EP4260327A1 - Procédé et dispositif de plate-forme de recherche pour traiter des interrogations de recherche adressées à une base de données contenant des données d'échantillon médical et/ou des échantillons - Google Patents

Procédé et dispositif de plate-forme de recherche pour traiter des interrogations de recherche adressées à une base de données contenant des données d'échantillon médical et/ou des échantillons

Info

Publication number
EP4260327A1
EP4260327A1 EP20835707.9A EP20835707A EP4260327A1 EP 4260327 A1 EP4260327 A1 EP 4260327A1 EP 20835707 A EP20835707 A EP 20835707A EP 4260327 A1 EP4260327 A1 EP 4260327A1
Authority
EP
European Patent Office
Prior art keywords
data
search
samples
avatar
sample
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
EP20835707.9A
Other languages
German (de)
English (en)
Inventor
Heiko Zimmermann
Andreas Kurtz
Antonie FUHR
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fraunhofer Gesellschaft zur Forderung der Angewandten Forschung eV
Universitaet des Saarlandes
Original Assignee
Fraunhofer Gesellschaft zur Forderung der Angewandten Forschung eV
Universitaet des Saarlandes
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fraunhofer Gesellschaft zur Forderung der Angewandten Forschung eV, Universitaet des Saarlandes filed Critical Fraunhofer Gesellschaft zur Forderung der Angewandten Forschung eV
Publication of EP4260327A1 publication Critical patent/EP4260327A1/fr
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/40—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for data related to laboratory analysis, e.g. patient specimen analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2457—Query processing with adaptation to user needs
    • G06F16/24578—Query processing with adaptation to user needs using ranking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3346—Query execution using probabilistic model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/338—Presentation of query results
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B50/00—ICT programming tools or database systems specially adapted for bioinformatics
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/20—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for electronic clinical trials or questionnaires
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H70/00—ICT specially adapted for the handling or processing of medical references
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Definitions

