DE102017007099A1 - Medical device for storage and evaluation of clinical data - Google Patents

Abstract

The invention relates to a medical device for storing medical data for information collection, information processing and iterative, information-based process optimization, which automatically accumulate knowledge and accumulate knowledge by an automated and anonymized collection of data from networked components of the operating room for minimally invasive surgery ( MIC), eg Includes insufflators, pumps, cameras, monitors, therapeutic instruments, as well as post-operative evaluations of treatment success associated with these data.

Description

Field of application of the invention

The object of the invention comprises a device for the use and provision of data from networked medical devices for a patient-individual and the workflow relieving device setting.

In this context, "networking" means the comprehensive use of information in the operating room of existing and relevant data. By "medical device" is meant any energy-powered or non-powered device used for diagnostic or therapeutic interaction with the human or animal body. Accessories and disposables are also explicitly designated as medical devices.

State of the art

Currently, information about the surgical procedure and the recovery process of the patient are not fed into a central pool of experience and are therefore not automatically available as a general knowledge base for future interventions. Technical process variables are hardly analyzed and evaluated in the context of the clinical outcome, the experiences of performed operations are not adequately evaluated, documented and published.

Medical guidelines provide evidence-based clinical trials that delay the medical procedures for diagnostic and / or therapeutic procedures. These guidelines are not sufficiently detailed as consensus papers of medical societies for the settings of medical devices or technical parameters. Here, the manufacturers of medical devices have in-depth knowledge, which is partly taught in training.

All of these ways from a standardized application of medical devices for a diagnostic or therapeutic procedure to good clinical results in the patient are tedious and associated with great effort.

Furthermore, studies often fail to track the complex relationships between device settings and the clinical success of the treatment because the documentation of the device parameters by the study physicians is usually not possible. Setpoint specifications are manually set by medical personnel based on individual experience

The EP1995679 describes a device and a method for networking and central operation and data transmission for documentation purposes and / or control or parameter adjustment of at least one device that is used during a medical procedure. Central is a database, which is kept on a control unit, an external server accessible via networks or a readable data medium. Paragraph [0030] describes access to the central database via local area network or Internet

According to paragraph [0008], this data base should contain "expert knowledge" and / or predetermined device parameters independent of the respective physician, which will be extended to propose the device combination mentioned in paragraph [0013] for the respective medical procedure and the physician, ie a type Assistance system for instrumentation. Paragraph [0023] describes that predetermined device configurations and parameters may be part of the database.

Paragraph [0016] describes that device parameters or device actions are triggered by a signal automatically generated by a sensor. However, there is no technical teaching, except that a device presetting is automatically adopted - due to which sensors and what cause this is not executed. A regulation of device parameters by sensor values in particular, as the solution according to the invention illustrated here carries out, is not described and not obvious.

Each acquisition of the database contents in a device is bound in the invention necessarily with a physician approval, so the operation of a confirmation element.

Paragraph [0028] describes that data of the patient (age, height, weight, existing or already recognized diseases, anatomical peculiarities) require changed parameter values and device configurations and optimal parameters depend on the devices and manufacturers used. Paragraph [0029] describes their provision in the database and various patient-dependent data sets or a formula for calculating patient-dependent parameters are described. The technical teaching is exemplified by the interpolation or extrapolation. The solution according to the invention provides an evidence-based approach, so this is also not described or obvious.

Inventive solution

For the solution of the defined problems, according to the invention, the collection and analysis of data in the medical device according to the present claims takes place by assumption from the intraoperative phase and collecting new data in the postoperative period. In a further step, from the currently transmitted data for already known scenarios, which are stocked with previously collected data, in the preoperative phase by derivation of the patient data, a derivation of device configurations or device settings can take place and instructions for the guideline-compatible treatment from the intraoperative data the medical device will be transmitted.

The prerequisite is the provision of data interfaces to sources of information in the operational environment, such as the hospital information system or the anesthesia system and the specification of relevant information sources and interfaces. These interfaces may be implemented electronically or otherwise, e.g. by image recognition and evaluation of a medical device with camera of data on forms or screens.

The field of application of the expert system may include further functional units in the operating room. Thus, the most important area of anesthesia greatly benefits from a constantly updated experience system and the connections to other clinical devices and systems.

