Method and system for collecting, representing, and evaluating personalised health data

Abstract

The invention relates to a method and a system based thereon for representing the individual personalised health and health-related data of at least one person. A digital 2D and / or 3D representation of the physical body of an individual person is used to explore the body of said person analogously to a landscape in order to derive conclusions regarding their state of health and / or their state of fitness, and recommendations are provided for treatment and / or improvement of these states. To this end, cluster-based analysis data from anonymised personal data of the system is used, obtained from a large amount of comparative data and normative data. Data volumes are employed in a targeted manner through the application of zooming in order to explore and treat the physical body of a person, and to create personalised behavioural recommendations and risk analyses.

Description

[0001] Method and system for recording, displaying and evaluating personalized

[0002] Health data

[0003] Short description

[0004] The invention relates to a method and a system based thereon for displaying the individual personalized health and health-related data of at least one person.

[0005] State of the art

[0006] Systems, methods, and modules are known for collecting health measurement data and processing this data to generate alerts and modify patient treatment plans. A system may be configured to receive data defining, for example, a plurality of health sensors and, based on this data, generate application logic that, in turn, is configured to cause a user device, in response to user input, to couple with the plurality of health sensors, thereby establishing communication between the user device and the health sensors.

[0007] A variety of recording and monitoring systems are also known, such as smartwatches, fitness watches, fitness trackers, apps, etc.

[0008] These can, for example, display information, inform the user through a vibration alarm or similar acoustic and / or visual functions, and detect the current situation via sensors. This allows the user to be monitored and / or supported through self-monitoring and situation-specific support, and through interactions, execute commands to control other devices. Smartwatches can typically connect wirelessly via Bluetooth to a smartphone with a suitable app and display and evaluate the data. Smartwatches are typically products for private users, but can also be used in industrial environments. Their areas of application are diverse and can generally be divided into sports, healthcare sectors, and advanced smartphone functions.

[0009] Certain smartwatches also create personalized training plans tailored to the user based on the extensive data collected. Routes and training plans for outdoor sports can be planned in advance and used as a navigation aid with GPS, or recorded and analyzed during the activity.

[0010] The sensors enable the detection of abnormalities, allowing applications for elderly people (fall detection, assistance, epilepsy), at-risk workplaces, or for medical applications. Smartwatches are also used to detect cardiac arrhythmias, such as atrial fibrillation. There are also models that can measure blood pressure. Some of these so-called smartwatches, or even separate devices, also offer a pulse oximeter measurement of blood oxygen saturation, which can also be recorded and displayed via an app. Nowadays, anyone can automatically record steps taken, heart rate, blood sugar, and much more using home tests or their smartphone and apps.

[0011] Independent of such data, their systematic collection, and analysis, health data exists on almost every person, which doctors hold in analog and / or digital health records. This personalized data on health, illnesses, monitoring results such as ECGs or X-rays, medication intake, vaccinations, etc., is available independently of and detached from personalized data from the aforementioned smartwatches or similar systems. Other sources of personalized data from different healthcare institutions are often not connected. This also includes general, person-specific behaviors such as eating habits, the individual's emotional state, the body's reactions to weather conditions, climatic changes, and / or changes of location due to travel, relocation, etc., are generally not recorded or are only available individually.

[0012] Task of the invention and solution

[0013] The object of the invention is to provide a method and system for collecting, displaying, and evaluating personalized data. A further object of the invention is to develop a system based on existing general, anonymized health data that can provide behavioral recommendations for the individual.

[0014] This problem is solved by a method, a system, and a computer program product having the features of the independent claims. Advantageous embodiments and further developments are specified in dependent claims.

[0015] Using a digital 2D and / or 3D representation of an individual's physical body, their body can be explored analogously to a landscape, conclusions can be drawn about their health and / or fitness levels, and recommendations can be made for treating and / or improving these conditions. Cluster-related analysis data from anonymized personal data in the system is also used for this purpose.

[0016] The system's cluster-based analysis data, derived from anonymized personal data, is created from a variety of comparative and normative data. By applying a method for zooming in on a region of interest, data sets are used in a targeted manner to research and treat a person's physical body, and to develop behavioral recommendations and risk analyses.

[0017] Further advantages and features of the invention will become apparent from the following description, in which exemplary embodiments of the invention are explained in detail with reference to the drawings. They show:

[0018] Fig. 1 shows a two-dimensional representation of health data in the form of health data elements of different persons, the representation also illustrating relationships between the health data elements;

[0019] Fig.2 a two-dimensional representation of health data from different

[0020] Persons in the form of differently colored and clustered health data elements; and

[0021] Fig. 3 shows a three-dimensional representation of health data of different individuals in the form of health data elements grouped in clusters and arranged on a three-dimensional landscape.