  • Method and search platform device for processing search requests to a database containing medical sample data and/or samples
  • the present invention relates to a method and a search platform device for processing search requests to a database that contains medical sample data and/or samples from a large number of samples from people (in particular patients).
  • the invention relates in particular to methods of application and devices for building an intelligent, self-learning platform that has a multi-directional mediator function between databases such. B. in laboratories, biobanks, and / or clinics, and the people whose sample data and / or samples are stored in the databases, and on the other hand researching entities, such. B. the pharmaceutical industry and (bio) medical research institutions.
  • the invention is used, for example, in the operation of databases with sample data and/or samples from a large number of samples from people and/or in the processing of search queries, in particular using machine learning and/or artificial intelligence (AI).
  • AI machine learning and/or artificial intelligence
  • a subject cohort includes subjects who have specific properties (parameters) for certain examinations, such as a certain common disease profile and/or certain genetic similarities.
  • Samples and data from subjects potentially suitable for forming a cohort are available in medical laboratory facilities and/or biobanks. So far, however, it has been extremely time-consuming for entities looking for test subjects to search through the various medical laboratories and biobanks for the required data and patients. There are also legal limits: The entities must not access personal data without authorization and directly, e.g. B. from patients access or use them.
  • search engines with special filter systems are available that help industry and research to find suitable patient samples, such as the commercial provider iSPECIMEN.
  • search engines can only work with standardized data in special databases and are not able to make predictive statements.
  • the requesting research institutions can only find those samples that are provided with identifiers (tagged) and for which a genome sequencing has already taken place.
  • the object of the invention is to provide an improved method for processing search requests to a database that contains sample data and/or samples from a large number of samples from people and/or a corresponding search platform device, with which disadvantages of conventional techniques are avoided.
  • the processing of search queries to a database should in particular make it possible to identify samples of interest with increased speed and/or reduced effort.
  • the processing of search queries to a database should furthermore in particular provide accelerated and/or simplified access to the sample data and/or samples and/or offer protection of the privacy of the persons to whom the sample data and/or samples belong.
  • a method for processing search queries to a database containing medical sample data and/or samples from a large number of samples from persons which comprises the steps of entering a search query with a requirements profile that the finding of at least one sample with sample data is directed, which have predetermined requested search parameters, in a AI processor device, search for at least one selected sample with the AI processor device, wherein the at least one selected sample meets the requirement profile, and output of identification data comprising at least one selected sample.
  • the database is z. B. provided by a medical institution, a laboratory and / or a sample bank of biological samples (biobank).
  • the requirement profile is preferably met at least with a predetermined probability, which is determined by the function of the AI processor device.
  • a search platform device configured to process search queries to a database containing medical sample data and/or samples from a plurality of samples from individuals, the search platform device having an input device that is set up to receive a search query with a profile of requirements that is aimed at finding at least one sample with sample data that have predetermined requested search parameters, an AI processor device that is coupled to the input device and set up to search for at least one selected sample , wherein the at least one selected sample meets the requirement profile, and an output device which is set up to output identification data of the at least one selected sample.
  • the search platform device or one of its embodiments is preferably set up to carry out the method according to the first general aspect of the invention or one of its embodiments.
  • the technique according to the invention advantageously offers more functions than a conventional search engine. It is a self-learning, predictive tool with built-in artificial intelligence.
  • This inventive technique can, for. B. support the pharmaceutical industry/(bio)medical research institutions in their search for suitable patient cohorts for studies and on the other hand database facilities, such as e.g. B. Laboratories/biobanks help to evaluate which of their samples are worth investing in, especially in terms of working time and costs, in sample processing, such as genome sequencing.
  • researching entities can have a direct influence on this evaluation by the learning artificial intelligence (AI) of the platform via their inquiries.
  • AI learning artificial intelligence
  • the queries of the researching entities are advantageously digitized and processed by the AI processor device.
  • the AI processor device includes a computer device that is capable of using a machine learning-based program, such as e.g. B. is set up by at least one artificial neural network.
  • the term AI processor device also refers to the artificial intelligence provided by the computer device.
  • Methods for setting up a class are described in WO 2020/233850 A1 (“Recursive coupling of artificially learning units”) and WO 2020/233851 A1 (“Coupling of several artificially learning units with a projection plane”), where in addition to building a learning system consisting of several AI units, the storage and independent application of ethical principles/rules by AI systems is also introduced.
  • the AI processor device learns through separate training and/or preferably by evaluating the queries directed to the AI processor device and the associated identification results from the database.
  • the AI processor device is used to search for a group of selected samples, with all selected samples meeting the requirements profile at least with the specified probability.
  • the group of selected samples advantageously provides a limited number of samples that can be further processed or examined or output for the respective query, which considerably simplifies the effort involved in operating the database.
  • the method according to the invention and the search platform device preferably fulfills at least one of the following two functions: a) Independent development of a ranking of the samples according to relevance for the research b) Independent evaluation (rating) of the parameters that are of interest for the research (which data of the patient are relevant for research)
  • the identification data of the group of selected samples are preferably output with ranking information (placement information), which assigns each of the selected samples a hit probability with which the requirements profile is met.
  • additional rating search parameters are preferably output that match the search query in such a way that a group of samples can be selected that meet the requirement profile with an increased hit probability.
  • the rating search parameters particularly preferably include a genetic profile, data on clinical treatments, previous illnesses, family predispositions to illnesses, personal diagnostic results, characteristics of lifestyle habits, eating habits, consumer behavior, sports and exercise characteristics, data on the consumption of drugs or other intoxicants, data on medication intake, Radiation exposure data, epigenetic data, geographic information, age, gender, ethnicity, allergies and/or mental illness.
  • the requirement profile of the researching entities is aimed at finding a group of samples with sample data, with the persons belonging to the group of samples forming a test person cohort.
  • the test person cohort can thus be generated directly in response to the query.
  • the AI processor device processes information from specialist literature, commercial market information from the industrial sector, in particular relating to pharmaceutical products, and/or information from approval databases relating to pharmaceutical products.
  • the platform thus advantageously forms a search portal for the researching entities of high practical value for the search for suitable patient cohorts by forwarding the search queries to the laboratories/biobanks.
