the second set: the name of the individual measurement, the type of the individual measurement, the body area measured in the individual measurement, the dose-length product of the individual measurement, the X-ray voltage of the individual measurement, the naming of the examination protocol, and the reason for the examination.

4. The method according to claim 3, The method step of determining whether (DET-1) has made one or more erroneous measurements comprises the following method steps: - determine (DET-11) the difference between the number of agreements for a single measurement and the number of checks for a single measurement, wherein the number of protocols for single measurements comprises the number of single measurements in the first set of single measurements, wherein the examined number of single measurements comprises the number of single measurements in the single measurements of the first set and in the single measurements of the subset of the second set, The check error information is based on a difference between the protocol quantity and the check quantity.

5. The method according to claim 1, wherein each individual measurement of the first set and the second set is assigned a consecutive number, wherein the inspection information comprises numbers of individual measurements of the first set and numbers of individual measurements of a subset of the second set, The method step of determining whether (DET-1) has made one or more erroneous measurements comprises the following method steps: - checking (CHECK-12) whether the numbers of the individual measurements in the check information are consecutive, The inspection error information includes the result of the inspection.

6. The method according to claim 1, wherein an erroneous measurement of the second set of single measurements corresponds to a single measurement of the first set of single measurements, The check error information includes the following information: the information indicates which individual measurements of the first set of individual measurements have corresponding erroneous measurements in the subset of the second set.

7. The method according to claim 1, wherein each individual measurement of the first set and the second set is assigned a name, where the names of the individual measurements of the first set are one-to-one, wherein the names of the error measurements of the second set correspond to the names of the corresponding single measurements of the first set, wherein the inspection information comprises names of individual measurements of the first set and of the individual measurements of the subset of the second set, The method step of determining whether (DET-1) has made one or more erroneous measurements comprises the following method steps: - confirm (DET-13) the same name in the inspection information, The check error information includes information based on the same name.

8. The method according to claim 1 or 2, The method step of determining whether (DET-1) has made one or more erroneous measurements comprises the following method steps: - applying (APP-14) the trained function to said examination information, This results in the check error message.

9. The method according to claim 8, The trained function includes a trained first sub-function and a trained second sub-function, wherein the trained first sub-function comprises an unsupervised learning algorithm, and wherein the trained second sub-function comprises a classification algorithm, The result of the trained first sub-function is input into the trained second sub-function.

10. The method according to claim 2, The statistical evaluation is performed in this case with respect to the frequency of erroneous measurements at the medical device and / or the frequency of erroneous measurements caused by an operator and / or the frequency of erroneous measurements during examinations of specific diseases.

11. A computer-implemented method for providing a trained function, the method comprising: - Provide (TPROV-1) training input data, wherein the training input data comprises at least one examination information about an examination, wherein each examination comprises one or more individual measurements, - Provide (TPROV-2) training output data, wherein the training output data includes at least one inspection error information of the inspection, wherein the training output data and the training input data are related to each other, - training (TRAIN) the trained function based on the training input data and the training output data, - providing (TPROV-3) said trained function, wherein the examination comprises a first set of individual measurements and a second set of individual measurements, wherein said first set comprises a single successful measurement, wherein said second set comprises error measurements made, wherein the inspection information comprises at least one information about a single measurement of the first set, and wherein the inspection information comprises at least one information about a single measurement of a subset of the second set, wherein individual measurements of the first set are predefined in the examination protocol, The method further comprises the following steps: - provide (PROV-3) the inspection protocol, and The step of determining (DET-1) whether one or more erroneous measurements were made during the examination based on the examination protocol.

12. The method according to claim 11, wherein the trained function is continuously further trained with the aid of feedback, The feedback is provided by the user who checks the error information.

13. A system (SYS) for providing error information about a plurality of individual measurements, comprising a calculation unit (SYS.CU) and an interface (SYS.IF), wherein the computing unit (SYS.CU) and / or the interface (SYS.IF) are designed to provide (PROV-1) a plurality of individual measurements, each of which is assigned to an examination, wherein each examination comprises one or more individual measurements; wherein the interface (SYS.IF) is further configured to receive (REC) from a user test information regarding the population to be tested for the examination; wherein the calculation unit (SYS.CU) is further configured to extract (EXT) examination information for each examination of the population based on the individual measurements belonging to the examination; wherein the computing unit (SYS.CU) is further configured to determine (DET-1) for each examination of the population based on the examination information whether one or more erroneous measurements were made in the examination; wherein the calculation unit (SYS.CU) is further designed to determine (DET-2) a test error information for each test of the population based on the result of the determination (DET-1) and the test information; wherein the calculation unit (SYS.CU) is further configured to compile (DET-3) the error information based on the check error information of the population to be checked; wherein the computing unit (SYS.CU) and / or the interface (SYS.IF) is further designed to provide (PROV-2) the error information to a user, wherein the examination comprises a first set of individual measurements and a second set of individual measurements, wherein said first set comprises a single successful measurement, wherein said second set comprises error measurements made, wherein the inspection information comprises at least one information about a single measurement of the first set, and wherein the inspection information comprises at least one information about a single measurement of a subset of the second set, wherein individual measurements of the first set are predefined in the examination protocol, The computing unit (SYS.CU) and / or the interface (SYS.IF) may further comprise: Provide (PROV-3) the inspection protocol, and Based on the inspection protocol, it is determined (DET-1) whether one or more erroneous measurements were made during the inspection.

14. A computer program product comprising a computer program that can be directly loaded into a memory (SYS.MU) of a system (SYS), the computer program product comprising a program segment so that when the program segment is executed by the system (SYS), all method steps of the method according to any one of claims 1 to 12 are performed.

15. A computer-readable storage medium having stored thereon program segments readable and executable by a system (SYS) so that, when the program segments are executed by the system (SYS), all method steps of the method according to any one of claims 1 to 12 are performed.

Images