Computer-implemented method, system or data processing unit and computer program product for controlling units in a medical information network

Abstract

Computer-implemented method, system or data processing unit (SYS) and computer program product for controlling units in a medical information network, comprising a computing unit (SYS.CU), a reporting station (AS), a data acquisition unit (SYS.DCU) and a storage unit (SYS.DB), wherein the computing unit (SYS.CU) is configured to automatically display current patient information (CPI) via a user interface (AS).UI), to automatically apply a trained function (TF2) to displayed current patient information (CPI) and to stored historical patient information (HPI), and to automatically provide selected stored historical patient information (HPI) depending on a determined relevance measure for this stored historical patient information (HPI) and / or depending on captured user activity data (AD) regarding a user's activity in relation to the displayed patient information (CPI).

Description

[0001] The invention relates to a computer-implemented method, a system or a data processing unit and a computer program product for controlling units in a medical information network.

[0002] The invention is particularly concerned with supporting physicians in the preparation of medical reports. Physicians in many disciplines are responsible for monitoring patients and conducting diagnostic examinations over a patient's treatment period. Examples of this include imaging examinations and the observation of various patient information, such as symptoms, blood values, tumor markers, and measurements related to the patient's physical condition (e.g., weight, pulse, body fat percentage).

[0003] Patient information is often distributed across multiple databases, such as medical information systems and various medical devices. Accessing diverse patient information can prove difficult due to incompatibilities between the information systems and medical devices and / or restricted access rights in practice, and is very time-consuming for the physician. Written patient information, such as diagnostic findings or symptom descriptions, in particular, can be very laborious to record and interpret and is difficult to interpret in the context of other patient information, such as blood test results or diagnostic images. This distributed patient information could contribute significantly to an assessment of the patient's health status, but is often inaccessible to the physician, or at least only with considerable effort.

[0004] Especially in the area of ​​imaging examinations, physicians currently work with multiple information systems and applications to analyze and process imaging examinations and also to access existing, stored clinical patient information. It is the responsibility of the user to authenticate themselves with the various systems and actively search for the relevant information for the current patient. It is clear that this manual process is time-consuming and error-prone, often causing delays in the evaluation of examination results, leading to the scheduling of unnecessary diagnostic tests, and even compromising the patient's entire treatment process.

[0005] From the earlier patent application DE 10 2022 201 630 A1 of the patent applicant, a computer-implemented method for providing information about a health status of a patient is known, comprising the steps of: receiving a plurality of patient information items of the patient by means of an interface, determining health information about the patient as a function of the plurality of patient information items and a first function by means of a computing unit and checking whether a trigger condition is met based on the determined health information about the patient, and providing the determined health information about the patient as a function of the trigger condition.The invention further relates to a system comprising a computing unit and a user interface, wherein the computing unit is configured to coordinate and carry out a method according to the invention and to provide determined information about a patient's health status via the user interface. According to this document, the patient's health status can be characterized by the plurality of patient information items. Furthermore, the health status can also be characterized by a correlation of one or more medical parameters and / or derived variables of the patient with one or more corresponding medical parameters and / or derived variables of a reference patient.A correlation can, in particular, comprise a comparison and / or quantification of a deviation of one or more medical parameters and / or derived variables of the patient with respect to one or more corresponding medical parameters and / or derived variables of a reference patient. According to one aspect, the first function is designed to detect multivariate outliers in patient information. In other words, in order to improve the provided health information, the aforementioned earlier application by the patent applicant provides for the automatic processing of the patient information of the present patient to search for similar patients for whom previously obtained health information is (pre-)known and, optionally, verified.

[0006] The technical teaching of the aforementioned prior patent application of the patent applicant DE 10 2022 201 630 A1, as well as the above technical features, are hereby incorporated into the disclosure of the present invention as possible developments and / or possible further or supplementary embodiments of the present invention. In particular, those individual technical aspects of the aforementioned prior patent application of the patent applicant that a person skilled in the art would consider helpful or supportive for the further development or implementation of the present invention are incorporated into the disclosure of the present invention.

[0007] It is an object of the present invention to provide an improved computer-implemented method, an improved system, and an improved computer program product for controlling units in a medical information network. In particular, it is an object of the present invention to provide an improved option for automated support of physicians in the creation of medical reports on a patient's health status based on a plurality of patient information items.

[0008] This object is achieved according to the invention by the subject matter of the independent patent claims. Advantageous embodiments and expedient further developments are the subject matter of the dependent claims. According to a first aspect of the present invention, a computer-implemented method for controlling units in a medical information network is provided, the medical information network comprising: a computing unit, as well as at least the following units connected to the computing unit: - at least one diagnostic station for the preparation of a medical report, - at least one data acquisition unit for recording current patient information, and - at least one storage unit for storing historical patient information.

[0009] The computer-implemented method according to the invention comprises at least the following steps: - automatic recording of current patient information of a patient using the data recording unit, - automatic display of the patient's current patent information via a user interface of the diagnostic station, - automatic retrieval of stored historical patient information - in particular patient information of the same patient and / or reference patients -, - automatic application of a trained, computer-implemented function executable in the processing unit or the diagnosis station to the displayed current patient information and to the determined stored historical patient information - in particular to patient information of the same patient and / or of reference patients -, - automatically determining a relevance measure for the determined stored historical patient information - in particular for patient information of the same patient and / or reference patients - based on the trained function, and - automatic provision of selected stored historical patient information - in particular patient information of the same patient and / or reference patients - by means of the user interface of the diagnostic station depending on the determined relevance level for this stored historical patient information.

[0010] In particular, in the aforementioned aspect of the invention and in any further aspect of the invention described in this document, the following steps can be provided alternatively or additionally, or the aforementioned steps of an aspect of the invention can be further developed as follows, in particular after an automatic display of the patient's current patent information by means of a user interface of the diagnostic station: - automatic determination of patient identification data of a patient from the displayed current patient information by means of a first computer-implemented function executable in the computing unit or the diagnosis station, - automatic retrieval of stored historical patient information of the same patient and / or reference patients based on the determined patient identification data, - automatic application of a trained second computer-implemented function executable in the processing unit or the diagnosis station to the displayed current patient information and to the determined stored historical patient information of the same patient and / or reference patients, - automatically determining a relevance measure for the determined stored historical patient information of the same patient and / or reference patients based on the trained second function, and - automatic provision of selected stored historical patient information of the same patient and / or reference patients.

[0011] An aforementioned inventive method according to the first aspect of the invention is designed as a computer-implemented method, and it can be realized in conjunction with any individual feature from any further aspect of the invention, in conjunction with any further aspect of the invention, or in conjunction with any combination or sub-combination of further features and / or aspects of the invention as described in this document.

[0012] The patient information mentioned in this document as well as any further information mentioned in this document can, within the scope of any aspect and / or within the scope of any feature of the present invention, be designed as data or data sequences or data packets, designed and suitable for data processing with the aid of a data processing device and / or a computer-implemented method and / or a computer program product, and such data or data sequences or data packets can, in particular, represent and / or contain information relating to at least one patient and / or relating to the state of health of a patient and / or relating to medical evidence and / or relating to medical findings.

[0013] The aforementioned patient information may in particular include current and / or historical information from clinical diagnostic examinations and / or therapeutic treatments, for example medical evidence which contains image data or image information from imaging examinations.

[0014] The patient information of reference patients mentioned in this document can be embodied as patient information of one or more real patients within the scope of any aspect and / or within the scope of any feature of the present invention. However, it can also be embodied as information of a generalized virtual or synthetic reference patient, created from a representative patient collective, for example in the form of averaging and / or by correlating with weightings and / or with a weighting function and / or with a frequency distribution or other relative weighting of information of individual patients in a patient collective, including the threshold values ​​of a criticality assessment extracted for this purpose.

[0015] The at least one diagnosis station for creating a medical diagnosis can be designed or further developed within the scope of any aspect and / or within the scope of any feature of the present invention, in particular for the integration and / or presentation and / or analysis of the aforementioned patient information - in particular current patient information and / or historical patient information.

[0016] The inventive aspects and / or features of the invention can be designed in particular within the scope of each aspect and / or within the scope of each feature of the present invention as - Method comprising one or more of the steps according to the invention in any combination or sub-combination of further features and / or aspects of the invention as described in this document, and / or - as a system or as data processing units designed to carry out a method according to the invention and / or - as a data processing program product or computer program product or as a plurality of data processing program products or computer program products or as a data processing program product system consisting of a plurality of data processing program product modules or computer program product system consisting of a plurality of computer program product modules, each with individual program steps, each individually or jointly designed to carry out corresponding steps of a method according to the invention, in particular in the form of one or more correspondingly designed and adapted computer program products which can be executed in data processing units in a medical information network.

[0017] According to a second aspect of the invention, a method according to the invention according to the previously described first aspect of the invention - and / or according to any further aspect of the present invention - can be further developed such that the step of automatically applying a trained function, in particular applying it to displayed current patient information and to determined stored historical patient information - in particular to patient information of the same patient and / or of reference patients - comprises the following steps: - automatic recording of user activity data relating to the activity of a user in relation to the current patient information displayed by means of the user interface by means of a recording device which is in data connection with the diagnosis station and with the computing unit, and - automatically applying the trained - in particular second - function to the displayed current patient information and to the determined stored historical patient information - in particular to patient information of the same patient and / or of reference patients - depending on - in particular in correlation with - the recorded user activity data.

[0018] A correlation of displayed or historical patient information with recorded user activity data can, within the scope of each aspect and / or within the scope of each feature of the present invention, particularly include a comparison of spatial and / or temporal parameter data of the display of patient information with respect to the user interface with spatial and / or temporal parameter data of the recorded user activity with respect to the user interface. This applies to the aforementioned second aspect of the invention and accordingly also to each further aspect of the invention.

[0019] This second aspect, as well as any further aspect of the present invention, can be designed, for example, to automatically capture user activity data relating to a user's activity during the viewing and / or processing and / or diagnosis of current evidence (such as the viewing and / or processing and / or diagnosis of current image information, alternatively or additionally of other components of the findings presentation, such as graphics, reports, tables).

[0020] In this case, within the scope of each aspect of the present invention, a detection device can be used to automatically track the region of particular attention of the user and / or the specific processing activity and / or processing region of the user and / or the user's focus of interest with regard to the patient information displayed by means of the user interface.

[0021] For example, within the scope of each aspect of the present invention, the location and / or movement and / or dwell time of an input element or operating element displayed on the user interface—such as a cursor or mouse pointer—can be detected by means of a detection device, and / or a body movement of the user can be detected by means of a detection device, such as eye tracking, gesture recognition, or touch detection. The detection device can be implemented as a computer program function (software) or as a physical device (hardware), or as a combination of both.

[0022] Based on the automatically recorded user activity data relating to a user's activity, within the scope of each aspect of the present invention, an anatomical position can be automatically localized in the current image information, for which a particular focus of interest can be recognized from the recorded user activity data, e.g. with the supplementary application of other standard medical technologies such as ALPHA technology, atlas mapping, referencing to standard schemas such as RadLex. This makes it possible, in particular based on image processing or analysis methods, to extract the anatomical context from the regions that the user is viewing and / or editing. If, for example, the user is obviously concerned with suspicious areas in the liver, the system recognizes this from the viewer's focus and / or the mouse position and / or the layout / report template.can derive at least an indication of the user's area of ​​interest from this. The system or computer-implemented method according to the invention recognizes or derives the location of the relevant region and can, for example, extract the relevant information from the patient's history and, if necessary, display it in a supplementary manner.