  • the inquiries z. B. the pharmaceutical industry uses the cl the platform preferably to create the sample ranking. In order to create this ranking, the platform's clerk can also observe the scientific literature in addition to inquiries from the pharmaceutical industry. Not only pharmaceutical companies can be connected to suitable sample material via this platform, academic, non-profit research institutions can also use this platform. In turn, they send information to the platform about which branches of research will be trending. In addition to the scientific publications, information from the industrial sector could also be of interest here: for example, which drugs are coming onto the market as new products, or which trends in pharmaceutical research can be read from the approval databases.
  • a conversion of the search query with a coding function into a fragmented search query is preferably provided.
  • the conversion preferably takes place using a coding device which is coupled to the AI processor device and the input device.
  • the personal information of people such as e.g. B. clinical data or sequencing data of their genome must not be freely accessible. You are subject to data protection. Nevertheless, a search for specific characteristics of the patient profile may be desirable so that the platform can work optimally and mediate between the two sides. This search can be carried out by using the coding function, for example with a hash function: The platform itself has no insight into a patient's genetic data, but can search within the databases of laboratories via hashes in the encrypted data for suitable genetic profiles for the researching entities are looking for without knowing the concrete genetic information.
  • the coding function can be used in particular in accordance with the procedure described in DE 10 2019 135 380.7 ("Method and data processing device for processing genetic data", unpublished on the priority date of the present disclosure), in which in an encrypted manner (e.g. via hashes) inaccessible data (such as the genetic information of a person) can be searched without the entire information of the DE 10 2019 135 380.7 is incorporated by reference into the present disclosure, in particular with regard to the processing of genetic data.
  • in an encrypted manner e.g. via hashes
  • inaccessible data such as the genetic information of a person
  • “Secure enclaves” are preferably installed in the databases for this type of application, via which the platform uses the coding function, e.g. B. the hashes, which can forward inquiries from the pharmaceutical industry/research institutes regarding certain samples.
  • the database such as B. the laboratory/biobank, confirms or denies the existence of matching samples without disclosing the underlying personal data. In this way, it is also not necessary for the platform to have personal patient data or a central ID system for the individual samples/patients. Instead, each currently relevant sample is stored decentrally and temporarily in an encrypted way (hashed) (as with the disease warning app).
  • the device does not store the specific genetic data of specific samples, patient clinical data, or other patient profile information. This personal data remains with the laboratories/biobanks or with doctors and patients.
  • the platform forwards the search in the form of requests transformed with the encoding function, in particular with "hashes", so that it does not need to get direct access to this personal data.
  • a request for consent to the use of data and/or samples by the associated persons is provided.
  • the request for consent is particularly preferably answered using questionnaires and/or informed consent documents and/or by an avatar device.
  • information can preferably be output, in particular to persons whose sample data are stored in the database, with the information output particularly preferably relating to pharmaceutical products, diseases, research results and/or recommendations for action.
  • FIG. 1 an overview of a search platform device and a search method with features according to preferred embodiments of the invention
  • FIG. 2 further details of the functions of the search platform device and the search method according to preferred embodiments of the invention
  • FIG. 3 is an illustration of advantages of using the invention in practice
  • FIG. 4 an overview representation of an avatar device with features according to preferred embodiments of the invention
  • FIG. 5 an illustration of the communication and the control of the communication between the avatar device and its environment (outside world);
  • FIG. 6 an illustration of the use of the first and second memory devices.
  • FIG. 7 an overview of an avatar method with features according to preferred embodiments of the invention.
  • a AI processor device can be configured in particular as described in WO 2020/233850 A1 ("Recursive coupling of artificially learning units") and WO 2020/233851 A1 ("Coupling of several artificially learning units with a projection plane").
  • WO 2020/233850 A1 and WO 2020/233851 A1 in particular with regard to the configuration the AI processor device and its peripherals, such as B. input and output devices and memory, are incorporated by reference into the present disclosure.
  • FIG. 1 illustrates the interaction of the AI processor device of the search platform device with research facilities, licensing authorities and scientific institutions.
  • the Kl processor device which is coupled to input and output devices (not shown), learns both through inquiries from the pharmaceutical industry or from other research institutions and through access to current scientific publications, new drugs and / or information about samples, which for the research are of particular interest and are therefore given which weighting.
  • the AI processor also learns in the same way to judge which parameters are important for research, such as e.g. B. Characteristics of the genome, age, gender, lifestyle, previous illnesses, etc. of the respective persons.
  • the requesting entities influence the weighting in terms of ranking and rating with their requests.
  • FIG. 2 illustrates the interaction of the AI processor device of the search platform device with databases such.
  • the AI processor device can provide information (identification information) for laboratories and biobanks as to which samples are likely to be worth genome sequencing and which parameters are of interest to the researching entities. Patients or doctors can also find out directly from the search platform device which research focuses currently exist.
  • the communication between the search platform device and the database can take place completely anonymously and/or using coding functions, in particular hash functions, via secure enclaves within the laboratories/biobanks.
  • An ethical avatar described below with reference to Figures 4 to 7, can be used as a digital representative of people to assist the search platform device in answering queries (predicting suitable samples), for example to determine whether a particular patient is available for study participation.
  • the search platform device is able to develop a ranking of samples (e.g. from "interesting” to "uninteresting” or using probabilities of suitability) and to identify parameters that are of interest for the research.
  • parameters can include, for example, the patient’s genetic profile, data on their clinical treatment, previous illnesses and family predispositions to illnesses, personal diagnostic results, lifestyle, eating habits, consumer behavior, sport and exercise, drug use, medication intake, possible radiation exposure, epigenetic data, geographic information, age , gender, ethnicity, allergies and/or mental illness.
  • the search platform device helps laboratories/biobanks both in assessing which of their samples have great current value for research and in evaluating those parameters that will be of interest for studies and which are also for the laboratories/biobanks to collect. ken could be worthwhile (see under Figure 2).
  • the voluntary consent of the patients concerned may be required for the use of data/samples from patients.
  • the ethical avatar could be used.
  • This avatar is an intelligent tool that has learned and can simulate real person's ethical attitudes.
  • the platform would be able to make a pre-selection regarding the ethical background of certain data/samples for the pharmaceutical industry: instead of having to contact real people to ask their attitudes to certain application and use options of their data/samples , a much faster pre-selection could be made by asking their ethical avatars.
  • the platform not only helps the pharmaceutical companies (or non-profit scientific institutions) in the search for patients with a suitable genetic and medical profile, but also makes a pre-selection of these patients who have a high probability of being willing to participate in R&D (Research and Development).
  • the ethical avatar acts as a predictor of the patient's willingness to participate in studies, derived from their personal values or religious beliefs or cultural background.
  • the prediction as to whether a patient is fundamentally open to participating in a study can also be made without an "ethical avatar" by evaluating corresponding questionnaires or informed consent documents by the AI of the platform.