[0022] Detailed description of the invention and embodiments

[0023] The object of the invention is achieved by at least one personalized data acquisition unit, at least one personalized data processing unit, at least one personalized data presentation unit, and at least one data evaluation unit, which are interconnected by an algorithm and, optionally, a cloud-based representation of the individual's personalized health and health-related data. Based on this, they provide a 2D and / or 3D representation of the individual's physical body. This representation(s) from existing databases of the relevant systems or data sets and the algorithms or procedural instructions linking them on this basis enable the individual to explore and represent their body and mind like a landscape.The method for zooming a region of interest from a digital image generated from the health-related data is applied according to the invention in such a way that, depending on the zoom scale, the resolution of digital images is reduced or enlarged in order to personalize properties of the physical body as well as the structure of the physical body (organs; circulatory system; bone structure, etc.) and to compare, analyze, research and provide treatment recommendations with the anonymized data of other persons or users.

[0024] It should be noted that the terms used here, such as data acquisition unit, data processing unit, cluster unit or analysis unit, do not necessarily refer to physical devices, but rather refer to logical units and their function, whereby the logical units can be distributed across several physical devices.

[0025] It should also be noted that the term computer program product not only includes a computer program stored on at least one data carrier, but also includes a computer program that can be downloaded and executed individually or in a distributed manner via a network on one or more data acquisition and processing units, cluster units or analysis units.

[0026] Health-related data can be obtained from examinations of the physical body. The following lists some examples of health-related data that can be processed according to the invention. This list is exemplary and by no means exhaustive; it primarily serves to illustrate that, due to the complexity and volume of health-related data, appropriate structuring of the data is desirable in order to grant the user of the invention suitable access to their health-related data.

[0027] In some embodiments, characteristics of the physical body are obtained from blood tests that analyze various cell types and cell components to draw conclusions about the health of the organism.

[0028] Analysis of various cell types may include the analysis of red blood cells, which collects one or more of the following data: count, hemoglobin concentration, hematocrit, mean cell volume (MCV), mean hemoglobin content (MCH), and mean hemoglobin concentration (MCHC). These analyses are often performed by automated cell counting with hematology analyzers, using methods such as light scattering or impedance measurement to determine cell size and density. Such measurements can provide clues to anemia or other hematological abnormalities.

[0029] Analysis of different cell types may include the analysis of leukocytes, which collects one or more of the following data: total white blood cell count, differentiation of subtypes such as neutrophils, lymphocytes, monocytes, eosinophils, and basophils. A differential blood count may be used, for example, to assess inflammation, immune reactions, or hematological diseases such as leukemia.

[0030] Analysis of various cell types may include platelet analysis, which collects one or more of the following data: platelet count and mean platelet volume (MPV). These parameters can indicate, for example, coagulation disorders or bone marrow diseases.

[0031] Serum and plasma samples offer further opportunities for analyzing biochemical parameters from which health-related data can be derived. For example, glucose concentration can be determined using enzymatic methods such as the glucose oxidase reaction to provide insights into blood glucose levels and metabolic health. Electrolytes such as sodium, potassium, calcium, and magnesium can be quantified using ion-selective electrodes to assess fluid and electrolyte balance. Lipid profiles, including cholesterol, LDL, HDL, and triglycerides, can be analyzed using enzymatic color assays or mass spectrometry to assess cardiovascular risk.

[0032] Furthermore, the concentration of total proteins, including albumin and globulins, as well as the activity of liver enzymes such as ALT, AST, and gamma-GT can be examined. Such measurements are often performed using spectrophotometry or enzymatic activity assays to obtain information about liver and kidney function. Renal function parameters such as creatinine, urea, and uric acid can be analyzed using enzymatic methods or Jaffe-based tests to assess kidney filtration capacity.

[0033] The investigation of intracellular markers offers further options. To assess mitochondrial redox status, for example, the NAD+ / NADH- ratio can be analyzed using fluorometric or mass spectrometric methods. Analysis of mitochondrial DNA (mtDNA) offers the possibility of identifying mutations, deletions, or copy number alterations. These are typically performed using quantitative PCR or next-generation sequencing to detect mitochondrial dysfunction or age-related changes. Mitochondrial function can also be assessed by measuring ATP production. This is done, for example, using luminometric tests that detect light emission during ATP hydrolysis. In addition, the activities of enzymes of the mitochondrial respiratory chain (complex IV) can be determined using enzymatic assays or by measuring the oxygen consumption rate.Substances such as FAD, cytochrome c, and coenzyme Q10 can be analyzed using high-performance liquid chromatography (HPLC) or mass spectrometry to assess the efficiency of the electron transport chain. The study of oxidative stress in mitochondria offers the possibility of analyzing reactive oxygen species (ROS) using chemiluminescent or fluorescent probes. In addition, antioxidant status can be assessed by measuring the glutathione ratio (GSH / GSSG). Calcium concentration in mitochondria can be analyzed using fluorometric methods, using special dyes to detect dysregulations in signal transduction.

[0034] For more detailed investigations, liquid biopsies are available. These can be used to detect cell-free mitochondrial DNA (cf-mtDNA) in plasma, using digital PCR or sequencing. In addition, antioxidant enzymes such as superoxide dismutase (SOD) and catalase can be analyzed using enzymatic activity assays to assess the protective mechanisms of the cells.