The expert system is a device for collecting data from medical devices and pre- and postoperative patient data, which combines patient data with process data and clinical treatment successes determined during the procedure, thereby ensuring the association of the data with each other, grouping such data groups and containing the process data and process information bundling the best clinical results and continuously weighting each other with new data input, determining the process data with the best clinical result as the optimum, and returning the determined optimum to the medical device as a presetting. The return involves clinical guidelines and clinical experience in relation to the data groups and selection rules. The medical device also has a data fallback level with failed data transmission.

Automated collection and response of device configurations is accomplished through a centralized data collection, data analysis, and storage device, an expert system so-called, accessible via a network from the medical device. This expert system consists of a selection and correlation module, a rules and expertise feed-in module with a self-learning sub-module and a data-retention module that maps, classifies, stores and controls content access. The contents are data sets consisting of anonymized patient data, anamnesis data, procedure information, information on medical devices used, intraoperative vital data and device data or device parameters, postoperative treatment evaluations. Furthermore, procedural rules that are used in the selection and correlation module are stored in the expert system. As one embodiment, a server may be reachable over the Internet and send back only parameter maps to the device, which provides the source information (device parameters, vital data, patient data).

Due to the central provision, e.g. Easily incorporate changes to guidelines or adjustments to changed procedures.

The concept of an experience and knowledge database updating with each intervention, which is made available to each surgeon, leads to a continuous optimization of minimally invasive surgery.

By linking the information from the expert system with the data from the anamnesis and device detection device configurations can be selected automatically. These settings can be optimized for each individual patient to the respective therapy and thus improve preoperatively minimally invasive surgery. The same applies to setpoint specifications such. the pressure in the operating area.

• • Datenkommunikationsschicht zur Vernetzung der Geräte im OP
• • Datenklassifizierung bzw. -gruppierung von Patientendatenaus der präoperativen Phase
• • Ermittlung bzw. Übertragung von für die medizinische Prozedur genutzten Medizingeräten
• • Ermittlung und Übertragung von Sollwertempfehlungen auf Grundlage von apriori-Wissen in das Medizingerät

The inventive solution for the expert system includes at least one

• • Data communication layer for networking devices in the OR
• • Data classification or grouping of patient data from the preoperative phase
• • Identification or transmission of medical devices used for the medical procedure
• • Identification and transmission of setpoint recommendations based on a priori knowledge of the medical device

• • Identifikationsmethode zur automatischen Erkennung von eingesetzten Geräten in der präoperativen Phase
• • Ermittlung und Übertragung Kennliniensatz zur Reglervoreinstellung auf Grundlage von apriori-Wissen in das Medizingerät
• • Datenübertragung aus der intraoperativen Phase in das Datenbanksystem
• • Ermittlung und Übertragung eines Kennliniensatzes zur Reglereinstellung auf Grundlage von intraoperativen Daten in das Medizingerät
• • Ermittlung und Übertragung von Sollwertempfehlungen auf Grundlage von intraoperativen Daten in das Medizingerät
• • Datenübertragung aus der postoperativen Phase in das Datenbanksystem
• • Einpflegen von Regeln die dem medizinischen Expertenwissen bzw. Leitlinien entstammen

Furthermore, the solution according to the invention may include

• • Identification method for automatic detection of used devices in the preoperative phase
• • Identification and transmission of the set of preselector values based on apriori knowledge of the medical device
• • Data transfer from the intraoperative phase to the database system
• • Identification and transmission of a set of controllers for regulator adjustment based on intraoperative data in the medical device
• • Identification and transmission of setpoint recommendations based on intraoperative data in the medical device
• • Data transfer from the postoperative phase to the database system
• • Inclusion of rules that derive from the medical expertise or guidelines

Automated information processing analyzes and classifies patient and device data and passes it to the database for further processing and logging. The classification procedure corresponds to the established procedures and a suitable classifier structure is created for each area of application. Training and validation of the classifier is done using data from the clinical environment.

The database uses data from established medical evaluation methods to evaluate the clinical outcome. A pre-and intra-operative phase and anonymized analyzer process data associated with the database combines expert system records with past intervention records and newly collected data on the operation and clinical outcome, and identifies successful combinations. The incorporation of medical expertise into the knowledge base provides recommendations for action in defined clinical scenarios based on current medical guidelines. This knowledge database is combined with the expert system, which feeds from the statistical evaluation of current and past interventions. This provides the information base for a constantly learning assistance system, which is available for the optimization of future interventions.