[0023] A further aspect of the invention relates to a computer-implemented method according to the invention for controlling units in a medical information network - in particular a method which can include one or more of the features according to the first or second aspect of the invention described above - wherein the computer-implemented method includes the following steps: - automatic display of current patient information via a user interface of a diagnostic station, - automatic recording of user activity data relating to the activity of a user in relation to the current patient information displayed by means of the user interface by means of a recording device which is in data connection with the diagnosis station and with a computing unit, and - Automatic application of a trained function - in particular a second, computer-implemented function - to the displayed current patient information and to stored historical patient information - in particular patient information of the same patient and / or of reference patients - determined in particular by means of a first, computer-implemented function, depending on - in particular in correlation with - the recorded user activity data for the purpose of automatically providing selected stored historical patient information - in particular patient information of the same patient and / or of reference patients.

[0024] One or more of the features or sub-combinations of the features of the first and second aspects of the present invention can be used for a practical embodiment as well as for a corresponding further development of this further aspect of the present invention.

[0025] The aforementioned patient information of reference patients can be determined within the scope of each aspect of the present invention, for example, using a method as described in the aforementioned document DE 10 2022 201 630 A1. In particular, a correlation of one or more medical parameters and / or derived variables of the patient with one or more corresponding medical parameters and / or derived variables of a reference patient can be carried out. A correlation can in particular comprise a comparison and / or a quantification of a deviation of one or more medical parameters and / or derived variables of the patient with respect to one or more corresponding medical parameters and / or derived variables of a reference patient.It may be provided that, based on the patient information of the patient in question, an automatic processing is used to search for similar patients for whom, in particular, one or more corresponding medical parameters and / or derived variables and / or corresponding health information are (pre-)known and - optionally - already verified.

[0026] The aforementioned inventive method according to the above-described further aspect of the invention is also designed as a computer-implemented method, and it can also be implemented in conjunction with each individual feature from each further aspect of the invention, in conjunction with each further aspect of the invention, or in conjunction with any combination or sub-combination of further features and / or aspects of the invention as described in this document.

[0027] This further aspect of the present invention can also be designed, for example, in such a way that user activity data relating to the activity of a user is automatically recorded during the viewing and / or processing and / or diagnosis of current evidence (such as the viewing and / or processing and / or diagnosis of current image information, alternatively or additionally of other components of the findings presentation, such as graphics, reports, tables), in particular according to one or more of the features, aspects, developments and technical details as already described above for the first and second aspects of the present invention.

[0028] Within the scope of the aspects of the invention described above, a further aspect of the invention may provide for the continuous and / or iterative automatic recording of user activity data and / or the continuous and / or iterative automatic application of the trained second function. In particular, a further aspect of the invention may include a computer-implemented method—in particular according to one or more of the above aspects of the invention or according to any subcombination of features of one or more of the above aspects of the invention—including the following steps: - automatically collecting second and / or further user activity data relating to a user's activity in relation to the current patient information displayed by the user interface and / or the selected stored historical patient information by means of the collecting device, - automatically - in particular repeatedly or multiple times - applying the trained second function to the displayed current patient information and to the determined stored historical patient information depending on - in particular in correlation with - the recorded second and / or further user activity data for adapting and / or verifying a determined relevance measure for the determined stored historical patient information, and - automatic selection and / or automatic provision of - in particular additional - selected stored historical patient information by means of the user interface of the reporting station depending on the adapted and / or verified determined relevance measure.

[0029] By means of the further aspect of the invention described above, in particular, a continuous and / or iterative refinement of provided historical patient information could be carried out. For this purpose, the user's activity could be continuously recorded, specifically the user's activity with respect to the current patient information and / or with respect to selected or provided stored historical patient information. On this basis, in particular, selected or provided stored historical patient information can be further filtered and / or optimized and / or verified, and additional relevant data—in particular, additional selected or stored historical patient information—can be provided if necessary.

[0030] According to a further aspect of the invention, a method according to the invention can be developed according to one of the previously described aspects of the invention - or according to any combination or sub-combination of features of the previously described aspects of the invention - such that - an automatic recording of user activity data regarding the activity of a user during the viewing and / or processing of displayed current patient information - in particular of displayed current evidence information - takes place, whereby - the detection device automatically detects the region and / or intensity of the viewing activity and / or the processing activity of the user in relation to the patient information displayed by means of the user interface - in particular in relation to the current evidence information displayed.

[0031] In particular, in the aforementioned further aspect of the invention, the detection device can be used to automatically detect the region and / or the activity and / or the intensity of the activity of an input and / or control element interacting with the user and / or with the user interface - in particular an input and / or control element displayed on the user interface - and controllable by the user - in particular with regard to displayed current evidence information.

[0032] This means that a recording device can be used to automatically track the region of particular attention of the user and / or the specific processing activity and / or processing region of the user and / or the focus of interest of the user in relation to the patient information presented by means of the user interface - in particular in relation to current evidence information presented.

[0033] For example, the detection device can be used to detect the location and / or movement and / or duration of an input and / or control element displayed on the user interface—such as a cursor or mouse pointer—and / or a detection device can be used to detect a user's body movement, such as eye tracking, gesture recognition, or touch detection. The detection device can be implemented as a computer program function (software) or as a physical device (hardware), or as a combination of both.

[0034] Based on the automatically collected user activity data regarding a user's activity, an anatomical position can be automatically located in the current image information for which a particular focus of interest can be identified from the collected user activity data (e.g., by supplementing the application of other standard medical technologies such as ALPHA technology, atlas mapping, referencing to standard schemas such as RadLex).

[0035] The aforementioned patient information within the meaning of one or more of the aforementioned aspects of the invention may in particular include the following: - Patient identification data to identify and / or mark specific patients, and - Evidence information regarding current and / or historical medical evidence, and / or - Findings information regarding current and / or historical medical findings.

[0036] The evidence information may include medical image information and / or medical document information and / or medical text information and / or medical parameters that include, depict, describe or otherwise represent one or more numerical or measured values ​​and / or a temporal sequence of a numerical or measured value.

[0037] Medical image information can include computer-generated image data representations of body regions, organs, cross-sections of organs, and / or visualizations of medical parameters. Medical document information and / or medical text information can include computer-generated representations of documents or text information concerning body regions, organs, and / or representations of medical parameters.

[0038] A medical parameter can, for example, be a blood value, a urine value, a pulse rate, blood oxygen saturation, sleep behavior, a tumor marker, weight, height, age, gender, or the like. However, it is equally conceivable that the medical parameter comprises any measured value from a diagnostic medical device. For example, the medical parameter can be cardiac activity from an electrocardiogram, a measured value from a molecular diagnostic analysis, or even the result of a histological examination. Furthermore, the term "medical parameter" also includes information about the patient's internal and external structures, which can be derived, for example, from imaging procedures such as magnetic resonance imaging, computed tomography, various X-ray procedures, ultrasound, positron emission tomography, or the like.For example, medical parameters can include organ volumes, such as heart or lung volume. Furthermore, medical parameters can include a location, number, and / or size of a change, particularly a pathological change, in the patient's body. These pathological changes can include, for example, lesions or nodules in tissue areas of the patient. Furthermore, the medical parameter can include one or more pieces of semantic information. In particular, medical parameters can include one or more words.

[0039] The findings information can represent a selection of medical parameters that characterize a patient's physical condition. A selection of medical parameters can, for example, represent a parameter configuration or a pattern of a patient's medical parameters. The health information can further comprise a parameter configuration that indicates a change in a patient's health status. It is also conceivable for the health information to indicate a probability of an illness and / or a, in particular critical, change in a patient's health status. The health information can, in particular, comprise a summarized or aggregated representation of selected medical parameters that contribute to a change in the patient's physical condition.It is also conceivable that the health information includes one or more key figures derived from one or more medical parameters and quantifying a change in a patient's physical condition.

[0040] Each further aspect of the invention mentioned above and each further aspect mentioned below and / or each further feature or features of the invention can also be designed in particular as - Method comprising one or more of the steps according to the invention in any combination or sub-combination of further features and / or aspects of the invention as described in this document, and / or - as a system or as data processing units designed to carry out a method according to the invention, and / or - as one computer program product or several computer program products with individual program steps, each individually or jointly designed to carry out corresponding steps of a method according to the invention, in particular in the form of one or more correspondingly designed and adapted computer program products which can be executed in data processing units in a medical information network.

[0041] In each of the aforementioned and below-mentioned aspects of the invention and in each of the aforementioned embodiments of the invention, the aforementioned steps of the inventive method can be automated, in particular computer-implemented, in particular in appropriately designed data processing units. This can mean that the inventive method is coordinated and executed by means of a computing unit—which can be designed as a central computing unit or as a distributed computing unit—for example, by means of a local computer, a network computer, a server, a cloud, or a comparable component. The execution of the method can be initiated, for example, as a result of manual activation by a user. Alternatively, the method can be started automatically, e.g., as a result of receiving patient information.

[0042] The medical network or medical information network can be configured to exchange medical information, in particular patient information and / or health information and the like. Furthermore, the medical information network can be configured to exchange control commands between the connected components. The medical network can, in particular, establish a communication and / or data connection between one or more diagnostic stations and a computing unit—which can be configured as a central computing unit or as a distributed computing unit. In addition, the medical network can establish a communication and / or data connection with other medical data processing devices, such as storage devices for storing patient information. The medical network can comprise an intranet and / or the internet.In other words, a diagnosis station or a plurality of diagnosis stations can be in communication and / or data connection with a - central or distributed - computing unit via the Internet.

[0043] In particular, communication or data connections can be based on the HL7 standard. Health Level 7 (HL7) is a group of international standards for the exchange of data between healthcare organizations and their computer systems. In particular, communication and / or data connections can be based on the FHIR standard. Fast Healthcare Interoperability Resources (FHIR) is a standard developed by HL7. It supports the exchange of data between healthcare software systems. By using the HL7 or FHIR standard, data can be transferred in a structured manner, eliminating the need for reformatting. The at least one diagnostic station can be configured, in particular, as a front-end computing device at which a user (in particular medical personnel such as a doctor) can access, view, and / or analyze patient information, and / or at which the user can access, view, and / or modify evidence in the form of medical reports, and at which medical reports can be created for patients. The diagnostic station can have a user interface for this purpose. The diagnostic station can be configured as a so-called client.

[0044] The computing unit can be configured either as a central or distributed back-end computing device, and in particular as a server system, and / or it can be located entirely or partially in the diagnostic station. The computing unit can comprise a cluster or group of computing devices and data storage units. The computing unit can be connected to the front-end computing device via the medical network. The computing unit can be connected to several different (but particularly similar) diagnostic stations via the medical network. The diagnostic station(s) can belong to a medical organization, such as a practice, a hospital, or a hospital network. The computing unit can also belong to the medical organization or be located outside the medical organization.The computing unit can be connected to several different diagnostic stations via the medical network, each of which belongs to different medical organizations.

[0045] Within the scope of the inventive methods, in the aspects of the invention described above and / or in further aspects of the invention described in this document, a trained, computer-implemented function executable in the computing unit or the diagnostic station is particularly automatically applied. Any further function described within the scope of the present invention can, in particular, also be embodied as such or as a comparable, trained function, but can also be embodied as another data processing function suitable and adapted for the respective data processing step.

[0046] A trained function generally maps input data to output data. The output data may, in particular, depend on one or more parameters of the trained function. The one or more parameters of the trained function may be determined and / or adjusted through training. The determination and / or adjustment of the one or more parameters of the trained function may, in particular, be based on a pair of training input data and associated training output data, wherein the trained function is applied to the training input data to generate training mapping data.

[0047] In particular, it can be provided that, within the scope of one or more aspects of the present invention, automatic, particularly computer-implemented, feedback loops are provided for modifying the training basis. This means that the training basis can be modified via feedback loops—i.e., it can improve over time—because feedback on the patient's health status allows the underlying model to be adapted during further development, and a better set of training data can be used for further evaluations of new patients.