  • Efficient operation of the search platform device gives the patient the opportunity to take part in studies that could be important to him (for example, because they relate to his illness).
  • the platform could also connect the patient to important sources of information, including medications they are taking, new research results about their condition, etc.
  • the AI processor device of the search platform device processes large amounts of data (requests from research institutions, current scientific publications, approval processes, current drug innovations), evaluates this data and is thus able to make predictions as to which samples could be relevant for research .
  • These prognostic results are of great value to the labs/biobanks, as they can assess which samples they have at their disposal are worth expensive full genome sequencing.
  • sequenced samples are of value for the pharmaceutical industry and (bio)medical research institutions, so that it is of great importance for laboratories/biobanks to sequence those samples that are later also purchased by research institutions.
  • the platform helps to mediate between laboratories/biobanks and potential sample buyers.
  • the platform supports research institutions in putting together suitable patient cohorts by forwarding requests for specific samples from research institutions to the relevant laboratories/biobanks. Through their inquiries, the research institutions have an influence both on the weighting within the sample ranking and on the assessment of the platform's Kl, which sample parameters (e.g. the patient's radiation exposure, previous illnesses, drug abuse, etc.) should be considered interesting.
  • the research institutions can use the platform to influence the sequencing behavior of the laboratories/biobanks and their focus on specific sample parameters for their own purposes.
  • FIG. 3 illustrates the advantages of using the search platform device using a business plan.
  • Users of the search platform device from the pharmaceutical industry pay a certain amount of money for the application, e.g. B. EUR 100,000 per user.
  • the platform forwards the user to laboratories/biobanks with samples that correspond to a requirement profile requested by the user, and the weighting of the sample ranking is influenced. This ranking has a direct impact on which samples are sequenced by laboratories/biobanks.
  • the search platform device can share in the sequencing costs. Scientific, non-profit research directions can use the platform free of charge.
  • the AI of the search platform device benefits from the inquiries because, based on this, it can adapt its ranking to the current research interest. Patients who participate in studies or who are willing to share personal data can be connected to relevant networks in return, e.g. B. relate to their illness.
  • a genome sequencing costs laboratories/biobanks around €500. These costs are only recorded by laboratories/biobanks for approx. 1% of their samples. Performing cruder sequencing methods (such as SNPs) or diagnostic tests cost around €50 each. Through the prediction of the search platform device, the laboratories/biobanks receive valuable information as to which samples could be relevant for research institutions, so that sequencing can lead to success more quickly and sequencing costs can be used more efficiently.
  • Inquiring pharmaceutical companies or other commercial organizations have to pay money (in the form of a "participation fee") into a fund for inquiries to the platform, because every inquiry means influencing the ranking of the platform and the evaluation of parameters.
  • Non-profit institutions such as universities or other non-profit (bio)medical research institutions can inquire free of charge. The value of these requests for the platform lies in the information they contribute: with each request, the platform's AI system learns which samples are currently in the research trend and can adjust its ranking and thus its predictions.
  • Inquiring laboratories/biobanks receive financial support from the platform's money fund if they sequence certain samples that are rated as particularly relevant to research and carry out certain diagnostics that provide parameters that are interesting for research. In this way, laboratories/biobanks are motivated to sequence certain samples, which in turn is in the interest of the requesting pharmaceutical industry/(bio)medical research institutions, which can thus compile attractive sets of samples.
  • the money in the fund of the platform is used to operate the platform (see Figure 3 for an overview of the business plan).
  • the present invention also relates to an avatar device ("ethical avatar”) set up to represent a person and to process personal data of the person, an avatar system with multiple avatar devices, applications of the avatar device and an avatar Method of Operating the Avatar Device.
  • the invention relates in particular to a method for developing and an apparatus for building a virtual representative of a person, which is also referred to herein as an ethical avatar or as the avatar device, the avatar device being, for example, individual moral attitudes, ethical principles, values and /or can learn other opinions and interests of the people and independently simulate them.
  • Applications of the invention are given, for example, in the operation of databases with personal data of persons and/or in the processing of personal data, in particular by means of machine learning and/or artificial intelligence (AI) and/or the processing of inquiries to the search platform described above.
  • Device given according to the invention are given according to the invention.
  • the object of this partial aspect of the invention is to provide an improved data-based, digital representation of a person and/or methods for its application, with the disadvantages of conventional techniques being avoided.
  • the digital representation of a person should in particular enable additional information and/or information in extended categories, such as e.g. B. provide non-physiological information about the person and / or offer advanced applications.
  • the digital representation of a person should in particular provide accelerated and/or simplified access to statements made by the person and/or offer protection of the person's private sphere.
  • an avatar device for this purpose, an avatar device, an avatar system with several avatar devices, applications of the avatar device and an avatar method for operating the avatar device with the features described below are used according to the invention.
  • An avatar device is preferably used, which is set up to represent a person and to process personal data of the person.
  • the avatar device includes an AI processor device for machine learning of individual mental Response characteristics of the person is set up by means of data training, the Kl processor being configured to generate semantic answers (in particular at least one sentence or individual words, such as yes or no) in response to questions depending on the mental response characteristics, a first Storage means (personal storage) coupled to the AI processor means and arranged to store the individual's mental response characteristics, an input device coupled to the AI processor means and for interacting with the individual and/or receiving questions regarding the individual is set up, and an output device which is set up to output signals which represent the responses generated by the AI processor device.
  • an AI processor device for machine learning of individual mental Response characteristics of the person is set up by means of data training
  • the Kl processor being configured to generate semantic answers (in particular at least one sentence or individual words, such as yes or no) in response to questions depending on the mental response characteristics
  • a first Storage means personal storage
  • an input device coupled to the AI
  • an application of the avatar device or one of its embodiments comprising at least one of forming a multidirectional platform for connecting medical laboratories and/or biobanks with industrial companies and/or scientific research institutions, forming a Platform for social research, opinion research and/or big data analysis, creation of a platform for the provision of patient declarations, in particular declarations of consent for the use of personal data, creation of a data protection platform, generation of a living will for the person, provision of input information for a Synthetic biology processes and control of technical systems.
  • an avatar method for operating the avatar device of a person comprising the steps of receiving at least one question relating to the person, processing the question with the AI processor device, and outputting a semantic answer to the at least one question .
  • an avatar system comprising a plurality of avatar devices, each avatar device being adapted for a different application.
  • the AI processor device includes a computer device that is used to apply a machine learning-based program, e.g. B. is set up by at least one artificial neural network.