[0035] Blood smears offer the opportunity to analyze the morphology of erythrocytes, leukocytes, and platelets. Abnormalities such as sickle cells or blisters can be detected using light microscopy, for example. Intracellular inclusions such as iron deposits or glycogen inclusions can be visualized using specific staining methods. Functional imaging techniques such as positron emission tomography enable the analysis of cellular metabolic activity, for example, through glucose uptake using radiotracers such as 18F-FDG. Magnetic resonance imaging offers the possibility of visualizing cell volume and tissue damage. Molecular analyses such as PCR, sequencing, or microarrays can be used to identify DNA mutations, gene expression, or epigenetic modifications such as methylation status.

[0036] Flow cytometry allows for the determination of leukocyte immunophenotypes, analysis of the CD4 / CD8 ratio, and evaluation of the cell cycle. Methods for quantifying reactive oxygen species or glutathione status can be used to assess oxidative stress. Metabolomic analyses offer options for measuring amino acids, lipids, or sugars using mass spectrometry or NMR spectroscopy.

[0037] Additionally, urine analyses can be used to measure creatinine, electrolytes, hemoglobin breakdown products, or hormones such as adrenaline and cortisol. Skin and mucous membrane swabs or biopsies can reveal cell morphology, inflammatory markers, or dysplasia.

[0038] The above analyses represent an exemplary data basis for the evaluation of cellular, mitochondrial and subcellular conditions, which can be processed as health-related data according to the invention.

[0039] The health data may, if necessary, also include other health-related data such as age, gender, height, body weight, data concerning external appearance such as skin, eye or hair colour, geographical data such as place of residence or place of work, or socio-economic information such as occupation, marital status or membership of a particular social group or class.

[0040] In some embodiments, the health-related data is processed anonymously. The anonymized data from a wide variety of sources, as well as the individual inputs of an individual, allows even personal, anonymized data released by the respective individual to be made available to the system according to the invention, in order to obtain insights and recommendations for the individual and also for the general public.

[0041] Figure 1 shows, in general terms and by way of example, the various sources that may contain data on a person or group of people. For example, the dots represent data from patient X's health records 11.

[0042] At point 12, data such as heart rate, number of steps, etc. from a fitness watch for a patient X are available. In Fig. 1, for example, point 12 illustrates the current heart rate along axis A and the current number of steps along axis B, whereby point 12 also represents further health data, such as a blood glucose value, the numerical value of which means that point 12 is not in the projection plane spanned by axes A and B, which corresponds to a specific blood glucose value from which the blood glucose value represented by point 12 deviates. The projection plane shown in Fig. 1 therefore represents only a subspace of the multidimensional taxometry, whereby the representation in the projection plane can depend on further subspaces of the multidimensional taxometry that are not contained in the projection plane.Interrelationships and relationships, or relations for short, between health data elements can be illustrated by assigning a suitable representation to each of the relations. In Fig. 1, for example, relations between some of the points are illustrated by connecting lines between the points. These relations can, for example, represent family relationships or commonalities in the health data elements, whereby the type of relationship can be expressed by a color, thickness, or line type. The representation of the relations can depend on dimensions of the multidimensional taxometry that are not contained in the projection plane.

[0043] Point 13 may contain data from the father of patient X, and point 14 may contain data from a professional group in a geographical region. So-called normative data for individuals (according to gender, height, weight, place of residence, behavior patterns, etc.) can be compared with the personalized, anonymized data of the individual patient X, but also with the anonymized data of a comparable group, in order to create analyses and behavioral recommendations for the individual and / or their doctor, fitness trainer, etc. Points 1 to 1+n (n as a natural number) represent the multitude of data sources from which information can be obtained for patient X. The 3D point representation can be designed using line thicknesses and colors so that identical or similar features are also easily recognizable visually.This makes it easier to use a zoom function in such a way that similar data from different levels can be overlaid in order to create cluster analyses.

[0044] If the multidimensional taxometry includes time in one dimension, the health data and the associated clusters can be related simultaneously in a temporal representation, for example, in a three-dimensional representation including the projection of the time axis. The health data elements represented by points in the multidimensional taxometry do not necessarily have to be assigned to a patient, but can also, for example, be assigned to a group of patients or, conversely, to individual body components of a single patient. The various health data elements can be hierarchically related, with one or more properties of a health data element being represented by hierarchically subordinate health data elements.For example, a patient's general health status can be represented by health data elements associated with the patient's individual organs. In this case, the multidimensional taxometry can be hierarchically structured to classify the plurality of health data elements according to their hierarchical structure. Classifying the plurality of health data elements can also involve selecting health data elements according to one or more selected hierarchy levels in the taxonomy, so that health data elements outside the one or more selected hierarchy levels of the representation are ignored.

[0045] The multidimensional taxometry can also include the spatial coordinate axes of a two- or three-dimensional representation of a schematic or current body of patient X and / or time as another coordinate axis of the multidimensional taxometry. This allows the body of patient X and, if applicable, its development or components of the body of patient X and their development over time to be represented.