[0048] In particular, the determination and / or adaptation can be based on a comparison of the training image data and the training output data. For the purposes of this invention, a trainable function, i.e., a function with parameters that have not yet been adapted, is also understood as a trained function. In particular, the trained function can be contained in a single filter component of the data filter. In addition, the data filter can have further filter components that do not include trained functions because they operate, for example, in a rule-based manner. Furthermore, the data filter can also have multiple trained functions.

[0049] A trained function within the meaning of this invention can be embodied as a trained mapping rule, a mapping rule with trained parameters, a function with trained parameters, an algorithm based on artificial intelligence, or a machine learning algorithm. An example of a trained function is an artificial neural network. Parameters within the meaning of the invention can, in particular, be weight parameters of neurons of an input layer and / or an output layer and / or of one or more hidden layers of such an artificial neural network.

[0050] For possibilities of designing such trained functions, in particular of appropriately trained neural networks, the person skilled in the art will refer to the technical teaching of the earlier patent application of the patent applicant DE 10 2022 201 630 A1 already mentioned at the beginning as well as to the document: Miotto, R., Li, L., Kidd, B. et al. Deep Patient: An Unsupervised Representation to Predict the Future of Patients from the Electronic Health Records. Sci Rep 6, 26094 (2016), and these two documents—singly or in combination—are incorporated into the disclosure of the present invention in this regard.

[0051] Values ​​described within the scope of the invention for a relevance measure, a probability measure and / or a probability score and / or for threshold values ​​of functions can be designed as error values ​​or cost values ​​of one or more error functions or cost functions of an aforementioned neural network, or as values ​​derived from such error values ​​or cost values.

[0052] A method according to one of the aforementioned aspects of the invention, as well as according to a further aspect as described in this document, can in particular enable the integration of multidisciplinary previous (i.e., historical in the sense of the invention) analytical, medical and / or diagnostic information and the automatic evaluation of the relevance of such information. A method according to one of the aforementioned aspects of the invention, as well as according to a further aspect as described in this document, can alternatively or additionally support, in particular, the diagnosis of current evidence, such as the diagnosis of current image examinations based on medical imaging methods, in particular through context-specific integration of relevant previous (i.e., historical in the sense of the invention) evidence of the same patient and / or of reference patients, such asprevious imaging information and clinical patient information in order to increase diagnostic certainty and to assist in the creation of reports and the subsequent quantification of findings. Until now, such previous information was "hidden" in a large number of unstructured documents, although it was sometimes also available in structured documents or accessible via program interfaces or programming interfaces APIs (Application Program Interfaces) of information servers. However, previously known systems according to the state of the art are in particular not capable of extracting relevant subsets of information and, in particular, integrating them into a user interface in a context-sensitive manner or making them available to a user via a user interface. For example, Large Language Models (LLMs) represent an exemplary possibility for structuring unstructured documents orto restructure them to make them more accessible.

[0053] In this case, a method according to the invention according to one of the aforementioned aspects of the invention can be based in particular on a computer-implemented workflow which automatically initiates and / or controls individual or all steps of the method according to the invention according to the first aspect and / or one or more further aspects of the present invention.

[0054] Such a workflow can, in particular, begin with the processing of patient information from a current imaging examination of a specific patient for whom previous (=historical) medical patient information is available in electronic form. This can be information from previous imaging examinations of the same and / or other modality and / or structured and / or unstructured evidence (e.g., medical reports) from any type of medical examination and / or medical discipline. Alternatively or additionally, this can be evidence and / or reported results provided by the patient, which the patient can make available via an electronic data interface - for example, using a data application (e.g., Noona) on a computer, notebook, or smartphone.For example, the patient can report the occurrence of symptoms, keep the treatment team informed about their general health status, or provide their own patient information when asked about their condition between individual examination or treatment appointments.

[0055] According to a further aspect of the invention, a method according to the invention can be developed according to one of the previously described aspects of the invention such that - the current patient information presented contains current evidence information of a patient regarding current medical evidence, - the selected stored historical patient information includes stored historical evidence information – in particular evidence information of the same patient and / or reference patients – relating to historical medical evidence, and - the stored historical evidence information is stored in the form of structured evidence information and / or unstructured evidence information.

[0056] The present invention can therefore, in particular, also take into account, within the context of the patient information, evidence information of the same patient and / or reference patients, such as previous medical findings of the same patient and / or reference patients, which are contained in the stored patient information - such as digital medical records of the patient in question - and which could be of importance for a finding (in particular diagnosis and / or reporting) of potential findings in the current area of ​​interest (in other words: automatic execution of an evidence query). In particular, it can be provided that on the basis of certain predefined criteria and / or criteria determined with the aid of computer-implemented methods (e.g. on the basis of known iestored historical diagnostic information, predefined computer-based workflow steps in patient care, current user role of the user currently active at the reporting station and alternatively or additionally information and / or data obtained from deep learning methods) historical evidence information - in particular evidence information of the same patient and / or reference patients - regarding historical medical evidence - in particular regarding evidence of the same patient and / or reference patients - can be selected.

[0057] According to the invention, this evidence query can be applied to both structured evidence information (e.g., previous DICOM SR or FHIR observations according to defined ontologies) and unstructured evidence information (e.g., narrative reports, video patient consultations). Automatic methods for the computer-implemented retrieval and / or processing of structured evidence information and / or unstructured evidence information are generally known from the state of the art. For example, a preprocessing step of structured evidence information and / or unstructured evidence information can be performed using natural language processing (NLP) techniques, in particular using unsupervised deep learning techniques, i.e., machine learning without pre-known target values, as described, for example, in the following document: Miotto, R., Li, L., Kidd, B. et al. Deep Patient: An Unsupervised Representation to Predict the Future of Patients from the Electronic Health Records. Sci Rep 6, 26094 (2016).

[0058] The technical teaching of the aforementioned document is hereby incorporated into the disclosure of the present invention. In particular, those individual technical aspects relating to natural language processing (NLP) techniques, particularly unsupervised deep learning techniques, are incorporated into the disclosure of the present invention that a person skilled in the art would consider helpful or supportive for further development or implementation of the present invention.

[0059] According to a further aspect of the invention, a method according to one of the previously described aspects of the invention can be further developed such that the steps of displaying the current patient information and automatically providing the selected stored historical patient information comprise the following: - Displaying the patient's current patent information in a first display area of ​​a user interface of the diagnostic station, and - Automatic provision of the selected stored historical patient information - in particular patient information of the same patient and / or of reference patients - in a second display area of ​​the user interface of the diagnostic station depending on the determined relevance level for this stored historical patient information.

[0060] With this development according to the invention, it can be achieved, in particular, that a reporting station in a medical information network automatically provides a user (e.g., a physician) with information about additional evidence—in particular concerning the same patient and / or reference patients—relevant for the interpretation of a current focus of interest via a user interface (e.g., a graphical user interface). The results of the evidence query are presented in the user interface while the user navigates, for example, in a computer-implemented viewing interface for displaying patient information, such as current evidence. Without any interaction, the user can evaluate relevant evidence directly in the user interface (and access more detailed information if necessary).The selected historical evidence information - in particular evidence information of the same patient and / or reference patients - is preferably presented in such a way that it does not hinder the viewing of the current patient information, such as current evidence information, so that the user can ignore the additionally displayed selected historical evidence information and view the current patient information - such as current evidence information of a patient - without interference.

[0061] According to a further aspect of the invention, a method according to the invention according to one of the previously described aspects of the invention can be further developed such that it further comprises the following steps: - automatically applying a trained third, computer-implemented function executable in the processing unit or the diagnosis station to the current patient information and to the provided selected stored historical patient information - in particular patient information of the same patient and / or reference patients, - automatically determining current and / or historical finding information and associated finding probability score information with respect to the current patient information based on the trained third function, and - Automatic provision of current and / or historical findings information via the user interface of the reporting station depending on the associated findings probability score information for this current findings information.

[0062] In this regard, particular reference is made to the information and technical facts presented at the beginning regarding trained functions, in particular neural networks as a special embodiment of trained functions.

[0063] With this development according to the invention, it can be achieved, in particular, that information regarding possible candidates for findings is automatically determined from the relevant patient information and such information regarding possible candidates for findings is automatically added to the existing patient information. In particular, computer-implemented preprocessing steps can be automatically performed with respect to patient information, in particular automatic computer-implemented image processing steps and / or automatic computer-implemented object recognition steps and / or automatic computer-assisted diagnostic steps (CAD steps). For example, certain conspicuous image regions in imaging evidence can be automatically identified, marked, or highlighted, and suggested findings can be automatically offered.

[0064] Based on the selected stored historical patient information provided, such as historical evidence information and / or historical findings information—in particular, findings information from the same patient and / or from reference patients—(e.g., previous illnesses, laboratory markers, symptoms), a probability score for the detected CAD findings can be calculated and / or refined to increase specificity and sensitivity and to highlight certain findings. In particular, the method according to the invention can be further developed to provide the user with additional explanatory information for the automatically generated findings suggestions, e.g.,by automatically highlighting and / or displaying or moving an optical marker (such as a PIN, a cursor or a mouse pointer) at the display location of the corresponding automatically generated suggested findings in displayed imaging evidence.

[0065] According to a further aspect of the invention, a method according to the invention according to one of the previously described aspects of the invention can be further developed such that it further comprises the following steps: - automatic linking of current findings information and stored historical findings information based on - in particular in correlation with - user activity data, and - Automatic application of a fourth, computer-implemented function, which can be executed in the processing unit or the reporting station, to the current findings information and stored historical findings information - preferably findings information of the same patient - to determine quantitative and / or time-dependent diagnostic evaluation information, preferably for the same patient.

[0066] With this development according to the invention, it can be achieved in particular that, in the event that the user considers historical findings information to be relevant for current findings information, for example as explanations and / or supplementary information for the creation of a current medical report for the corresponding patient, the user can generate corresponding user activity data through corresponding activities on the user interface of the diagnosis station (for example text input, voice input, mouse clicks, gestures, or the like), which are then automatically recorded. In a further computer-implemented step, this activity data can be recorded, and on the basis of this data, the relevant historical findings information can be automatically added to the current patient information, in particular added to a current medical report.

[0067] With this further development according to the invention, it is also possible, in particular, to determine quantitative and / or time-dependent finding evaluation information based on the current finding information and relevant historical finding information—in particular based on finding information from the same patient and / or from reference patients. For example, the method or a system according to the invention can cause historical previous finding quantifications to be transferred to a current medical study or a current medical report and recalculated in order to document a temporal progression and / or changes in a temporal progression.

[0068] According to a further aspect of the invention, a method according to the invention according to one of the previously described aspects of the invention can be further developed such that it further comprises the following step: - automatically training and / or optimizing parameters of the trained second function and / or the trained third function based on current finding information and associated finding probability score information and / or based on selected stored historical patient information and associated relevance measure information and / or user activity data.

[0069] With this development according to the invention, it can be achieved, in particular, that the additional collected findings can be used, based on current historical information as well as on information from user activity data, which provides indications of feedback, evaluations, and / or confirmations from the user during the current processing process, to optimize and retrain at least one trained function—for example, a neural network—in order to improve the performance of the trained function. For example, an "ROC performance" (ROC = Receiver Operating Characteristic) can be improved, which serves to measure the performance of a machine learning method or a forecasting method that classifies objects into at least two classes, A and B.

[0070] According to a further aspect of the invention, a method according to one of the previously described aspects of the invention can be further developed such that the step of automatically providing the selected stored historical patient information comprises one of the following steps or both of the following sub-steps: - automatic provision via the user interface of selected stored historical patient information whose determined relevance measure exceeds a predefined threshold towards high relevance, in conjunction with a high relevance label and / or high relevance highlighting and / or high relevance positioning, and / or - automatic provision via the user interface of selected stored historical patient information whose determined relevance measure falls below a predefined threshold towards low relevance, in conjunction with a low relevance label and / or low relevance highlighting and / or low relevance positioning.