  • a machine learning-based program e.g. B.
  • the term AI processor device also refers to the artificial intelligence provided by the computer device.
  • the avatar device advantageously provides a digital twin of the person to whom the avatar device relates, the avatar device e.g. B. represents individual opinions, interests and / or moral attitudes of the real person.
  • the avatar device makes it possible to simplify and/or speed up the management of personal data.
  • the avatar device allows to protect the privacy of the data subject and to take their interests into account as much as possible.
  • the avatar device advantageously represents an intelligent, digital and predictively usable tool that can act as a representative of real people.
  • the avatar device can be used, for example, as an advisory link between people on the one hand and requesting entities, such as research facilities or institutions or companies, on the other hand, while the anonymity of the person concerned is preserved (direct contact with the person is no longer necessary ) and at the same time the interests of the person can be represented towards others.
  • the avatar device is a tool that can be used to predict statements, especially opinions, of its real counterpart and in all types of databases/analysis where previously thousands of people had to be interviewed in a time-consuming and costly manner.
  • the ethical avatar represents an artificial intelligence (AI) that is able to be trained by people, in particular with regard to their moral attitude, and then to anticipate answers to new questions on the basis of this training.
  • AI artificial intelligence
  • the avatar class learns about the person's moral attitudes, values, interests, and principles.
  • associative class With the help of a learning, associative class, he can now apply these principles/values/attitudes when it comes to answering new (ethical) questions.
  • the avatar device is thus not only able to reproduce what it has learned, but also to apply what it has learned in new contexts in an intelligent and creative manner. She can intelligently predict, based on her knowledge of the person, what that person would mean and judge in a given new context if asked.
  • the AI processor device is set up for machine learning of the individual mental response characteristics, which comprise the person's moral attitudes, attitudes, opinions, values, interests and/or ethical principles.
  • the response characteristics include subjective characteristics of the person, which characterize how the person reacts to a question, fact, experience, situation, another person, a social group and/or an idea, where reactions e.g. B. can include the articulation of statements, affects and/or behaviors.
  • the first storage device is configured to store the mental reaction characteristics in the form of rule catalogues, micro-contracts and/or blockchains.
  • the implementation of the invention is not limited to these variants, but is possible with other digitizable formats.
  • the first memory device is preferably set up to store individual additional data on which the mental reaction characteristics depend and which include in particular a religious affiliation, a group affiliation, a club membership and/or a party membership of the person.
  • the AI processor device is set up to understand and use ethically relevant terms.
  • a second memory device (public memory, shared memory) is preferably provided, which is set up to store general additional data which represent knowledge relevant to the mental reaction characteristics.
  • the second memory device is particularly preferably coupled to an update device with which the general additional data can be renewed.
  • the avatar device in particular the input device, has a communication channel which is set up for communication between the avatar device and the person.
  • this allows the person to access the avatar device, for example to test its function, train the AI processor, or change stored data.
  • the avatar device is thus not only trained by the person in question, but the real person whose moral attitudes it simulates also preferably has the opportunity at any time to check and, if necessary, correct the statements, in particular decisions, of their ethical avatar to train him further, or to make changes to stored values/interests/moral principles.
  • the person also has the option of transferring the decision-making power entirely to their avatar and having them represent them completely.
  • the real person's autonomy is not restricted, since the avatar merely acts as an assistant representative, under the person's control at all times and based only on the mental response characteristics, such as e.g. B. individual opinions and interests, the person takes on advisory functions.
  • the input device is provided with an authorization device which is set up to direct the receipt of questions relating to the person to predetermined authorized entities, such as e.g. B. pharmaceutical companies and / or research institutions, and / or to predetermined permissible content.
  • the authorization device particularly preferably comprises a communication authentication memory and a standard communication protocol function.
  • the avatar does not allow access to the data by unauthorized persons and, even when answering questions asked of him, only provides personal data to a limited extent and only to authorized entities .
  • the avatar device further includes a cryptographic memory configured to store access keys.
  • a cryptographic memory configured to store access keys.
  • the avatar device has a log memory which is set up to store a history of changes in the contents of the first and/or second memory devices.
  • a timer device can advantageously be provided, which is set up to provide a time scale for the functions of the avatar device.
  • the avatar device particularly preferably comprises a communication device which is set up for contacting and/or communicating with the person.
  • z. B. enables verification of the semantic response supplied by the AI processor device.
  • FIG. 4 shows an embodiment of the avatar device and its mode of operation, the sections (steps and/or components) a to g shown in FIG. 4 having the following meanings.
  • Section a includes storing what has been learned and/or accessing what has been learned
  • section b includes logging the times of the training units
  • section c is access to the second storage device in order to be able to design a survey more intelligently and with a timer device (clock system) to compare
  • section d includes an interaction between key management and authentication memory
  • section e it is determined who is entitled to ask questions and/or information
  • section f includes influencing the activation or prevention of communication with the environment, in particular a communication protocol
  • section g refers to a channel for training the class by a person, a questioning of the class and/or information from the class.
  • the communication (see FIG. 7) and control of the communication between the avatar device and its environment (outside world, in particular requesting entities or real person training the avatar) comprises the following steps.
  • Step a is a com- communication of the real person.
  • the AI processor device verifies, using its cryptographic memory, whether the person is authorized to make changes to and/or communicate with the personal memory.
  • step c decisions are made yes/no to allow training and/or yes/no to provide information.
  • Step d is an incoming request on a predetermined topic from an entity such as B. a research institution.
  • step e the avatar's AI processor device uses a directory of all authorized entities in its communication authentication memory to check whether it is entitled to receive information from this entity and on which topics it is entitled to provide information to it, in particular a protocol of communication in the communication authentication -Memory is used.
  • step f the result of the check is delivered.
  • FIG. 6 illustrates the storage of data and the retrieval of data from the first storage device (personal storage) in combination with the use of the second storage device (shared public storage) for chronological classification of the training content.
  • the AI processor of the ethical avatar is not only able to store the current mental reaction characteristics, e.g. a real person's moral attitudes/opinions/interests (if authorized requests come, see Figure 2), but can also store older versions of the real person's mental response characteristics in a historical log and timed via a clock system classify absolutely. This chronological assignment also makes a comparison with the second memory device possible.
  • the second memory device can also use the AI processor device to make the training by the real person more intelligent, for example by the AI processor device asking about the denomination of the person and understanding the answer "Christian" as access to the second memory device, since which captured what "being a Christian” means in relation to moral attitudes at a given point in time.