• • Evaluation der operativen Resultate (postoperatives Outcome bzw. klinischer Behandlungserfolg)
• • Statistische Auswertung und Clusterung anonymisierter klinischer Daten
• • Sammlung von medizinischen Handlungsempfehlungen aus Expertenwissen
• • Rückkopplung von Erfahrungen auf die Gerätevoreinstellungen .
• • Neben der Nutzung vorhandener Informationen, werden Prozessinformationen gesammelt, analysiert und für andere vernetzte, medizinische Teilsysteme bereitgestellt.

Information from the intraoperative phase is collected, anonymized and analyzed to build an expert system in the postoperative period. For this purpose, in the postoperative phase, among others:

• • Evaluation of the operative results (postoperative outcome or clinical treatment success)
• • Statistical evaluation and clustering of anonymized clinical data
• • Collection of medical recommendations from experts
• • Feedback from experience to the device default settings.
• • In addition to using existing information, process information is collected, analyzed, and shared with other networked medical subsystems.

Data classes (for example, patient type, operation type) or groupings of data records are created and the associated patient and anamnesis data are analyzed. From the data analysis, setpoint recommendations and / or controller presettings of the medical device are derived by combining the device parameters present in the data sets from the data sets with the same data classes with the same data classes in relation to the patient and the procedure in accordance with prescribed rules and recommending or setting the device be returned to the medical device via the network.

Such a predetermined rule could be to give preference to the lowest possible target pressure value for the body cavity as a returned device parameter. Another rule could be an age-dependent setpoint pressure, which is fixed. These rules should be designed according to medical requirements and risk assessment. The records are collected continuously. Returned device settings can be changed by the surgical team. In addition to the device setting values that react directly to vital data, presets can be used to determine device presets or controller behavior.

In the automated update of the expert system, the course of therapy is quantified and relevant aspects are logged. The feedback of experience from the aftercare of future operations creates a digital supply chain. An automated optimization of minimally invasive surgery is used.

According to the invention, source and control parameters of insufflators and pumps in minimally invasive surgery in the medical device are changed directly or indirectly by data input, whereby an improved device function is ensured, which improves the view in the surgical field, reduces bleeding, and the burden on the patient's circulation minimized.

The combination of all pre-, intra- and postoperative patient and surgical information proposed as a solution according to the invention makes it possible to relieve the surgeon or the clinical staff of technical configuration procedures of the devices used.

• - Verbesserung der Therapie durch Klassifizierung und Nutzung von apriorischem Wissen und Erfahrungen
• - Quantifizierung des Behandlungserfolgs
• - Sammlung von Expertenwissen
• - Verbesserung zukünftiger Eingriffe durch Rückkopplung von Erfahrungswerte und Expertenwissen in eine Datenbank

mittels

• - Methoden zur automatischen Datenverarbeitung und Klassifikation
• - Methoden zur apriorischen Reglereinstellung und Sollwertempfehlung
• - Methoden zur automatisierten Auswertung und Bewertung klinischer Daten
• - Iterative Optimierung der MIC durch automatische Erfahrungsakkumulation.

Thus, a

• - Improving therapy by classifying and using a priori knowledge and experience
• - Quantification of treatment success
• - Collection of expert knowledge
• - Improvement of future interventions by feedback of empirical values and expert knowledge into a database

by means of

• - Methods for automatic data processing and classification
• - Methods for apriori controller adjustment and setpoint recommendation
• - Methods for the automated evaluation and evaluation of clinical data
• - Iterative optimization of MIC through automatic accumulation of experience.

Anamnesis data is collected and logged before each surgery. Up to now, the use of this a priori data is only done manually by the clinical staff. An automated evaluation by classification methods is to be researched in order to be able to better tailor the therapy to the individual patient.

All a priori data from the preoperative phase and the experience and knowledge database according to the invention should be used for automatic device adjustment. The device configuration should be automated in the form of controller parameters and setpoint specification.

The solution according to the invention likewise includes a holistic optimization of process sequences by linking electronically available data with the control of quality-relevant process variables for optimizing the device settings during the course of the operation with the data stored in the databases before and after an operation. The basic idea of the invention is the iterative optimization of MIC procedures through data collection.

According to the invention, an automated anamnesis evaluation, which includes a fusion of current patient data with an expert system. The device systems used during the procedure should be automatically recognized by means of identification methods. Furthermore, action recommendations for the surgeon are generated from all available apriori data.