[0071] In particular, it can be provided that earlier findings, i.e. historical patient information, which cannot be directly linked to a specific focus of interest, i.e. whose relevance for the current diagnosis of the patient in question is lower, are nevertheless made available or presented to the user. This earlier knowledge, i.e. historical patient information, which cannot be directly linked to a specific focus of interest, can, for example, be presented in a more general form and / or labeled and / or positioned - temporally and / or spatially and / or data-wise and / or in tabular form - in order to inform the user about a general health condition and / or about a medical history and / or about medical evidence and / or findings from other medical disciplines.This can be done, for example, to inform the user that there is an increased likelihood of metastases in certain body regions based on previous diseases and laboratory markers / genetic risk factors, whereby this information is presented with a special label and / or a special highlight and / or in a special position in an automatic computer-implemented processing workflow, for example by positioning this information - temporally and / or spatially and / or data-wise and / or in a table - at the beginning (at least in the first half, preferably in the first quarter) of a processing workflow for the diagnosis of a patient, and thus additional historical patient information for the diagnosis of a patient is made available or presented to the user at the beginning of an automatic computer-implemented processing workflow.

[0072] According to a further aspect of the invention, a method according to the invention according to one of the previously described aspects of the invention can be further developed such that it further comprises the following steps: - automatically marking the displayed current evidence information and / or the selected historical evidence information using current evidence relevance information determined on the basis of user activity data; and - automatically optimizing stored historical evidence information and / or stored historical evidence relevance information based on the determined evidence relevance information.

[0073] In particular, it can be provided that the current evidence information and / or selected historical evidence information displayed are automatically marked as relevant by the user through active or implicit interaction with the user interface of the reporting station (in particular explicitly through active marking by the user, e.g. "thumbs up" and / or highlighting and / or spatial and / or data-related and / or tabular arrangement, whereupon corresponding marking data are determined as user activity data and evidence relevance information can be determined from this, and / or implicitly, when the user transfers evidence information into a current medical finding or report and then evidence relevance information is automatically determined from such a correlation orAssignment of evidence information to findings information correlated with or derived from corresponding user activity data).

[0074] Accordingly, it can also be provided that the displayed current evidence information and / or selected historical evidence information is automatically marked as irrelevant by the user through active or implicit interaction with the user interface of the reporting station (in particular explicitly through active marking by the user, e.g. thumbs down and / or highlights and / or spatial and / or data-related and / or tabular arrangement, and / or implicitly if the user does not transfer evidence information into a current medical finding or report and then automatically determines evidence-relevance information via a computer-implemented method step). This information or markings can in particular be used to provide the system orto provide feedback to the computer-implemented procedure so that stored historical evidence information and / or stored historical evidence relevance information can be optimized, for example so that corresponding evidence identifications in a backend can be improved through retraining.

[0075] As already stated at the outset, the aspects and / or features of the invention can be designed in particular as - Method comprising one or more of the steps according to the invention in any combination or sub-combination of further features and / or aspects of the invention as described in this document, and / or - as a system or as a data processing unit designed to carry out a method according to the invention and / or - as a data processing program product or computer program product or as a plurality of data processing program products or computer program products or as a data processing program product system consisting of a plurality of data processing program product modules or computer program product system consisting of a plurality of computer program product modules, each with individual program steps, each individually or jointly designed to carry out corresponding steps of a method according to the invention, in particular in the form of one or more correspondingly designed and adapted computer program products which can be executed in data processing units in a medical information network.

[0076] A further aspect of the present invention is therefore a data processing unit or a system, in particular designed to carry out a method according to one or more of the previously described aspects, wherein the data processing unit or the system has a computing unit and wherein the computing unit is in data connection with at least the following units: - at least one diagnosis station trained to display and / or provide patent information and to create a medical report, - at least one data acquisition unit trained to collect current patient information, and - at least one storage unit designed to store historical patient information, wherein the computing unit is configured to execute computer-implemented program codes and to generate control commands to carry out the following steps: - automatically applying executable, computer-implemented program codes representing a trained function to displayed current patient information and to stored historical patient information - in particular to patient information of the same patient and / or reference patients -, - automatically determining a relevance measure for the determined stored historical patient information - in particular for patient information of the same patient and / or reference patients - based on initial values ​​of the applied trained function, and - Generation of control commands for the automatic provision of selected stored historical patient information - in particular patient information of the same patient and / or of reference patients - by means of a user interface of the diagnostic station depending on the determined degree of relevance for this stored historical patient information. - In particular, in the aforementioned aspect of the invention and in any further aspect of the invention described in this document, it can be provided alternatively or additionally that the computing unit is additionally configured or further developed to execute computer-implemented program codes and to generate control commands to carry out the following further or further developed steps according to the aforementioned aspect of the invention: automatically determining patient identification data of a patient from displayed current patient information by means of computer-implemented program codes that can be executed in the computing unit and that represent a first function; - automatically applying executable, computer-implemented program codes representing a trained second function to displayed current patient information and to stored historical patient information of the same patient and / or reference patients, - automatically determining a relevance measure for the determined stored historical patient information of the same patient and / or reference patients based on initial values ​​of the applied trained second function, - Generation of control commands for the automatic provision of selected stored historical patient information of the same patient and / or reference patients by means of a user interface of the diagnostic station depending on the determined relevance level for this stored historical patient information.

[0077] A further aspect of the present invention is a data processing unit or a system, in particular a data processing unit or a system according to the previous aspect of the invention, in particular designed to carry out a method according to one or more of the previously described aspects, wherein the data processing unit or the system comprises the following: a computing unit, and at least the following units connected to the computing unit: - at least one diagnosis station with at least one user interface, designed to display and / or provide current and historical patent information and to create a medical report, - at least one data acquisition unit designed to collect current patient information; - at least one storage unit designed to store historical patient information, and - at least one recording device designed to automatically record user activity data relating to the activity of a user in relation to the current patient information displayed by means of the user interface, wherein the computing unit is configured - to execute computer-implemented program codes representing a trained function on current patient information and on historical patient information - in particular on patient information of the same patient and / or of reference patients - depending on - in particular in correlation with - recorded user activity data, and - for generating control commands for the automatic provision of selected stored historical patient information - in particular patient information of the same patient and / or of reference patients - by means of a user interface of the diagnostic station depending on initial values ​​of the trained function applied.

[0078] A further aspect of the present invention is a data processing program product or data processing program system comprising a plurality of data processing program product modules, comprising instructions which, when the data processing program product or the data processing program product modules are executed by a data processing unit or a system, cause this data processing unit or the system to carry out the steps of a method according to one or more of the previously described aspects, in particular in cooperation with a data processing unit or a system according to one or more of the previously described aspects, and in particular in the form of a correspondingly designed computer program product or a plurality of computer program product modules, comprising computer-implemented program codes,which can be executed in at least one computing unit of data processing units in a medical information network.,

[0079] A further aspect of the present invention is a computer program product or computer program product system consisting of a plurality of computer program product modules, comprising instructions which, when the computer program product or the computer program product modules are executed by at least one data processing unit or a system having at least one computing unit, cause the at least one data processing unit or the system to execute the following program steps, in particular to execute the steps of a method according to one or more of the previously described aspects and / or in particular in cooperation with a data processing unit or a system according to one or more of the previously described aspects: - automatic computer-implemented recording of a patient's current patient information, - automatic computer-implemented display of the patient's current patient information via a user interface of at least one diagnostic station, - automatic computer-implemented retrieval of stored historical patient information - in particular patient information of the same patient and / or reference patients -, - automatically applying a trained computer-implemented function to the displayed current patient information and to the determined stored historical patient information - in particular to patient information of the same patient and / or reference patients -, - automatic computer-implemented determination of a relevance measure for the determined stored historical patient information - in particular for patient information of the same patient and / or reference patients - based on initial values ​​of the applied trained function, and - automatic computer-implemented provision of selected stored historical patient information - in particular patient information of the same patient and / or of reference patients - by means of the user interface of the diagnostic station depending on the determined degree of relevance for this stored historical patient information.

[0080] In particular, in the aforementioned aspect of the invention and in any further aspect of the invention described in this document, it can be provided alternatively or additionally that the computer program product or computer program product system is additionally configured or further developed to carry out the following further or further developed program steps according to the aforementioned aspect of the invention, in particular after the program step of an automatic computer-implemented display of the current patient information of the patient by means of a user interface of at least one diagnosis station: - automatic computer-implemented determination of patient identification data of a patient from the displayed current patient information by means of a first, computer-implemented function, - automatic computer-implemented retrieval of stored historical patient information of the same patient and / or reference patients based on determined patient identification data, - automatically applying a trained second computer-implemented function to the displayed current patient information and to the determined stored historical data of the same patient and / or reference patients, - automatic computer-implemented determination of a relevance measure for the determined stored historical patient information of the same patient and / or reference patients based on initial values ​​of the applied trained second function, and - automatic computer-implemented provision of selected stored historical patient information of the same patient and / or reference patients via the user interface of the diagnostic station depending on the determined relevance level for this stored historical patient information.

[0081] A further aspect of the present invention is a computer program product or computer program product system comprising a plurality of computer program product modules according to the above-described aspect of the invention, comprising instructions which, when the computer program product or the computer program product modules are executed by at least one data processing unit or by a system having at least one computing unit, cause the at least one data processing unit or the system to execute the following program steps: - automatic computer-implemented recording of user activity data relating to the activity of a user in relation to the current patient information presented by means of the user interface by means of at least one computer-implemented and / or hardware-implemented recording device; and - automatic computer-implemented application of the trained - in particular second - computer-implemented function to the displayed current patient information and to the determined stored historical patient information - in particular to patient information of the same patient and / or of reference patients - with computer-implemented correlation of the trained - in particular second - computer-implemented function with the recorded user activity data.

[0082] A further aspect of the present invention is a computer program product or computer program product system comprising a plurality of computer program product modules, designed to control units in a medical information network, comprising commands which, when the computer program product or the computer program product modules are executed by at least one data processing unit or a system having at least one computing unit, cause the at least one data processing unit or the system to execute the following program steps, in particular to execute the steps of a method according to one or more of the previously described aspects and / or in particular in cooperation with a data processing unit or a system according to one or more of the previously described aspects: - automatic computer-implemented display of current patient information via a user interface of a diagnostic station, - automatic computer-implemented recording of user activity data relating to the activity of a user in relation to the current patient information presented by means of the user interface by means of at least one computer-implemented and / or hardware-implemented recording device, and - automatically applying a computer-implemented trained - in particular second - function to the displayed current patient information and to stored historical patient information - in particular patient information of the same patient and / or of reference patients - determined in particular by means of a first, computer-implemented function - depending on - in particular in correlation with - the recorded user activity data for the purpose of automatically providing selected stored historical patient information - in particular patient information of the same patient and / or of reference patients.

[0083] A further aspect of the invention is a computer program product or a computer program product system comprising a plurality of computer program product modules, each according to one of the aspects of the invention described above and each designed to control units in a medical information network, comprising instructions which, when the computer program product or the computer program product modules are executed by at least one data processing unit with at least one computing unit, cause the at least one data processing unit or a system to execute the following program steps: - automatic computer-implemented recording of user activity data relating to the activity of a user during the viewing and / or processing of displayed current patient information, in particular displayed current evidence information; and - automatic computer-implemented detection of the region and / or intensity of the user's viewing activity and / or editing activity with regard to the patient information displayed via the user interface, in particular with regard to current evidence information displayed, - in particular automatic computer-implemented detection of the region and / or the activity and / or the intensity of the activity of an input and / or control element interacting with the user and / or with the user interface - in particular an input and / or control element displayed on the user interface - and controllable by the user - in particular with regard to displayed current evidence information.