[0046] The ability to "dive beneath the skin," for example, and to "travel" between passages such as the intestine, an artery, or vein, and into the organs (stomach, heart, etc.), also makes it possible to share certain parts of the database(s), for example, with physicians of different specialties, or to add new data to the database over time by incorporating data from PET / CT scans or colonoscopies, for example, and to create a medical history and treatment recommendations as well as personalized behavioral patterns. It allows the individual (or another member of an algorithm) to explore changes over time (due to aging, etc.), predict likely deteriorations, and identify the need for preventative countermeasures (e.g., a training program to strengthen specific muscles or structures).

[0047] Algorithms such as a "zoom" function can enable the visual representation of increasing detail at a specific location in the body. For example, the user first sees their heart, then details of the cardiac muscle, then cells and the intracellular matrix, and finally the internal compartments of a single cell, such as mitochondria.

[0048] A CT and / or MRI scan can serve as the basis for the overall data set and representation, with information such as video images from a colonoscopy being integrated as appropriate. Ultrasound images can also be integrated, and commonly available scans and models of organs can be used to fill in the gaps in the individual data set, thus archiving a representation that comes as close as possible to the individual's probable drive.

[0049] Algorithms such as "follow the path" instructions can be used to "travel" along tubular structures such as a vein or intestine, observing the lining from within or detecting abnormalities along their length, such as an aneurysm or diverticulitis. A PC mouse, for example, can be used as a control instrument.When the peculiarity of the organ in question, such as a high content of fat cells in the liver, is detected by an algorithm ("compare the standard" algorithm) by comparing the individual's presentation with a healthy and average one, an additional algorithm ("suggest remedy") shows options such as lifestyle changes ("consume fewer carbohydrates, reduce alcohol consumption") and possible medications, and adds options for further analyses to observe changes (for example, pulsed ultrasound to monitor and measure fibrosis of life).

[0050] Treatment centers and specialists can also be suggested and identified. The estimated costs and timeframe of treatments can be provided. An additional algorithm ("Overall Benefits / Risks") highlights the locations and shows the possible and likely effects of such an intervention or lifestyle change on the individual's body and mind. Based on the person's medication and dietary intake, the likely short- and long-term effects on body structure, organ effects, and the likely effects on the person's mind and mood are displayed, and risks or side effects can be highlighted.

[0051] Fig. 2 shows exemplary health data elements in the form of point clouds, which can result, for example, from the health data of patients with cardiac arrhythmias. Cardiac arrhythmias can be caused by recreational drugs such as alcohol, nicotine, or coffee. Disturbances in mineral balance can also be the cause of cardiac arrhythmias. A mineral deficiency, particularly potassium and magnesium, can, for example, trigger or exacerbate cardiac arrhythmias. Another possible cause of cardiac arrhythmias is an overactive thyroid. Finally, high blood pressure, obesity, and congenital or acquired heart defects can also lead to cardiac arrhythmias. Individuals' health data elements can be classified according to these causes, and the multidimensional taxometry can be entered.In Fig. 2, two of these causes are plotted along axes A and B in the two-dimensional representation, which represents a subspace of the multidimensional taxometry. A further subspace of the multidimensional taxometry, not spanned by the two axes A and B, is represented in Fig. 2 by coloring the health data elements, for example, by encoding the value of a dimension not shown by a color value of the points in the projection plane.

[0052] Color, location, and size, as well as overlaps of freely selectable clusters, allow conclusions to be drawn about possible treatment outcomes. On this basis, individual, personalized comparative studies can be created for patients. For example, a point cloud 23 can depict all male patients with cardiac arrhythmias. Point cloud 21, for example, can contain patients with a specific body mass index (BMI). Comparisons of treatment methods can be derived from intersections 22 of the overlapping point clouds 21 and 23. For this purpose, an area 22 can be zoomed in to inspect further relevant comparative data, which may arise, for example, from the patient's medical history or geographical features.

[0053] As a result of zooming into area 22, a data group 34 can result, as shown in Fig. 3, which in turn is formed as a result of the overlaps of data groups 31, 32, and 33. Data group 31 could include patients experiencing cardiac arrhythmias in the atrium. Furthermore, a distinction could be made between cardiac arrhythmias with a heart rate that is too slow (bradycardia, below 60 beats per minute) and those with a heart rate that is too fast (tachycardia, above 100 beats per minute), which would be captured, for example, by data group 32.

[0054] The underlying cause of cardiac arrhythmias can often be a lack of energy in the heart muscle cells. This disrupts the function of the cell's powerhouses, the mitochondria. This dysfunction can be caused, among other things, by reduced blood flow to the heart muscle and a lack of oxygen, but many prescribed heart medications (e.g., beta-blockers, cholesterol-lowering drugs, or antibiotics) can also damage the mitochondria.