  • the avatar device has the following characteristics illustrated in Figure 4:
  • the avatar device includes (1) a learning component that can learn moral attitudes from real people via personal training, wherein
  • the results of this training are stored in the first memory device and the AI of the avatar can access this information (the first memory device is not public, ie access from the environment of the avatar device is blocked).
  • the mental response characteristics such as B. moral beliefs, the person can be stored in the form of rule catalogues, micro-contracts and / or blockchains in the first memory device of the avatar.).
  • Information about the person’s religious affiliation or group/club/party memberships can also be stored, from which the ethical avatar is able to draw conclusions about the person’s opinions (by accessing a separate memory (second memory facility, see (4)).
  • the avatar device is set up to respond automatically to new questions (input) on the basis of its stored knowledge of the mental reaction characteristics, such as e.g. B. Attitudes to generate new answers (output) for the person.
  • the mental reaction characteristics such as e.g. B. Attitudes to generate new answers (output) for the person.
  • Such an avatar is preferably able to understand and apply ethically relevant terms, which is made possible by a corresponding ontology (in particular systematic, formalized representation of entities, such as concrete and abstract objects, properties, facts, events, processes). becomes.
  • the avatar (4) can preferably access a separate, second memory device in which (through previous professional training) more fundamental ones for the reaction characteristics, such as e.g. B. Morality, relevant knowledge is stored.
  • This can, for example, involve knowledge about various ethical theories and their implications, about religions or other groups (such as parties or associations) and their codes and rules, but also about knowledge about ethically relevant terms and ethical dilemmas known and discussed in moral philosophy. situations.
  • This second storage facility public storage, e.g.
  • the avatar device could be used by the avatar device during its training by individuals to make the individual's interrogation more intelligent, to make the individual aware of certain questions, and to categorize the individual's responses to be able to
  • the avatar only uses this general knowledge during the training, but not when he later answers ethical questions on his own behalf. In this case, he makes decisions based only on what he knows about their mental response characteristics, particularly individual values/interests/principles.
  • the second storage device can receive regular updates, for which a clock system within the avatar (see (9)) is used.
  • a communication channel is provided between the ethical avatar and its real counterpart, the person, so that the latter can train the avatar and preferably control the actions of their digital representative at any time or re-initiate the training if they so wish.
  • Communication with requesting entities is also made possible, but is preferably subject to a limitation.
  • Authorization takes place • (6) via a communication authentication memory including standard communication protocol function. This is a directory of those entities with which the avatar is allowed to communicate, combined with conditions about which context the questions of the requesting entity may relate to in order to be answered.
  • the authorizations could be given either by the real person himself or by a central institution. Only authorized authorities may be permitted information within a clearly defined framework: For example, an authorized medical institution may question the avatar, but only questions with a medical connection are allowed (representation of the communication between the avatar and the outside world in Figure 5).
  • the avatar can preferably check which instance has which information rights and to what extent. These keys also ensure that only the real person is authorized to make changes to the avatar's personal storage.
  • the avatar (8) preferably contains a memory (protocol memory) in which the historical course the changes are stored (e.g. via blockchains or similar).
  • the avatar has
  • Both codes are stored in the second memory device (4) linked to the respective points in time, as are the training units in the personal memory (2) via the clock system with the training point in time be stored linked.
  • This enables an absolute chronological classification of the personal attitudes of the person (representation in FIG. 6).
  • the ethical avatar can (10) have a unit that relates to the basic accessibility of the real person, links a contact option, or provides information about the person's ability to be contacted.
  • the real person has exactly one corresponding avatar device (ethical avatar), preferably with a corresponding identification tag, to which inquiries can be sent by various entities. All information about this person is processed via an avatar device.
  • avatar device ethical avatar
  • a corresponding identification tag e.g., a corresponding identification tag
  • All information about this person is processed via an avatar device.
  • avatar system multiple avatar devices (avatar system), generated in different application areas, can be assigned to a real person.
  • the successful synchronization of different avatars to an individual could either be via an avatar identification number (provided by an avatar inquiry authority), a global unique identifier (like the patient ID or the biosample ID) or a (hashed) individual identifier (like a picture [ex developed the concept of a "gravatar” - a gravatar is an image that accompanies the user from website to website and acts as an identifier (see: https://de.gravatar.com).] or genetic "fingerprints" such as SNPs - single nucleotide polymorphisms - or STRs - short tandem repeats) If a person has several avatars, these avatars are preferably synchronized with one another so that all avatars involved are up to date with regard to the mental reaction contents, e.g. attitudes and opinions, of their common real person A comparison between the different avatars of a real person should be guaranteed.
  • an avatar identification number provided by an avatar inquiry authority
  • a global unique identifier like the patient ID
  • the platform can support both sides: It advises the laboratories on which samples/data would be worth expensive full sequencing of the genome (necessary for research) (because it knows which samples/data are of interest for research/economy) , and can help researchers put together suitable cohorts.
  • the patients' ethical avatars help with this pre-selection, also because they know the moral attitudes of their real counterparts.
  • the consent of the affected patients is required for the use of data/samples from patients. Ethical avatars could be used to make it easier to obtain these explanations (see below).
  • Social science research Avatars could help with the pre-selection of which people could be interested in which research topics and survey content, or which people would object immediately
  • the person concerned can rely on the fact that in the event of their own inability, the ethical avatar trained by them can anticipate decisions in their interest Unforeseen future research areas/areas of application: With a broader training of the avatar, in addition to specific questions of "Informed Consent " also learn more general moral attitudes/value concepts/principles of the person concerned and in future even anticipate answers to questions that go beyond those of "informed consent”.
  • the avatar would have access to the corresponding hashes and could answer yes/no to questions that arose without having to know the encrypted information.
  • This power would be particularly useful in the area of compiling patient/subject cohorts with certain common characteristics. Often z.
  • cell lines are required whose donors have specific, for example genetic, similarities.
  • the respective avatars could, upon request, confirm or not confirm the suitability of the cell lines whose donors they represent, without revealing the genetic information of their real-life counterpart.
  • the search for suitable patients/subjects could be done very quickly (answering questions about a person's specific genetic characteristics), and the avatar could also help with the pre-selection without revealing the underlying sensitive data of the corresponding persons.
  • the driver does not have to submit to the dictum of average moral standards, but can implement their personal ethical values on the road (of course within the framework of the legal framework) (just as they would do if they drove themselves).
  • the responsibility of the person is supported by the help of the ethical avatar, the key would be managed and stored professionally, a loss is unlikely Monitoring of data protection:
  • the avatar represents a link in the communication between people and (medical) facilities or research institutions .