The transmission of patient and anamnesis data is not only patient specific, but also case-specifically related to intraoperatively collected data, i. based on the time period and the clinical procedure performed, and this data set in turn the postoperative quantitative assessments of clinical treatment success and / or adverse effects or complications.

The transmission of the data can be automated by a human operator without the intervention of a human operator or with the initiation of a human operator, or manually by input to a data entry screen.

The feedback of device settings to the medical device can be made via a previously established secure connection or as an addressed and encrypted data packet with a defined destination. The reception of the data packet can be acknowledged by the receiving and designated as the medical device.

Certain procedures that are based on the medical process or for other reasons, a specific reaction of the medical device can be made as a rule formulated deposited in the evaluation of the expert system. These may include, for example, a reduction in joint pressure after a defined time above a reference pressure, flushing (media exchange) of the expanded body cavity after a defined period of time or more complex relationships, such as the warning after a certain pressure-time combination (with a longer time with low pressure and a maximum pressure-time limit should not be exceeded) or the gradual reduction of the desired pressure as the procedure progresses or after a set increase above a defined threshold value. Further rules are conceivable and can be answered via the device parameters sent by the medical device from the server with coordinated device settings.

Like the patient and anamnesis data, the postoperative evaluations of the success of the treatment are also fed back into the expert system and the other data in a patient-specific and case-specific manner. The assignment may be made anonymously according to the prior art via ordering features generated previously (e.g., a case number of the hospital) or over time (date) and patient data (name, date of birth, and / or health insurance number).

Of course, it is important for the implementation of a centralized expert system that medical devices also fail to connect in the run-up to commissioning and in the event of a network connection failure or malfunction, the operation continues in a safe state. If a connection to the server is not available or the data transmission is interrupted in the course of the procedure, a fallback level comes into action in the medical device, which ensures a safe state for the operation of the medical device. It will keep certain sizes, if necessary, change other sizes to a safe mode.

For this purpose, the previously implemented in the medical devices control logic or control logic is kept active and set by the possibility of inserting input via the network interface, the function of the control or regulation logic is not overridden. This can be accomplished by preserving the injected values and always effecting a safe state, or by detecting a connection failure and changing the changed values to a safe value, or a mixture of both.

On the basis of an expert system, the clinical personnel are issued by an inventive assistance system action recommendations and / or pointed to critical conditions.

Setpoint recommendations and / or controller presettings of the medical devices derived from the data analysis flow into an assistance system to support the clinical staff. The assistance system assists the surgeon by linking a constantly expanding expert system with records of individual patient data and technical metrics to achieve optimal clinical outcomes.

The assistance systems can also derive the device settings suitable for particular medical devices from the information of the inserted or recognized medical devices in the operating room and transmit them back to the requesting medical device. This is an experiential knowledge automatically retrievable.

Due to the networked provision of device parameters and device data, detection of certain critical states can be automated outside the device and the surgeon and / or device operator in the operating room can be assured of a secured device configuration or settings that can be rejected or accepted at the push of a button. An automated takeover is also conceivable, but this requires a higher amount of risk management or validation of the function.

Further assistance functions can be in a trend analysis of values that generates a warning and thus changes the further progress according to the trend by reaction of the surgical team or changed device settings.

It is also conceivable to visualize processes as assistance, that is to say graphic progressions of device parameters with an inserted ideal course or border areas.

Another aspect of the expert system connection is a collection of data provided for the documentation of the individual surgical procedure to the hospital or evaluated as quality assurance, used anonymously for medical studies, or used anonymously as a basis for the development of medical device controls.

Pre-sorting based on clinical expertise is intended to divide the anamnestic data into data classes (e.g., patient type, surgical or procedure type).