[0055] As an example of damage caused by antibiotics, data group 33 is shown in Fig. 3. Data group 34, resulting from the overlap of data groups 31, 32, and 33, can be re-evaluated on this basis with positive medical treatments and / or personalized behaviors. For example, treatment with a so-called neurostimulator could have been administered over defined zones in a patient's ear. The treatment success over a period t with this auricular neurostimulator thus provides conclusions about the effects (duration; intensity of stimulation) of the ECG values ​​available during this period t and / or the personalized data from documented records of an individual patient. This provides a reflection of lifestyle (eating habits, alcohol, occupational activities (stress symptoms), pedometer, sleep behavior, etc.).

[0056] Data groups 35 and 36 are examples of so-called landscape formation. In this case, the health data elements are represented by corresponding heights according to their frequency of occurrence in different patients. If the heights in the representation represent the frequency of occurrence of health data elements not for the entire group of patients, but for a single patient X, hills can grow during landscape formation, which can be constructed from the temporally added number of identical health data elements for organs, blood counts, ECG data, etc. Through the growth and formation of so-called mountains and valleys, a personalized health atlas can be created over time, which can be compared with other health atlases in terms of contour, shape, height, peaks, etc. Color contours of the time periods facilitate selection in cluster analysis.

[0057] The extraction of individual personal data can also be carried out using terms and / or image captures from the full texts of treatment and / or medical examination protocols, in order to then be standardized on the basis of syntactic and semantic rules (natural language processing).

[0058] Each personalized concept is weighted based on the organ or body part in question (i.e., real-time state, general health history, personalized history; etc.) in which it was identified and how often it occurred: the extracted concepts reflect the patient's semantic content.

[0059] The visual representation through a map is created with relevant concepts and can be built up in the following steps.

[0060] First, a vector model of multidimensional taxometry is created, which primarily involves the extraction and weighting of concepts. This is followed by clustering, a cluster analysis in which a procedure is carried out to discover similarity structures in data sets. The groups of "similar" objects thus found are referred to as clusters, and the group assignment is referred to as clustering. By reducing the dimensions of multidimensional taxometry, personalized, anonymous data are compared and presented, and the associated personalized, anonymous clinical pictures and / or behavioral patterns and / or fitness data, etc., are determined.

[0061] By projecting the generated health data elements using a mapping algorithm, a graphical representation of the health data elements and their relationships is generated. This is followed by visual analysis and presentation. An example of this is maps that depict a specific region of the body, for example, and to which the actual appearances and data can be adapted through appropriate scaling. This helps in representing the health data of multiple patients in a 2D space (map) or a 3D representation, so that the distances can reflect the similarity distances of individual personalized health data as accurately as possible.

[0062] Each display point or a display point cloud (data set) can correspond to a person with his or her patient data or to a group of people with identical or very similar characteristics, so that the position of the points depends on the similarity of the patients.

[0063] The more health data elements and the more similarities between the health data elements they have in common, the closer they are to each other, providing clues to diseases and possible treatments with probable success rates, recommendations, etc. The entire system can be set in relation to a healthy body as a normative size. The distance of a point representing a person from the normative size can be assumed to be a measure of the health status of patient X. The distances between the points assigned to individual patients are calculated and presented pictorially or conceptually in a multidimensional representation (N patients à N1 dimensions).

[0064] A single colored cluster contains those patients (represented by dots) who share concepts / characteristics and are connected by spatial proximity and similarities. At the same time, the distance to the standard size of a healthy body or a point cloud of the healthy body is shown as a reference point.

[0065] During projection, the system calculates the position of each cluster or point cloud.

[0066] A three-dimensional representation can also be used to show a connection with a subspace that is not shown.

[0067] If patients of the same predefined cluster or a point cloud with similar representation (size, shape, color, etc.) are located on several so-called peaks (3D representation of height differences of the patient data or clouds) or are located on a peak, this means that the system has determined that some patients also have connections to another taxometric dimension. The representation of a connection with a taxometric dimension can thus be shown not only by the color of the health data elements, as in Fig. 2, but also by height modeling. The coloring can also interact with the height modeling. For example, a green color in a peak area can illustrate a particularly strong connection with another taxometric dimension that is otherwise not shown.One of the main advantages is that all information can be presented in a format that is easy for every user to understand, allowing each user to learn about and observe their body, organs, and their functions. Through anonymized presentation and a so-called sharing function, they can enter their personalized data into the overall system and make it available anonymously.

[0068] By using the system according to the invention, each user benefits from the personalized, anonymous data of every other user to gain insights into their own person.

[0069] Technical or medical terms used in the description of a medical report, which are usually written in Latin or medical abbreviations, can be translated into everyday language, referred to here as conceptual language, and synchronized. Using a semantic database, this conceptual language can also be translated into medical jargon or into specialized formulations. A "great saphenous vein" would suddenly become a concrete visual impression rather than an incomprehensible Latin term.

[0070] For this purpose, artificial intelligence (AI) algorithms are used to connect the different user groups and user levels. This allows specific elements or body parts and / or organs or their functions to be included in the representation of the human body and are understandable to the average user. This, in turn, helps the user, for example, to enter personal data into the system as real-time data (e.g., body temperature; headache; sweating; date / time, etc.). The visualization of a procedure, such as the closure of the great saphenous vein in a specific person and the effects this has on the person's blood flow distribution, can be observed, and this would eliminate considerable explanation time before a treatment.A potential patient could finally see what effect such a procedure would have on blood circulation and how this could improve the healing of a chronic wound on his leg.