[0084] A further aspect of the present invention is a computer-readable storage medium on which a data processing program product or data processing program system consisting of a plurality of data processing program product modules or a computer program product or a computer program product system consisting of a plurality of computer program product modules according to one of the aspects of the invention described above is stored or on which program sections that can be read and executed by a computing unit are stored in order to carry out steps of a computer-implemented method for controlling units in a medical information network according to one of the aspects of the invention described in this document when the program sections are executed by the computing unit.

[0085] A further aspect of the present invention is a computer-implemented method, system or data processing unit and computer program product for controlling units in a medical information network, comprising a computing unit, a diagnosis station, a data acquisition unit and a storage unit, wherein the computing unit is configured to automatically display current patient information by means of a user interface, to automatically apply a trained function to displayed current patient information and to stored historical patient information,and for automatically providing selected stored historical patient information depending on a determined relevance measure for this stored historical patient information and / or depending on recorded user activity data regarding a user's activity with respect to the displayed patient information. The properties, features, and advantages of this invention, as well as the manner in which they can be achieved, are explained below by way of example using specific embodiments described in conjunction with the drawings. These specific embodiments do not limit the invention to these embodiments. In different figures, identical components are provided with identical reference numerals. The figures are generally not to scale.

[0086] They show: Fig. 1 a schematic representation of a specific embodiment of an arrangement according to the invention in a medical information network; Fig. 2 a schematic representation of a possible special development of the embodiment according to Fig. 1; Fig. 3 a schematic representation of a possible special development of an embodiment according to Fig. 1 or after Fig. 2; Fig. 4 a schematic representation of a specific embodiment of a method according to the invention, in particular a computer-implemented method, in particular in the form of a computer program product; and Fig. 5 a schematic representation of a possible special development of an embodiment according to Fig. 4. Fig. 6 a schematic detailed representation of a possible special development of an embodiment according to Fig. 1 or after Fig. 2 or after Fig. 3;

[0087] The embodiment according to Fig. 1 schematically shows, by way of example, part of a medical information network comprising a system SYS for controlling data processing units in a medical information network. The illustrated system SYS can include the following units: a computing unit SYS.CU, and the following units connected to the computing unit SYS.CU: a data acquisition unit SYS.DCU for acquiring current patient information CPI, at least one data interface SYS.IF, via which patient information CPI can be fed to the data acquisition unit SYS.DCU, and at least one storage unit SYS.DB for storing patient information CPI, HPI.

[0088] The SYS.CU computing unit is connected to at least one AS reading station, which has at least one AS.UI user interface. Individual or all of the aforementioned units—i.e., the SYS.CU computing unit, the SYS.DCU data acquisition unit, and the SYS.IF data interface—can also be located directly in the AS reading station.

[0089] The patient information CPI and HPI stored in the SYS.DB storage unit can, in particular, represent historical patient information HPI, which can, in particular, include historical evidence information HEI, historical evidence relevance information HERI, and historical findings information HMRI. Furthermore, the historical patient information HPI contains patient identification data PID for the unique identification of the respective patient to whom the stored historical patient information HPI is assigned.

[0090] In the same storage unit SYS.DB or in a comparable additional storage unit SYS.DB, current patient information CPI can be permanently stored or temporarily stored. This can include, in particular, current evidence information CEI, current evidence relevance information CERI, and current findings information CMRI. Furthermore, the current patient information CPI also contains patient identification data PID for the unique identification of the respective patient to whom the current patient information CPI is assigned.

[0091] The computing unit SYS.CU can be designed as the central computing unit of the SYS system and / or a diagnosis station AS. However, the computing unit SYS.CU can also be designed as a sub-unit of a distributed computing unit of the SYS system and / or a diagnosis station AS. A first computer-implemented function F1 and / or a trained second computer-implemented function TF2 can be hosted in the computing unit SYS.CU. The first computer-implemented function F1 can, in principle, also be designed as a trained computer-implemented function, but it can also be designed as a simple search algorithm. For the specific design of trained computer-implemented functions, reference is made to the explanations already described at the beginning of this document.

[0092] The user interface AS.UI of the reporting station AS can have at least a first display area DA1, which is used to display current patient information CPI - such as in particular current evidence information CEI in conjunction with a unique identification of the corresponding patient using patient identification data PID. Depending on requirements or permanently, the user interface AS.UI of the reporting station AS can have at least a second display area DA2, which can be used temporarily or permanently to display historical patient information HPI, and whose size and / or position and / or display duration and / or display design in the user interface AS.UI can vary over time and / or location, in particular depending on the type, scope and significance of the historical patient information HPI to be displayed.

[0093] Furthermore, according to the embodiment according to Fig. 1, a recording device DU is provided, which is in data communication with the diagnosis station AS and with the computing unit SYS.CU. The recording device DU can be implemented as a computer program function (i.e., software-implemented) or as a physical device (i.e., hardware-implemented), or as a combination of both. The recording device DU is designed to automatically record user activity data AD regarding a user's activity in relation to the current patient information CPI displayed via the user interface AS.UI.For example, the detection device DU can be used to detect the location and / or movement and / or duration of an input element or control element displayed on the user interface - such as a cursor or mouse pointer - and / or the detection device DU can be used to detect a body movement of the user, such as eye tracking, gesture recognition or touch detection.

[0094] The schematic representation according to Fig. 4 shows a specific embodiment of a method according to the invention, in particular a computer-implemented method, in particular in the form of a computer program product, which can be used with an arrangement according to the embodiment according to the Fig. 1 to 3, and whose individual steps can in principle be combined in any suitable combination and / or any suitable order.

[0095] The specific embodiment of a method according to the invention according to Fig. 4 comprises a computer-implemented method for controlling data processing units in a medical information network, in particular as in Fig. 1 to 3 and includes in particular the following steps: - automatic recording according to step S1 of current patient information CPI of a patient by means of the data recording unit SYS.DCU; - automatic display according to step S2 of the current patient information CPI of the patient by means of a user interface AS.UI of the diagnosis station AS, wherein the step of displaying S2 of the current patient information CPI can in particular comprise: • Displaying the patient's current patent information CPI in a first display area DA1 of a user interface AS.UI of the diagnosis station AS), - optionally an automatic determination according to step S3 of patient identification data PID of a patient from the displayed current patient information CPI by means of a first computer-implemented function F1 executable in the computing unit SYS.CU or the diagnosis station AS, - automatic determination according to step S4 of stored historical patient information HPI of the same patient and / or of reference patients, in particular on the basis of the optionally determined patient identification data PID, - automatically applying, according to step S5, a trained second computer-implemented function TF, which can be executed in the computing unit SYS.CU or the diagnosis station AS, to the displayed current patient information CPI and to the determined stored historical patient information HPI of the same patient and / or of reference patients, wherein this step S5 can comprise the following two sub-steps: • automatic recording according to sub-step S5.1 of user activity data AD regarding the activity of a user in relation to the current patient information CPI displayed by means of the user interface AS.UI by means of a recording device DU, which is in data connection with the diagnosis station AS and with the computing unit SYS.CU; and • automatic application according to sub-step S5.2 of the trained second function TF2 to the displayed current patient information CPI and to the determined stored historical patient information HPI of the same patient and / or reference patients depending on - in particular in correlation with - the collected user activity data AD; - automatically determining, in accordance with step S6, a relevance measure for the determined stored historical patient information HPI of the same patient and / or reference patients on the basis of the trained second function TF2; and - automatic provision according to step S7 of selected stored historical patient information HPI of the same patient and / or of reference patients by means of the user interface AS.UI of the diagnosis station AS depending on the determined relevance measure for this stored historical patient information HPI, wherein this step S7 can in particular comprise the following step: • Automatic provision of the selected stored historical patient information HPI of the same patient and / or reference patients in a second display area DA2 of the user interface AS.UI of the diagnosis station AS depending on the determined relevance level for this stored historical patient information HPI, and wherein step S7 may alternatively or additionally comprise the following two sub-steps: • automatic provision according to sub-step S7.1 using the user interface AS.UI of selected stored historical patient information HPI, the determined relevance measure of which exceeds a predefined threshold value towards high relevance, in conjunction with a high relevance label and / or high relevance highlighting and / or high relevance positioning; and / or • automatic provision according to sub-step S7.2 by means of the user interface AS.UI of selected stored historical patient information HPI, the determined relevance measure of which falls below a predefined threshold value in the direction of low relevance, in conjunction with a low relevance label and / or low relevance highlighting and / or low relevance positioning.

[0096] Within the scope of the specific embodiment of a method according to the invention according to Fig. 4, it can be provided in particular or optionally as a further development of the specific embodiment of a method according to the invention that continuously and / or iteratively an automatic recording S5.1 of user activity data AD and / or continuously and / or iteratively an automatic application S5.2 of the trained second function TF2 takes place, as by the recursion loop around the steps S5.1 and S5.2 (or in total around the step S5) in Fig. 4 illustrates this as an optional process flow. A suitable termination condition can be defined for such a recursion loop, for example, as a predefined parameter or one learned using a trained function, or in the form of a suitable automatic comparison function.

[0097] In particular, such a development of the specific embodiment of a method according to the invention can be Fig. 4 the computer-implemented method includes the following steps: - automatic recording S5.1 of second and / or further user activity data AD relating to the activity of a user in relation to the current patient information CPI displayed by means of the user interface AS.UI and / or the selected stored historical patient information HPI by means of the recording device DU, - automatic - in particular repeated or multiple - application S5.2 of the trained second function TF2 to the displayed current patient information CPI and to the determined stored historical patient information HPI depending on - in particular in correlation with - the recorded second and / or further user activity data AD for adapting and / or verifying a determined relevance measure or an automatic determination S6 of a relevance measure for the determined stored historical patient information HPI, and - automatic selection and / or automatic provision S7 of - in particular additional - selected stored historical patient information HPI by means of the user interface AS.UI of the reporting station AS depending on the adapted and / or verified determined relevance measure.

[0098] Fig. 5 shows a schematic representation of a possible special development of an embodiment according to the invention according to Fig. 4, in particular a computer-implemented method, in particular in the form of a computer program product, which is provided with an arrangement according to the embodiment according to the Fig. 1 to 3 can be realized.

[0099] The special training according to Fig. 5 of the specific embodiment of a method according to the invention according to Fig. 4 comprises further possible steps for a computer-implemented method for controlling data processing units in a medical information network, in particular as in Fig. 1 to 3, which can be used individually or in any suitable combination and / or any suitable sequence with each other and / or with the specific embodiment of a method according to the invention according to Fig. 4 can be combined.

[0100] The special training according to Fig. 5 includes in particular at least one or several of the following steps: - automatic application S8 of a trained third computer-implemented function TF3, executable in the SYS.CU processing unit or the AS reporting station, to the current patient information CPI and to the provided selected stored historical patient information HPI of the same patient and / or reference patients, - automatic determination S9 of current finding information CMRI and of associated finding probability score information PSI with respect to the current patient information CPI based on the trained third function TF3, - automatic provision S10 of the current findings information CMRI using the user interface AS.UI of the reporting station AS depending on the associated findings probability score information PSI for this current findings information CMRI, - automatic linking S11 of current findings information CMRI and stored historical findings information HMRI based on user activity data AD, - automatic application S12 of a fourth, computer-implemented function F4, which can be executed in the SYS.CU processing unit or the AS reporting station, to the current CMRI findings information and stored historical HMRI findings information, preferably of the same patient, to determine quantitative and / or time-dependent MREI findings evaluation information, preferably of the same patient, - automatic training and / or optimization S13 of parameters of the trained second function TF2 and / or the trained third function TF3 on the basis of current findings information CMRI and associated findings probability score information PSI and / or on the basis of selected stored historical patient information HPI and associated relevance measure information and / or user activity data AD, - automatic labelling S14 of the displayed current evidence information CEI and / or the selected historical evidence information HEI using current evidence relevance information CERI, which is determined on the basis of user activity data AD, - Automatic optimization of S15 stored historical evidence information HEI and / or stored historical evidence relevance information HERI based on the determined evidence relevance information CERI.