For example, patient therapy is significantly improved by reducing bleeding and / or preventing unwanted tissue damage from excessive pressure in the surgical site during medical procedures in the MIS. The limits are difficult to determine and therefore well to determine from procedures with good to very good treatment success according to the law of large numbers. This approach is known in medicine as "evidence-based." The law of large numbers is necessary because each patient responds differently to the interventions and therapeutic measures and therefore results in a dispersion. With bundling and classification, many collected data sets can be merged into one type and in this group between class the device parameters are analyzed and correlated with the quantified treatment success. A pure correlation would probably generate wrong values, so an additional rule-based selection can take place. The expert system determines the optimal device settings reported back to the medical device based on a statistical evaluation of the surgical data and the determined operation evaluations. This establishes a connection between the type of patient and the type of surgery, the procedure performed and the result achieved. The statistical evaluation is done on an anonymous data basis. The information collection, evaluation and information-based process optimization are anonymized and yet patient-specific to realize. For this purpose, methods of, for example, multivariate regression analysis are used, but also other suitable methods. These procedures provide an interface between the post- and preoperative phases. The automated Linking preoperative knowledge with the intraoperative procedure under automated, postoperative evaluation of surgical procedures carries the risk of individual over-adjustments. The definition of data and model structures and the non-linear optimization is carried out according to the prior art, eg according to the work of or.net. The result is always dependent on the data set used and the structure chosen, with the risk of overfitting in particular. The medical conclusions derived from the automated evaluation must be validated and made available for all subsequent surgical procedures. This should be done on an expert system superimposed on the MIC process. From this system, the course of an intervention can also be planned and monitored.

The evaluation of an operative intervention can take place according to different aspects. Sometimes subjective impressions take a high priority. An objective, automated evaluation method is required, which forms the data base for the learning database.

When deriving recommendations for action from an expert system, it is imperative to identify correlations between relevant clinical process variables and a clinical outcome. The challenge is to derive causality from correlations. If this does not succeed, there is a risk of deriving incorrect recommendations for action from the experience and knowledge database.

The standard computer systems in use are currently characterized by a strict separation of the real and virtual worlds. Future MIC system solutions will be characterized by intelligent surgical units with networked system components, intelligent sensors and actuators. The widely used embedded systems today do not meet these requirements. A paradigm shift towards dedicated systems for these feedback and assistance functions is required.

According to the invention, an iteratively growing knowledge and experience database that can be available to the surgeon as an assistance system during any intervention in the area of minimally invasive surgery. This concept offers great potential for other areas in the clinical environment and is to be used for areas such as anesthesia, rehabilitation therapy or the field of dialysis.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The 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

• EP 1995679 [0007]

Claims (6)

Medical device for storing medical data, containing At least one computing unit, consisting of a selection and correlation module, a feed module with self-learning submodule, At least one memory unit which stores measurement data and / or patient data obtained via the interface and which optionally contains at least one data record as a fallback period, At least one interface to at least one other medical device, wherein the interface transmits measurement data and / or patient data to the memory unit, Optionally at least one interface to at least one hospital information system (patient file) and / or an image recognition system for reading patient information. Medical device according to Claim 1 , where the arithmetic unit contains • at least one system consisting of hardware and software for data communication for networking the devices in the OR, • at least one system consisting of hardware and software for data classification or grouping of patient data from the preoperative phase • at least one system consisting of hardware and software for identifying the medical device (s) used for the medical procedure, • at least one system consisting of hardware and software for data transmission to the medical devices used for the medical procedure, • at least one system consisting of hardware and software for Determination and transmission of setpoint recommendations based on a priori knowledge of the medical device, as well as optional • at least one system consisting of hardware and software for the identification of devices used in the preoperative phase • at least one system consisting of hardware and software for detection and transmission a characteristic set for setting the regulator on the basis of previously known a priori knowledge in the medical device • at least one system consisting of hardware and software for data transfer from the intraoperative phase to the database system • at least one system consisting of hardware and software for the determination and transmission of a set of characteristics for Controller setting on the basis of intraoperative data in the medical device • at least one system consisting of hardware and software for the determination and transmission of setpoint recommendations based on intraoperative data in the medical device • at least one system consisting of hardware and software for data transmission from the postoperative phase in the database system • at least one system consisting of hardware and software for incorporating rules derived from medical expertise or medical guidelines. Medical device according to Claim 1 or 2 in which the stored data is a mixture of anonymized patient data, anamnesis data, procedure and device data or device parameters, postoperative treatment evaluations and procedural rules. Medical device according to Claim 3 wherein the fluid pump is an insufflator or a fluid pump. Medical device according to at least one of the claims Claim 1 - 4 wherein at least one of the interface transmits measurement data of a blood pressure measurement device and / or patient data of an electronic patient record. Medical device according to at least one of the claims Claim 1 - 5 , wherein the operating parameters fluid pressure, fluid flow and / or fluid temperature are controlled.