[0071] However, even if the implementation of the method according to the invention and the associated system is a medium- to long-term project and goal, short-term improvements over the state of the art, in which medical information is scattered in different locations in different doctor's offices, hospitals and in the patient's own file system, are easily achievable in both analog and generally incompatible digital formats.

[0072] Current attempts, such as the attempt to create an electronic patient record system (current attempt to improve the state of the art), are taken into account by the invention, as it provides a comprehensive framework and data structure that can be understood by the patient, as well as creating motivation and the opportunity, as well as helping to inform and educate the patient about the effects of various measures on his body and on his health.

[0073] The ability to single out specific subsystems, such as the circulatory system or a specific neural feedback loop, could also be integrated. Making certain layers and systems transparent while highlighting others is therefore another feature of representation (or a structured database).

[0074] Basic data such as age, gender, weight, blood pressure and simple visualization ("body maps" with ZOOM), with continuous addition of existing and new data (ECG, data that can be automatically imported from Withings and smartwatches, etc.) can also be processed

[0075] In a specific embodiment of the invention, the method and system according to the invention can be associated with a method and system that links newly acquired data and data analyses in the form of an adapted geographical visualization on a graphical user interface (GUI), for example on a remote client computing device, in order to obtain new data models therefrom.

[0076] Databases with patient-specific data can be used for reference and comparison.

[0077] The solution to the problem may also include using an app to make it easier for the user to visualize their health status and enable them to compare their own health data with reference values.

[0078] If a person continuously records their data, such as blood pressure and ECG (using a health app), simple data analyses can provide faster and more accurate diagnoses (e.g., separating generalized high blood pressure from systolic hypertension). Smartwatches or similar fitness watches can provide diagnoses such as "sinus rhythm" or "inconclusive" very quickly (e.g., within 30 seconds) via an app on these devices.

[0079] The components of a comprehensive system could be: Input (manual and automatic):

[0080] • Blood tests, MRI, PET, Ultrasound, DNA, ECG

[0081] • Data such as the values ​​provided by the Withings scale on weight, muscle and fat percentage, etc.

[0082] • Blood pressure measurements

[0083] • Age, gender

[0084] • Physical activities

[0085] • Eating habits

[0086] • Other (smoking, alcohol, sugar)

[0087] • Medications with details)

[0088] • Inputs on mental fitness

[0089] Visualization through status and comparisons with best comparison data / standard sizes are possible.

[0090] Projections based on this could be personalized and defined as recommendations:

[0091] • Sports, fitness, exercise

[0092] • Nutrition

[0093] • Medications

[0094] • Operations

[0095] • Reactions to treatments (positive and negative)

[0096] • etc.

[0097] A communication interface with doctors, hospitals, medical facilities and devices, and / or clouds, etc., can be integrated. Based on this, the patient—or user—has a personalized app for managing their personal health data, including medical and fitness data. The app serves as a replacement for unreliable paper records or various electronic systems that contain only fragments of their medical history. With this app, all important information is stored in one place.

[0098] The invention also encompasses a system and method for acquiring, displaying, evaluating, and providing personalized instructions from data sets that may result from cluster analysis(s) of sleep data, sleepwalking, personalized dream experiences and their personal interpretations, from personal descriptions of heat sensations, food intolerances, headache symptoms, body aches, restlessness, mood states, etc., which may result in conjunction with real-time displays from existing data from smartwatches or fitness watches.

[0099] This system and procedure can also be applied to mental illnesses, including their cause research and treatment recommendations. This would allow for more targeted research and treatment of so-called "burnout disorders." The consequences and conclusions of remote work could also be better researched.

[0100] The invention is also particularly applicable to further research into autoimmune diseases, nervous system disorders such as multiple sclerosis, and rare diseases for which the necessary research effort is not always sufficient due to a small number of patients (e.g., Asperger's syndrome, Barrett's esophagus, vascular dementia, osteochondritis, etc.). In particular, the zoom function of the data analysis related to genetic influences, in conjunction with geographical data analyses, with lifestyles of the individual groups, the type and extent of drug treatments, and the resulting intersections for personalized treatment and personalized behaviors, are of immense importance here.

[0101] The early detection of possible diseases, changes in a user's body, risks of spread of infectious diseases, and predictive models for the emergence and spread of impending pandemics can also be researched and their emergence and spread can be better combated in real time.

[0102] A prerequisite for this implementation is the anonymization of the collected user data. The user can release the personalized data by granting it to specific user groups, such as their family doctor.