[0101] Accordingly, Fig. 2 a schematic representation of one or more of the above-mentioned steps of further training according to Fig. 4 and Fig. 5 adapted arrangement of a possible special development of the embodiment according to Fig. 1 and Fig. 3 shows a schematic representation of one or more of the above-mentioned steps of a further training according to Fig. 4 and Fig. 5 adapted arrangement of a possible special development of an embodiment according to Fig. 1 or after Fig. 2.

[0102] This shows Fig. 2 in particular a schematic representation of a special development of the embodiment according to Fig. 1, which is adapted and configured such that a trained third, computer-implemented function TF3, which can be executed in the computing unit SYS.CU - or alternatively in the diagnosis station AS - is automatically applied to the current patient information CPI and to the provided selected stored historical patient information HPI of the same patient and / or reference patients, and an automatic determination of current diagnosis information CMRI and associated diagnosis probability score information PSI with respect to the current patient information CPI takes place on the basis of the third function TF3 trained in the computing unit SYS.CU - or alternatively in the diagnosis station AS. According to the schematic representation in Fig. 2, the special development of the embodiment is further adapted and designed such that the current finding information CMRI is then automatically provided by means of the user interface AS.UI of the diagnosis station AS depending on the associated finding probability score information PSI for this current finding information CMRI, wherein the associated finding probability score information PSI can also be provided to the user by means of the user interface AS.UI of the diagnosis station AS.

[0103] According to the schematic representation in Fig. 2, the special development of the exemplary embodiment is further adapted and designed such that - in particular by means of a (hardware-implemented and / or software-implemented) feedback loop data connection - automatic training and / or optimization of parameters of the trained second function TF2 and / or the trained third function TF3 can be carried out on the basis of current finding information CMRI and associated finding probability score information PSI and / or on the basis of selected stored historical patient information HPI and associated relevance measure information and / or user activity data AD.

[0104] A trained function TF2, TF3 within the meaning of this invention can be designed, in particular, as a trained mapping rule, a mapping rule with trained parameters, a function with trained parameters, an algorithm based on artificial intelligence, or a machine learning algorithm. An example of a trained function is an artificial neural network. Parameters within the meaning of the invention can, in particular, be weight parameters of neurons of an input layer and / or an output layer and / or of one or more hidden layers of such an artificial neural network. For further development of such trained functions, in particular of correspondingly designed neural networks, the person skilled in the art will refer to the technical teaching of the aforementioned earlier patent application of the patent applicant DE 10 2022 201 630 A1, as well as to the aforementioned document: Miotto, R., Li, L., Kidd, B. et al. Deep Patient: An Unsupervised Representation to Predict the Future of Patients from the Electronic Health Records. Sci Rep 6, 26094 (2016), and these two documents—singly or in combination—are hereby incorporated into the disclosure of the present invention.

[0105] A second trained function TF2, TF3 can, in one or each of the aforementioned embodiments, comprise, for example, a multi-layer neural network. For example, one or each trained function TF2, TF3 comprises a multi-layer neural network that is adapted based on a backpropagation method by means of the computing unit SYS.CU. Subsequently, a difference can be calculated between an actual output of one or each trained function TF2, TF3 and a target output of one or each trained function TF2, TF3 of the multi-layer neural network, which is considered an error. The error can then be backpropagated from an output layer to an input layer of the multi-layer neural network. In this case, a configuration of the multi-layer neural network, in particular a weighting of connections between neurons, can be changed depending on their influence on the error.Using appropriate techniques, the error between the target and actual output of the multilayer neural network for an input pattern can be minimized. One or each adapted trained function TF2, TF3 can be stored in a memory unit of the system SYS after adaptation.

[0106] Fig. 3 shows in particular a schematic representation of a special development of the embodiment according to Fig. 1 or Fig. 2, which is adapted and configured such that an automatic linking of current findings information CMRI and stored historical findings information HMRI takes place on the basis of user activity data AD, and an automatic application S12 of a fourth, computer-implemented function F4, executable in the computing unit SYS.CU - or alternatively in the reporting station AS - to the current findings information CMRI and stored historical findings information HMRI, preferably of the same patient, takes place, in order to determine quantitative and / or time-dependent findings evaluation information MREI, preferably of the same patient. The aforementioned information, ieCurrent findings information CMRI and / or historical findings information HMRI and / or quantitative and / or time-dependent findings evaluation information MREI and / or current evidence information CEI can be provided to the user individually and / or in combination and / or in a suitable spatial arrangement and / or suitable chronological sequence using the user interface AS.UI of the reporting station AS.

[0107] According to the schematic representation in Fig. 3, the special development of the exemplary embodiment is further adapted and designed such that - in particular, an automatic marking of the displayed current evidence information CEI and / or the selected historical evidence information HEI can be carried out by means of current evidence relevance information CERI, which can be determined on the basis of user activity data AD and - in particular by means of a (hardware-implemented and / or software-implemented) feedback loop data connection - an automatic optimization of stored - in particular stored in the storage unit SYS.DB - historical evidence information HEI and / or stored historical evidence relevance information HERI can be carried out on the basis of the determined evidence relevance information CERI.

[0108] Single or several or any combination or sequence of steps of the above-described specific embodiment of a method according to the invention according to Fig. 4 as well as individual or several or any combination or sequence of steps of the previously described possible special development of the embodiment according to Fig. 4 according to the wide steps after Fig. 5. can be implemented in particular in the form of a data processing program product or in the form of a data processing program system consisting of several data processing program product modules, which comprise instructions which are executed during the execution of the data processing program product or the data processing program product modules by a data processing unit - for example in a system SYS according to one of the Fig. 1 to 3 - cause this data processing unit to carry out one or more steps of a method described above according to Fig. 4 or Fig. 5, in particular in conjunction with a data processing unit or a system SYS according to one of the Fig. 1 to 3, and in particular in the form of a correspondingly designed computer program product or several computer program product modules, comprising computer-implemented program codes which can be executed in at least one computing unit SYS.CU of data processing units in a medical information network.

[0109] Each of the above steps Fig. 4 or Fig. 5 can in particular be designed as an automatic computer-implemented step, ie for example as computer-implemented recording S1 of current patient information CPI of a patient, automatic computer-implemented display S2 of the current patient information CPI of the patient by means of a user interface AS.UI of at least one diagnosis station AS, automatic computer-implemented determination S3 of patient identification data PID of a patient from the displayed current patient information CPI by means of a first, computer-implemented function F1, automatic computer-implemented determination S4 of stored historical patient information HPI of the same patient and / or of reference patients on the basis of the determined patient identification data PID,Automatic application S5 of a trained second computer-implemented function TF2 to the displayed current patient information CPI and to the determined stored historical patient information HPI of the same patient and / or reference patient, Automatic computer-implemented determination S6 of a relevance measure for the determined stored historical patient information HPI of the same patient and / or reference patient based on initial values ​​of the applied trained second function TF2, Automatic computer-implemented provision S7 of selected stored historical patient information HPI of the same patient and / or reference patient by means of the user interface AS.UI of the diagnosis station AS depending on the determined relevance measure for this stored historical patient information HPI,Automatic computer-implemented acquisition S5.1 of user activity data AD relating to the activity of a user in relation to the current patient information CPI displayed by means of the user interface AS.UI by means of at least one computer-implemented and / or hardware-implemented acquisition device DU, and automatic application S5.2 of the trained second computer-implemented function TF2 to the displayed current patient information CPI and to the determined stored historical patient information HPI of the same patient and / or of reference patients, with computer-implemented correlation of the trained second computer-implemented function TF2 with the acquired user activity data AD.

[0110] The storage unit SYS.DB according to the Fig. 1 and Fig. 3 illustrates a specific embodiment of a computer-readable storage medium on which a data processing program product or data processing program system described above, consisting of a plurality of data processing program product modules, or a computer program product or a computer program product system consisting of a plurality of computer program product modules described above, is stored, or on which program sections readable and executable by a computing unit SYS.CU are stored in order to execute steps of a computer-implemented method for controlling data processing units in a medical information network according to a computer-implemented method described above when the program sections are executed by the computing unit SYS.CU.Such a computer-readable storage medium can also be designed as a mobile data storage device such as a mobile hard disk or USB stick or as a correspondingly designed memory semiconductor component.

[0111] The present invention offers the following applications and advantages in particular: - Increasing diagnostic certainty when viewing patient information and / or evidence without compromising efficiency, - Increased efficiency in complex cases (speed of information access), - Increasing the efficiency of reporting (acquisition and dissemination of information), - Increasing the efficiency of information access in multidisciplinary bodies, - greater focus on the most important anatomical parts of displayed imaging information (based on the patient's medical history).

[0112] The steps described above and / or arrangements adapted to these steps can be further developed in particular as follows, and a further possible specific embodiment and application of the present embodiments and / or the present invention can be explained with the aid of the following further, also exemplary description:

[0113] It can be done in a first exemplary step - in particular in a step S1 according to Fig. 4 - for example, a receipt of previous medical documents (imaging examinations, reports, structured data, ...) for a current patient and an imaging examination, which are subject to a further exemplary step - in particular in a step S2 according to Fig. 4 - are to be presented and reviewed, for example, by a physician as a user (reading by a radiologist, retrieval by physicians in another medical consultation, e.g. tumor board).

[0114] In a further exemplary (pre-processing) step, not explicitly shown in the figures, a structuring of previous unstructured evidence (e.g. reports) can be carried out according to defined ontologies, which allow an automatic determination of the data of such evidence in a further exemplary step - in particular in a step S4 according to Fig. 4 - enables.

[0115] Alternatively or additionally, in a further exemplary step, at least the location of relevant information in an unstructured document can be determined indirectly by comparing it with "similar content types" in relation to a current report and / or in relation to a known ontology. This could, for example, include highlighting and displaying findings in previous images and reports - in particular in a step S7.1 and / or 7.2 according to Fig. 4 - (for example, in relation to the liver organ, when the radiologist, as user, examines the liver using the AS.UI user interface), but also the highlighting of related information when the radiologist examines organs that present secondary pathologies related to the primary liver findings highlighted in the previous report.

[0116] Unstructured information, such as plain text documents, typically lacks the structured information necessary to perform quantitative comparisons desired or required for medical image interpretation and follow-up care. However, this information can be extracted using the trained functions described above, particularly by AI systems such as the previously described artificial neural networks, especially large language models (LLM) or natural language processing (NLP). For example, the AI ​​system ChatGPT has demonstrated high levels of abstraction and comprehension capabilities and could be used as an exemplary concrete AI application for extracting structured information from an unstructured source.

[0117] Furthermore, clinical ontologies can be very valuable inputs for such systems, as they provide information about relationships between information units and enable further abstraction and association.

[0118] In a further exemplary step - in particular in a step S5.1 according to Fig. 4 - the user's current focus of interest can be recorded, for example, interests in specific images or image areas, special areas of interest, activated tools, workflows, or other information of interest such as laboratory values ​​in a table / graph.

[0119] In a further exemplary step, a first exemplary step - in particular in a step S6 according to Fig. 4 and / or step S14 and / or step S15 according to Fig. 5 - for example, previous evidence can be correlated with the current focus of interest in the context of the known user role and / or workflow and / or clinical context and / or a relevance index can be determined and / or calculated.

[0120] For example, a method, computer program product or system according to the invention can use various indicators to track the current focus of interest of the user - in particular in a step S5 or S5.1 according to Fig. 4.

[0121] The detailed representation of the Fig. Figure 6 shows an example of an automatic detection of the focus of interest by means of a detection of user activity data AD regarding the activity of a user during the viewing and / or processing of displayed current patient information CPI, in the example according to Fig. 6 of the current evidence information (CEI) presented, in particular imaging data containing imaging information that is uniquely identified by patient identification data (PID) and can be assigned to a specific patient. This imaging data contains, in particular, imaging information representing organs O1, O2, and O3.