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Theoretical Computer Science (AREA)
  • Public Health (AREA)
  • Physics & Mathematics (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Business, Economics & Management (AREA)
  • Business, Economics & Management (AREA)
  • Biophysics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Pathology (AREA)
  • Bioethics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Biotechnology (AREA)
  • Evolutionary Biology (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

Un procédé de traitement d'interrogations de recherche adressées à une base de données contenant des données d'échantillon médical et/ou des échantillons provenant d'une pluralité d'échantillons de personnes comprend les étapes consistant à entrer une interrogation de recherche avec un profil de requête visant à extraire au moins un échantillon avec des données d'échantillon qui ont des paramètres de recherche interrogés prédéterminés dans un dispositif de processeur AI, rechercher au moins un échantillon sélectionné à l'aide du dispositif de processeur AI, l'échantillon ou les échantillons sélectionnés satisfaisant le profil de requête au moins avec une probabilité donnée, et délivrer en sortie des données d'identification pour l'échantillon ou les échantillons sélectionnés. L'invention concerne également un dispositif de plateforme de recherche configuré pour traiter des interrogations de recherche adressées à une base de données contenant des données d'échantillon médical et/ou des échantillons provenant d'une pluralité d'échantillons de personnes.
EP20835707.9A 2020-12-11 2020-12-11 Procédé et dispositif de plate-forme de recherche pour traiter des interrogations de recherche adressées à une base de données contenant des données d'échantillon médical et/ou des échantillons Pending EP4260327A1 (fr)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/EP2020/085818 WO2022122172A1 (fr) 2020-12-11 2020-12-11 Procédé et dispositif de plate-forme de recherche pour traiter des interrogations de recherche adressées à une base de données contenant des données d'échantillon médical et/ou des échantillons