[0103] Further advantageous embodiments of the invention are listed in the following numbered embodiments (NAB):

[0104] NAB 1. System for the recording, presentation and evaluation of personalized health data comprising: one or more servers which are set up to receive, link and evaluate health data of a person A upon a request for data analysis by at least one personalized data recording unit, at least one personalized data processing unit, at least one personalized data presentation unit and at least one data evaluation unit, a personalized input unit which is able to link newly recorded health data for the person A with the already existing personalized health data of the person A, at least one data unit with which personalized health data of any number of users B can be processed by cluster analyses, at least one data unit which is set upto link, analyze, display, translate, and standardize the health data of person A and the health data of user B.

[0105] NAB 2: System for the recording, presentation and evaluation of personalized

[0106] Health data according to NAB 1, whereby the personalized data set of at least one person A is compared with persons B from the data sets (n+1) by means of a cluster analysis, which is carried out based on 2D and / or 3D representation of the physical body of the individual in the data set (n+1).

[0107] NAB 3. System for the collection, presentation and evaluation of personalized health data according to NAB 1 or NAB 2, whereby a geographically based cluster is compared with the health data, evaluated and recommendations for action are generated.

[0108] NAB 4. Method for operating a system for recording, displaying and evaluating personalized health data, in which the health data of a person A are received, processed, linked and / or evaluated for data analysis with the aid of at least one personalized data recording unit, at least one personalized data processing unit, at least one personalized data display unit, and at least one data evaluation unit, and in which

[0109] With the help of at least one personalized input unit, newly recorded health data of person A is linked in real time with the already existing personalized health data of person A, at least one data unit processes the personalized data of any number of users B through cluster analyses, and with the help of at least one data unit, the health data of person A and the health data of user B are linked, analyzed, displayed, translated and / or standardized, in particular for use in the context of medical treatment and / or for creating a personalized fitness program.

[0110] NAB 5. Method for operating a system for recording, displaying and evaluating personalized health data according to NAB 4, wherein at least one data analysis related to the physical body is carried out by zooming in on an area of ​​interest from a digital image in order to compare the genetic and / or disease-causing influences and / or the physical and / or psychological characteristics of a body with geographical and / or environmentally specific and / or climatic data analyses and / or with lifestyles of groups of people and / or the clinical picture of a person or group of people, in order to determine the type and extent of medicinal treatment and to create a personalized treatment and / or behavioral recommendation for person A from the resulting data intersections.

[0111] NAB 6. Method for operating a system for the acquisition, presentation, and evaluation of personalized health data according to NAB 4 or NAB 5, whereby individual personal data of person A is obtained via terms and / or images from the full texts of treatment and / or medical examination protocols in order to standardize these on the basis of syntactic and semantic rules. NAB 7. Method in which an auricular neurostimulator is used for treatment over defined zones in the ear of a patient, whereby the treatment is carried out over a period of time t with this auricular neurostimulator and conclusions are drawn regarding the effects (duration; intensity of the stimulation) on ECG values ​​available during this period t and / or the personalized data from documented patient records.

[0112] NAB 8. Computer program with program code for carrying out the method according to claims 4 to 7, when the computer program is executed on a computer unit.

[0113] NAB 9. A computer-assisted method for navigating through a plurality of health data elements, the method comprising:

[0114] Selecting a multidimensional taxonomy for health data elements;

[0115] Classifying the plurality of health data elements such that each health data element is associated with a point in a space spanned by at least a first portion of the dimensions of the multidimensional taxonomy; for each health data element, assigning a first representation to the point in space associated with the respective health data element, wherein the associated first representation depends on one or more properties of the health data element that are classified by at least a second portion of the dimensions of the multidimensional taxonomy;for each relation between related health data elements, assigning a second representation to the respective relation, wherein the assigned second representation depends on one or more properties of the respective related health data elements that are classified by at least a third part of the dimensions of the multidimensional taxonomy, wherein the first, second and third parts of the dimensions may overlap;

[0116] Causing an output of a portion of a projection of the space corresponding to an output of the first and second representations of an identified subset of health data elements and their relations;

[0117] Providing a user interface for changing the snippet and for causing an output of the changed snippet.

[0118] NAB 10. The method of NAB 9, wherein at least a portion of the plurality of health data elements is in a hierarchical relationship, wherein one or more properties of a health data element are represented by hierarchically subordinate health data elements, and wherein the multidimensional taxonomy is hierarchically structured to classify the plurality of health data elements according to their hierarchical structure.

[0119] NAB 11. The method of NAB 10, wherein classifying the plurality of health data elements comprises selecting health data elements according to one or more selected

[0120] hierarchy levels in the taxonomy, so that health data items outside the one or more selected hierarchy levels are ignored in the output.

[0121] NAB 12. The method of NAB 10 or NAB 11, wherein providing the user interface comprises providing a zoom function by means of which the one or more hierarchy levels are selected and the output of the first and second representations is adjusted accordingly.

[0122] NAB 13. Method according to one of NAB 9 to NAB 12, wherein the first part of the dimensions comprises three dimensions which are output in a two-dimensional projection of the space.

[0123] NAB 14. Method according to NAB 13, wherein the first representation of up to three properties of

[0124] health data elements that are output as a color value, allowing the user to identify commonalities or similarities based on clusters of similar colors.