[0122] By means of the detection device DU, an automatic detection of the region R and / or the intensity of the viewing activity and / or the processing activity of the user can now be carried out with regard to the patient information CPI displayed by means of the user interface AS.UI, in particular with regard to displayed current evidence information CEI, and in particular with regard to the imaging data containing the imaging information that represents displayable organs O1, O2, O3. By means of the detection device DU, an automatic detection of the region R and / or the activity and / or the intensity of the activity of an input and / or control element ICU displayed on the user interface AS.UI and controllable by the user - in particular with regard to displayed current evidence information CEI - can be carried out.

[0123] The user's focus of interest, for example a particular image region such as the aforementioned region R and / or data region or information region, which in particular represents a specific anatomical location or region of the body of the respective patient, can be estimated, for example, on the basis of the following data or information, individually or in any combination: - the current and / or previous image display (e.g., series and slice position), its viewing features (e.g., bone / soft tissue window preset, selected organs O1, O2, O3, and / or presets in kinematic views), technical settings or adjustments to the display on the user interface that indicate the location of interest (e.g., zoom settings and / or pan settings, location and / or movement and / or speed and / or track of a cursor or mouse pointer as an example of a user-controllable input and / or control element in the ICU, or eye tracking and / or gesture recognition and / or touch detection via touch-sensitive devices) - the current report template (medical report) and / or the current section (e.g., section for incidental findings) and / or a specific textual and / or semantic characterization of patient information (e.g., about a specific organ O1, O2, O3 of the patient).

[0124] The user's current focus of interest can be further narrowed down, for example, by tracking the currently selected component of the user interface AS.UI, and the selected display of previous medical findings can be selected and optimized for this component - in particular according to a step S2 or S7 according to Fig. 4. For example, - an image viewer can be used to display and highlight locations of previous evidence on images from imaging studies, - a tabular representation and / or graphical representation of biomarkers is used to display biomarkers of interest to the user and to select biomarkers of interest either automatically or through user activity using the AS.UI user interface, and, - a report section may be used to display a suggested set of text and / or a table and / or a graph summarising a change across multiple studies.

[0125] The context for which additional information is provided to the user therefore comes, for example, from the interaction of the user - e.g. a radiologist - with images or image data of an imaging examination. This means that the focus of the user's current interest depends on the images, in particular the section and / or aspect that the radiologist is looking at. For example, the user has 'moved' to a certain point within a certain series of images / image data (e.g. after selecting a reconstruction, orientation, kernel / filter, contrast agent phase) (i.e. scrolled to the slice, zoomed in if necessary, positioned the mouse somewhere, re-windowed). This set of information can be used by the system to decide which additional information about the patient is currently of interest. Using existing technology (e.g. alpha), the system can - if necessaryDetermine – even pixel-precisely – which organ O1, O2, or O3 the user is currently actively viewing. Based on the imaging, the current viewing settings, and, if applicable, the report template used, the system can support filtering of additional information from the patient history.

[0126] For this purpose, in addition to or instead of data and information from the current patient, data and information from real, fictitious, or synthetic reference patients can also be used. This allows the system to learn not only from the history of a single current patient, but also to improve its training based on many reference patients. The system can use this knowledge not only to retrieve and display a selection of information from the same patient, but also in combination with additional information that is helpful to the user: "Why is this provided information important?" In particular, the system can provide insights from a group of reference patients, such as: "The current patient shows the following deviations from normal values ​​in their history, and in 80% of cases from a group of reference patients, this deviation indicates that..."For example, that a certain course of disease has occurred. This allows, in particular, to show the user only information about the current patient, and in particular, to relate this information to the knowledge and reason for the system's decision to consider the presented information important and to display it, as well as, in particular, to provide the relevant evidence as to why this information is important.

[0127] Overall, a method, computer program product or system according to the invention can take into account and / or use a current workflow step (e.g., quick review at the scanning workstation, reading or viewing, tumor board preparation and / or demonstration) and the role of the user (technologist, junior / senior radiologist, oncologist, ...) as well as personal preferences and / or the personal level of experience of the respective user in order to limit the display of evidence and / or findings to information that may be of interest to the user.

[0128] The system knows and / or learns the previous and current evidence in a structured and semantic way (“what” is available) - in particular according to a step S14 and / or S15 according to Fig. 5 -, it knows and / or learns the current task of the user (“where” the user performs a task), it knows and / or learns the user’s preferences (“how” the user will appreciate the support), the system can calculate a relevance index - in particular according to a step S6 according to Fig. 4 - and select and provide evidence - in particular according to a step S7 according to Fig. 4 - to support the user.

[0129] The relevance index - in particular according to a step S7 according to Fig. 4 - can be calculated, for example, for a mapping table that correlates the semantic evidence model with the user and / or with the workflow step and / or with the user interface component to enable automatic notification and evidence presentation while the user performs, for example, routine tasks using the user interface AS.UI.

[0130] In a further exemplary step, a coupling with computer-assisted recognition and quantification systems and / or a provision of additional supporting evidence and / or a calculation of an overarching relevance index can take place. The method, computer program product, or system according to the invention can, for example, automatically display AI-generated findings based on the current imaging examination - in particular according to a step S9 and / or S10 according to Fig. 5 - and use these findings automatically to visualize and display previous findings - in particular according to a step S11 and / or S12 according to Fig. 5.

[0131] The method, computer program product or system according to the invention can also understand previous findings - in particular according to a step S11 according to Fig. 5 - and / or learn and / or be trained thereon - in particular according to a step S13 according to Fig. 5 - and thus, for example, put an optimized, newly generated AI algorithm into context. For example - in particular using an exemplary step according to a step S12 according to Fig. 5 - it may turn out that a tiny lung nodule in a patient with no history of cancer - i.e. without relevant historical patient information - may not be alarming, whereas in a cancer patient - i.e. with relevant historical patient information - the finding needs to be followed up.

[0132] Furthermore, the method, computer program product, or system according to the invention can automatically provide the user with this interpretation and recommendations for further processing. Furthermore, the system can use stored previous findings or information and / or information derived from stored and / or current information as input for optimizing the AI ​​algorithm - in particular according to a step S13 and / or step S15 according to Fig. 5 - in order to increase the ROC (Receiver Operating Characteristic) performance. For example, in particular according to one or more of the steps S9 to S12 according to Fig. 5 - the original AI algorithm does not detect any finding based on the current imaging information, but by - supplementing - using known stored previous findings, i.e. based on stored historical patient information HPI, the AI ​​algorithm can automatically detect that there was a finding nearby or that markers indicate an increased risk of findings in a particular part of the body.

[0133] In a further exemplary step - in particular according to a step S2 or S7 according to Fig. 4 - the relevant previous findings can be integrated into the AS.UI user interface without hindering the user's workflow when reading medical information and / or evidence and / or when viewing information from imaging examinations and / or during the diagnosis, but rather to give the user the opportunity to quickly review them and, if necessary, provide further details and / or transfer and / or update findings for a current report. This could, for example, be done - in particular according to a step S2 or S7 according to Fig. 4 - by displaying selected information in a diagnostic cockpit, a separately displayed, for example "floating" information search window or by automatic suggestions in the currently processed medical report (similar to an auto-completion function).

[0134] In a further exemplary step - in particular according to one or more steps S9 to S12 according to Fig. 5 - a transfer of previous, i.e. stored historical findings and / or stored historical quantifications to a current diagnosis and / or imaging study and / or an update of measurements and / or trend information can be carried out, in particular by means of a data interface to a corresponding computer-implemented reporting system.

[0135] In a further exemplary step - not explicitly shown in the figures - a connection to a computer-implemented knowledge gateway subsystem can be made, for example by providing data or information on a computer-implemented description of a "center of interest" that is enriched with previous findings, i.e. data from stored historical patient information HPI or with information derived from it from various clinical disciplines, in order to display the results of relevant knowledge sources (e.g. atlas, textbooks, guidelines, risk calculators, etc.) that have been refined according to the enriched patient context by corresponding computer-implemented processing steps.

[0136] Regardless of the grammatical gender of a particular term, persons with male, female or other gender identities are included. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2022 201 630 A1 [0005, 0006, 0025, 0050, 0104] Cited non-patent literature

[0000] Miotto, R., Li, L., Kidd, B. et al. Deep Patient: An Unsupervised Representation to Predict the Future of Patients from the Electronic Health Records. Sci Rep 6, 26094 (2016 [0050, 0057, 0104]