Publications (1)

Publication Number Publication Date
EP4260327A1 true EP4260327A1 (fr) 2023-10-18

Family

ID=74125148

Family Applications (1)

Application Number Title Priority Date Filing Date
EP20835707.9A Pending EP4260327A1 (fr) 2020-12-11 2020-12-11 Procédé et dispositif de plate-forme de recherche pour traiter des interrogations de recherche adressées à une base de données contenant des données d'échantillon médical et/ou des échantillons

Country Status (6)

Country Link
US (1) US20240004887A1 (fr)
EP (1) EP4260327A1 (fr)
JP (1) JP2024501194A (fr)
KR (1) KR20230156302A (fr)
CN (1) CN116648701A (fr)
WO (1) WO2022122172A1 (fr)

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10796781B2 (en) * 2015-12-07 2020-10-06 Clarapath, Inc. Spatial genomics with co-registered histology
WO2020233851A1 (fr) 2019-05-21 2020-11-26 Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. Couplage de plusieurs unités à apprentissage artificiel avec un plan de projection

Also Published As

Publication number Publication date
KR20230156302A (ko) 2023-11-14
WO2022122172A1 (fr) 2022-06-16
CN116648701A (zh) 2023-08-25
JP2024501194A (ja) 2024-01-11
US20240004887A1 (en) 2024-01-04

Similar Documents

Publication Publication Date Title
Wang When artificial intelligence meets educational leaders’ data-informed decision-making: A cautionary tale
Easterby-Smith et al. Absorptive capacity: A process perspective
Sun et al. Positive youth development, life satisfaction and problem behaviour among Chinese adolescents in Hong Kong: A replication
Halpern-Manners et al. Panel conditioning in the general social survey
Wagner et al. Enhancing the power of household panel studies: The case of the German Socio-Economic Panel Study (SOEP)
van De Ven et al. Alcohol and other drug (AOD) staffing and their workplace: examining the relationship between clinician and organisational workforce characteristics and treatment outcomes in the AOD field
Frank et al. Developments in qualitative mindfulness practice research: a pilot scoping review
Lewis et al. Relocation, realignment and standardisation: Circuits of translation in Huntington’s disease
Arnout et al. Ethnographic research method for psychological and medical studies in light of COVID‐19 pandemic outbreak: Theoretical approach
Omotayo et al. Adoption and use of electronic voting system as an option towards credible elections in Nigeria
Savage Elizabeth Bott and the formation of modern British sociology
Mahoney et al. Self‐and cross‐citations in the Journal of Applied Behavior Analysis and the Journal of the Experimental Analysis of Behavior: 2004‐2018
Laurence Inside the child's head: Histories of childhood behavioural disorders
Beckingham Bureaucracy, case geography and the governance of the inebriate in Scotland (1898–1918)
Mathlin et al. Factors associated with successful reintegration for male offenders: a systematic narrative review with implicit causal model
EP4260327A1 (fr) Procédé et dispositif de plate-forme de recherche pour traiter des interrogations de recherche adressées à une base de données contenant des données d'échantillon médical et/ou des échantillons
Nishimoto et al. Career stage‐specific predictors of reflective ability among clinical nurses
Townsend The moderating role of social support on the relationship of perceived stress and life satisfaction of psychology graduate students
WO2022122173A1 (fr) Appareil d'avatar et procédé de représentation d'une personne et de traitement de données à caractère personnel de la personne
Holt et al. The criminogenic family: families as the cause of crime in research and policy
Cerezo et al. Queering psychology research methods
Bader Qualitative Research Methods and Clinical Practice Techniques for Social Work
Schwab et al. Biobanks and the human microbiome
Somit et al. Political Science and Biology: Then, Now, and Next
Erdös School of Business and Economics Vrije Universiteit Amsterdam

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20230613

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)