Claims

Patent claims 1. A method for operating a system for recording, displaying and evaluating personalized health data, in which, for a data analysis of the health data of a person A, the health data of the person A are recorded with the aid of at least one personalized data recording unit, the health data of the person A are processed, linked and / or evaluated by at least one personalized data processing unit, the personalized data of a large number of users B are processed by means of cluster analyses by at least one cluster unit, the health data of the person A and the health data of the users B are linked, analyzed, translated and / or standardized by at least one analysis unit, and the results of the analysis unit, in particular for use in the context of medical treatment and / or for the creation of a personalized fitness program.

2. The method according to claim 1, wherein the personalized data processing unit, the analysis unit and / or the cluster unit performs / performs the following: Selecting a multidimensional taxonomy for the health data; Classifying a plurality of health data elements such that each health data element is associated with a point in a space spanned by at least a first portion of the dimensions of the multidimensional taxonomy; for each health data element, assigning a first representation to that point in space associated with the respective health data element, wherein the associated first representation depends on one or more properties of the health data element that are classified by at least a second part of the dimensions of the multidimensional taxonomy.

3. The method of claim 2, wherein the data representation unit causes the output of a portion of a projection of the space that corresponds to an output of the first representations of an identified subset of health data elements.

4. The method according to claim 2, wherein the personalized data processing unit, the analysis unit and / or the clustering unit further performs the following: for each relation between related health data elements, assigning a second representation to the respective relation, wherein the assigned second representation depends on one or more properties of the respective related health data elements that are classified by at least a third part of the dimensions of the multidimensional taxonomy, wherein the first, second and third parts of the dimensions may overlap, and wherein the data representation unit performs the following: Causing an output of a portion of a projection of the space corresponding to an output of the first and second representations of an identified subset of health data elements and their relations; 5. The method according to any one of claims 2 to 4, wherein at least a portion of the plurality of health data elements is in a hierarchical relationship, wherein one or more properties of a health data element is represented by hierarchically subordinate health data elements, and in which the multidimensional taxonomy is hierarchically structured in order to classify the plurality of health data elements according to their hierarchical structure.

6. The method of claim 5, wherein classifying the plurality of health data items comprises selecting health data items according to one or more selected hierarchy levels in the taxonomy such that health data items outside the one or more selected hierarchy levels are ignored in the output.

7. The method according to any one of claims 3 to 6, wherein the data representation unit provides a user interface for changing the section and for causing an output of the changed section.

8. The method of claim 7, wherein providing the user interface comprises providing a zoom function by means of which the one or more hierarchy levels are selected and the output of the first and optionally second representations is adjusted accordingly.

9. Method according to claim 8, wherein by zooming an area of interest from a digital image at least one data analysis is carried out relating to the physical body of person A, wherein the genetic and / or disease-causing influences and / or the physical and / or psychological characteristics of the body of person A are linked to geographical and / or environmental and / or climatic data analyses and / or to lifestyles of groups of users B and / or the The clinical picture of person A or user B is related in order to determine the type and extent of medicinal treatment and, from the resulting data intersections, to create a personalized treatment and / or behavioral recommendation for person A.

10. The method according to any one of claims 2 to 9, wherein the first part of the dimensions comprises three dimensions output in a two-dimensional projection of the space.

11. The method of claim 10, wherein the first representation depends on up to three properties of health data elements that are output as a color value, so that the user can determine commonalities and / or similarities based on clusters of similar coloring.

12. Method according to one of claims 1 to 9, in which the personalized health data of person A are related to the health data of user B by means of a cluster analysis and in which the health data of person A related to the health data of user B are displayed in a 2D and / or 3D representation of the physical body of person A.

13. Method according to one of claims 1 to 12, in which a geographically based cluster of users B is generated by the cluster unit and the analysis unit relates the geographically based cluster to the health data of person A, evaluates the relationship and generates recommendations for action for person A based on the evaluation of the relationship.

14. Method according to one of claims 1 to 13, in which, in order to provide individual health data of person A, terms and / or image recordings from the full texts of treatment and / or medical examination protocols are evaluated by the data recording unit and are standardized on the basis of syntactic and semantic rules.

15. Method according to one of claims 1 to 15, wherein newly recorded health data of person A are linked in real time with the already existing personalized health data of person A with the aid of at least one personalized input unit.

16. System for recording, displaying and evaluating personalized health data according to a method according to one of claims 1 to 14, comprising: one or more servers which are set up to receive, link together and / or evaluate health data of a person A in response to a request for data analysis by at least one personalized data recording unit, at least one personalized data processing unit, at least one personalized data display unit and at least one data evaluation unit, at least one cluster unit with which personalized health data of any number of users B can be processed by cluster analyses, at least one analysis unit which is set up to standardize, link together and / or display the health data of person A and the health data of users B.

17. System according to claim 15 for carrying out a method according to claim 16, which is provided with a personalized input unit which is capable of To link newly recorded health data of Person A with the existing personalized health data of Person A.

18. Computer program product with program code for carrying out the method according to one of claims 1 to 15, when the program code is executed on a computer unit.

Images