Claims

[1] A computer-implemented method for controlling units in a medical information network, comprising: a computing unit (SYS.CU); and at least the following units connected to the computing unit (SYS.CU): at least one diagnostic station (AS) for preparing a medical report; at least one data acquisition unit (SYS.DCU) for recording current patient information (CPI); and at least one storage unit (SYS.DB) for storing historical patient information (HPI); wherein the computer-implemented method comprises the following steps: - automatic recording (S1) of current patient information (CPI) of a patient using the data acquisition unit (SYS.DCU); - automatic display (S2) of the patient's current patient information (CPI) via a user interface (AS.UI) of the diagnosis station (AS); - automatic retrieval (S4) of stored historical patient information (HPI); - automatic application (S5) of a trained, computer-implemented function (TF2) executable in the computing unit (SYS.CU) or the diagnosis station (AS) to the displayed current patient information (CPI) and to the determined stored historical patient information (HPI); - automatically determining (S6) a relevance measure for the determined stored historical patient information (HPI) based on the trained function (TF2); and - automatic provision (S7) of selected stored historical patient information (HPI) via the user interface (AS.UI) of the diagnostic station (AS) depending on the determined relevance level for this stored historical patient information (HPI). [2] Method according to claim 1, wherein the step of automatically applying (S5) a trained second function (TF2) comprises: - automatic recording (S5.1) of user activity data (AD) relating to the activity of a user in relation to the current patient information (CPI) displayed by means of the user interface (AS.UI) by means of a recording device (DU) which is in data connection with the diagnosis station (AS) and with the computing unit (SYS.CU); and - automatic application (S5.2) of the trained function (TF2) to the displayed current patient information (CPI) and to the determined stored historical patient information (HPI) depending on - in particular in correlation with - the recorded user activity data (AD). [3] Computer-implemented method for controlling units in a medical information network, in particular a method comprising one or more of the features according to claim 1 or 2, wherein the computer-implemented method comprises the following steps: - automatic display (S2) of current patient information (CPI) using a user interface (AS.UI) of a diagnostic station (AS); - automatic recording (S5.1) of user activity data (AD) relating to the activity of a user in relation to the current patient information (CPI) displayed by means of the user interface (AS.UI) by means of a recording device (DU) which is in data communication with the diagnosis station (AS) and with the computing unit (SYS.CU); and - automatically applying (S5.2) a trained - in particular a second, computer-implemented - function (TF2) to the displayed current patient information (CPI) and to stored historical patient information (HPI) - in particular determined by means of a first, computer-implemented function (F1) - depending on - in particular in correlation with - the recorded user activity data (AD) for the purpose of automatically providing (S7) selected stored historical patient information (HPI). [4] Method according to one of claims 1 to 3, wherein continuously and / or iteratively an automatic acquisition (S5.1) of user activity data (AD) and / or an automatic application (S5.2) of the trained second function (TF2) takes place, in particular wherein the computer-implemented method includes the following steps: - automatically capturing (S5.1) second and / or further user activity data (AD) relating to the activity of a user in relation to the current patient information (CPI) displayed by means of the user interface (AS.UI) and / or the selected stored historical patient information (HPI) by means of the capturing device (DU); - automatically - in particular repeatedly or multiple times - applying (S5.2) the trained second function (TF2) to the displayed current patient information (CPI) and to the determined stored historical patient information (HPI) depending on - in particular in correlation with - the recorded second and / or further user activity data (AD) for adapting and / or verifying a determined relevance measure for the determined stored historical patient information (HPI); and - automatic selection and / or automatic provision (S7) of - in particular additional - selected stored historical patient information (HPI) by means of the user interface (AS.UI) of the reporting station (AS) depending on the adapted and / or verified determined relevance measure. [5] Method according to one of claims 1 to 4, wherein - an automatic recording (S5.1) of user activity data (AD) regarding the activity of a user during the viewing and / or processing of displayed current patient information (CPI), in particular of displayed current evidence information (CEI); where - the detection device (DU) automatically detects the region (R) and / or the intensity of the viewing activity and / or the processing activity of the user in relation to the patient information (CPI) displayed by means of the user interface (AS.UI), in particular in relation to the current evidence information (CEI) displayed; whereby - in particular by means of the detection device (DU) an automatic detection of the region (R) and / or the activity and / or the intensity of the activity of an input and / or control element (ICU) interacting with the user and / or with the user interface (AS.UI) - in particular one displayed on the user interface (AS.UI) - and controllable by the user - in particular with regard to displayed current evidence information (CEI) - is carried out. [6] Method according to one of claims 1 to 5, wherein - the current patient information (CPI) presented includes current evidence information (CEI) of a patient regarding current medical evidence; - the selected stored historical patient information (HPI) includes stored historical evidence information (HEI) relating to historical medical evidence; and - the stored historical evidence information (HEI) is stored in the form of structured evidence information and / or unstructured evidence information. [7] Method according to one of claims 1 to 6, wherein the steps of displaying (S2) the current patient information (CPI) and automatically providing (S7) the selected stored historical patient information (HPI) comprise: - Displaying (S2) the current patient information (CPI) in a first display area (DA1) of a user interface (AS.UI) of the diagnosis station (AS); and - automatic provision (S7) of the selected stored historical patient information (HPI) in a second display area (DA2) of the user interface (AS.UI) of the diagnostic station (AS) depending on the determined relevance measure for this stored historical patient information (HPI). [8] A method according to any one of claims 1 to 7, further comprising the following steps: - automatically applying (S8) a trained third computer-implemented function (TF3) executable in the computing unit (SYS.CU) or the diagnosis station (AS) to the current patient information (CPI) and to the provided selected stored historical patient information (HPI); - automatic determination (S9) of current and / or historical findings information (CMRI, HMRI) and associated findings probability score information (PSI) with respect to the current patient information (CPI) based on the trained third function (TF3); and - automatic provision (S10) of the current and / or historical findings information (CMRI, HMRI) via the user interface (AS.UI) of the reporting station (AS) depending on the associated findings probability score information (PSI) for this current findings information (CMRI). [9] The method of claim 8, further comprising the steps of: - automatic linking (S11) of current findings information (CMRI) and stored historical findings information (HMRI) based on user activity data (AD); and - automatic application (S12) of a fourth computer-implemented function (F4) executable in the computing unit (SYS.CU) or the reporting station (AS) to the current finding information (CMRI) and stored historical finding information (HMRI) of the same patient to determine quantitative and / or time-dependent finding evaluation information (MREI). [10] A method according to claim 8 or 9, further comprising the step of: - automatic training and / or optimization (S13) of parameters of the trained second function (TF2) and / or the trained third function (TF3) based on current finding information (CMRI) and associated finding probability score information (PSI) and / or based on selected stored historical patient information (HPI) and associated relevance measure information and / or user activity data (AD). [11] Method according to one of claims 1 to 10, wherein the step of automatically providing (S7) the selected stored historical patient information (HPI) comprises: - automatic provision (S7.1) via the user interface (AS.UI) of selected stored historical patient information (HPI) whose determined relevance measure exceeds a predefined threshold towards high relevance, in conjunction with a high relevance label and / or high relevance highlighting and / or high relevance positioning; and / or - automatic provision (S7.2) by means of the user interface (AS.UI) of selected stored historical patient information (HPI) whose determined relevance measure falls below a predefined threshold towards low relevance, in conjunction with a low relevance label and / or low relevance highlighting and / or low relevance positioning. [12] A method according to any one of claims 1 to 11, further comprising the following steps: - automatically labelling (S14) the displayed current evidence information (CEI) and / or the selected historical evidence information (HEI) using current evidence relevance information (CERI) determined on the basis of user activity data (AD); and - automatic optimization (S15) of stored historical evidence information (HEI) and / or stored historical evidence relevance information (HERI) based on the determined evidence relevance information (CERI). [13] Data processing unit or system (SYS), in particular designed to carry out a method according to one of claims 1 to 12, wherein the data processing unit or the system (SYS) has a computing unit (SYS.CU) and wherein the computing unit (SYS.CU) is in data connection with at least the following units: - at least one diagnosis station (AS) designed to display (S2) and / or provide (S7) patent information (CPI, HPI) and to create a medical diagnosis; - at least one data acquisition unit (SYS.DCU) designed to acquire (S1) current patient information (CPI); and - at least one storage unit (SYS.DB) designed to store historical patient information (HPI); wherein the computing unit (SYS.CU) is configured to execute computer-implemented program codes and to generate control commands to perform the following steps: - automatically applying (S5) executable, computer-implemented program codes representing a trained function (TF2) to displayed current patient information (CPI) and to stored historical patient information (HPI); - automatically determining (S6) a relevance measure for the determined stored historical patient information (HPI) based on initial values of the applied trained function (TF2); and - Generation of control commands for the automatic provision (S7) of selected stored historical patient information (HPI) by means of a user interface (AS.UI) of the diagnosis station (AS) depending on the determined relevance measure for this stored historical patient information (HPI). [14] Data processing unit or system (SYS), in particular according to claim 13 and in particular designed to carry out a method according to one of claims 2 to 12, wherein the system (SYS) comprises the following: a computing unit (SYS.CU); and at least the following units connected to the computing unit (SYS.CU): - at least one diagnosis station (AS) with at least one user interface (AS.UI), designed to display (S2) and / or provide (S7) current and historical patent information (CPI, HPI) and to create a medical diagnosis; - at least one data acquisition unit (SYS.DCU) designed to acquire (S1) current patient information (CPI); - at least one storage unit (SYS.DB) designed to store historical patient information (HPI); and - at least one recording device (DU) designed to automatically record (S5.1) user activity data (AD) relating to the activity of a user in relation to the current patient information (CPI) displayed by means of the user interface (AS.UI); wherein the computing unit (SYS.CU) is configured - to execute computer-implemented program codes representing a trained function (TF2) on current patient information (CPI) and on historical patient information (HPI) depending on - in particular in correlation with - recorded user activity data (AD); and - for generating control commands for the automatic provision (S7) of selected stored historical patient information (HPI) by means of a user interface (AS.UI) of the diagnosis station (AS) depending on initial values of the applied trained function (TF2). [15] Data processing program product or data processing program system consisting of several data processing program product modules, comprising instructions which, when the data processing program product or the data processing program product modules are executed by a data processing unit, cause this data processing unit to carry out the steps of a method according to one of claims 1 to 12; in particular in cooperation with a data processing unit or a system (SYS) according to one of claims 13 or 14; and in particular in the form of a correspondingly designed computer program product or several computer program product modules, comprising computer-implemented program codes which can be executed in at least one computing unit (SYS.CU) of a data processing unit or a system (SYS) in a medical information network. [16] Computer program product or computer program product system consisting of several computer program product modules, comprising instructions which, when the computer program product or the computer program product modules are executed by at least one data processing unit or a system (SYS) with at least one computing unit (SYS.CU), cause the at least one data processing unit or the system (SYS) to execute the following program steps; in particular the steps of a method according to one of claims 1 to 12 and / or in particular in cooperation with a data processing unit or a system (SYS) according to one of claims 13 or 14: - automatic computer-implemented recording (S1) of current patient information (CPI) of a patient; - automatic computer-implemented display (S2) of the patient's current patient information (CPI) by means of a user interface (AS.UI) of at least one diagnosis station (AS); - automatic computer-implemented retrieval (S4) of stored historical patient information (HPI); - automatic computer-implemented application (S5) of a trained computer-implemented function (TF2) to the displayed current patient information (CPI) and to the determined stored historical patient information (HPI); - automatic computer-implemented determination (S6) of a relevance measure for the determined stored historical patient information (HPI) based on initial values of the applied trained function (TF2); and - automatic computer-implemented provision (S7) of selected stored historical patient information (HPI) by means of the user interface (AS.UI) of the diagnostic station (AS) depending on the determined relevance measure for this stored historical patient information (HPI). [17] Computer program product or computer program product system consisting of several computer program product modules according to claim 16, comprising instructions which, when the computer program product or the computer program product modules are executed by at least one data processing unit or a system (SYS) with at least one computing unit (SYS.CU), cause the at least one data processing unit or the system (SYS) to execute the following program steps: - automatic computer-implemented recording (S5.1) of user activity data (AD) relating to the activity of a user in relation to the current patient information (CPI) displayed by means of the user interface (AS.UI) by means of at least one computer-implemented and / or hardware-implemented recording device (DU); and - automatic application (S5.2) of the trained computer-implemented function (TF2) to the displayed current patient information (CPI) and to the determined stored historical patient information (HPI) under computer-implemented correlation of the trained computer-implemented function (TF2) with the recorded user activity data (AD). [18] A computer program product or computer program product system comprising a plurality of computer program product modules, designed to control units in a medical information network, comprising commands which, when the computer program product or the computer program product modules are executed by at least one data processing unit with at least one computing unit (SYS.CU), cause the at least one data processing unit or a system (SYS) to execute the following program steps; in particular, to execute the steps of a method according to one of claims 1 to 12 and / or in particular in cooperation with a data processing unit or a system (SYS) according to one of claims 13 or 14: - automatic computer-implemented display (S2) of current patient information (CPI) using a user interface (AS.UI) of a diagnostic station (AS); - automatic computer-implemented recording (S5.1) of user activity data (AD) relating to the activity of a user in relation to the current patient information (CPI) displayed by means of the user interface (AS.UI) by means of at least one computer-implemented and / or hardware-implemented recording device (DU); and - automatically applying (S5.2) a computer-implemented trained - in particular second - function (TF2) to the displayed current patient information (CPI) and to stored historical patient information (HPI) - in particular determined by means of a first, computer-implemented function (F1) - depending on - in particular in correlation with - the recorded user activity data (AD) for the purpose of automatically providing (S7) selected stored historical patient information (HPI). [19] Computer program product or computer program product system consisting of several computer program product modules, according to claim 17 or 18, designed to control units in a medical information network, comprising commands which, when the computer program product or the computer program product modules are executed by at least one data processing unit with at least one computing unit (SYS.CU), cause the at least one data processing unit or a system (SYS) to execute the following program steps: - automatic computer-implemented recording of user activity data (AD) relating to a user's activity while viewing and / or editing displayed current patient information (CPI), in particular displayed current evidence information (CEI); and - automatic computer-implemented detection of the region (R) and / or the intensity of the user's viewing activity and / or editing activity in relation to the patient information (CPI) displayed by means of the user interface, in particular in relation to the current evidence information (CEI) displayed; - in particular automatic computer-implemented detection of the region (R) and / or the activity and / or the intensity of the activity of an input and / or control element (ICU) interacting with the user and / or with the user interface (AS.UI) - in particular one displayed on the user interface (AS.UI) - and controllable by the user - in particular with regard to displayed current evidence information (CEI). [20] Computer-readable storage medium on which a data processing program product or data processing program system consisting of a plurality of data processing program product modules according to claim 15 or a computer program product or a computer program product system consisting of a plurality of computer program product modules according to one of claims 16 to 19 is stored or on which program sections readable and executable by a computing unit (SYS.CU) are stored in order to carry out steps of a computer-implemented method for controlling units in a medical information network according to one of claims 1 to 12 when the program sections are executed by the computing unit (SYS.CU).

Images