WO2010057891A1 - Procédé mis en oeuvre par ordinateur pour l'aide au diagnostic médical - Google Patents

Procédé mis en oeuvre par ordinateur pour l'aide au diagnostic médical Download PDF

Info

Publication number
WO2010057891A1
WO2010057891A1 PCT/EP2009/065333 EP2009065333W WO2010057891A1 WO 2010057891 A1 WO2010057891 A1 WO 2010057891A1 EP 2009065333 W EP2009065333 W EP 2009065333W WO 2010057891 A1 WO2010057891 A1 WO 2010057891A1
Authority
WO
WIPO (PCT)
Prior art keywords
diagnosis
diagnostic
medical
risk
patient
Prior art date
Application number
PCT/EP2009/065333
Other languages
German (de)
English (en)
Inventor
Frank Gotthardt
Dierk Heimann
Original Assignee
Compugroup Holding Ag
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 Compugroup Holding Ag filed Critical Compugroup Holding Ag
Priority to EP09752195A priority Critical patent/EP2347358A1/fr
Priority to US13/128,185 priority patent/US8548827B2/en
Publication of WO2010057891A1 publication Critical patent/WO2010057891A1/fr

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
    • 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
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • 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
    • G16H40/60—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 operation of medical equipment or devices
    • G16H40/63—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 operation of medical equipment or devices for local operation

Definitions

  • the invention relates to a computer-implemented method for medical diagnosis support, a data processing system and a computer program product.
  • Medical information systems document, among other things, diverse, patient-related, administrative and medical data. Although the use of medical information systems makes it possible to document patient data essentially comprehensively and to store the patient data, the problem that a treating physician often has is that time problems in medical practices and hospitals arise It is rarely possible for a patient to have a complete overview of the patient's course of treatment by reviewing a patient's medical record before the start of a treatment appointment. In this case, a treating physician often only has time to spend a lot of time on the health problems and diagnoses of the patient, which have occurred in the recent past.
  • a symptom associated with a disease may vary from patient to patient.
  • a symptom may be indicative of a variety of different diseases, and each disease may be caused by a multitude of not always clear, symptoms are characterized.
  • the available medical expertise for the various diseases is very unevenly distributed.
  • the causes and symptoms of some diseases are well known and well described, while the causes of other diseases are still completely obscure.
  • at least correlation studies are available that show a statistical correlation between certain environmental factors, eating habits, physical activity, a particular genotype or the presence of other diseases (comorbidities).
  • Some diseases can be clearly attributed to one or a few causes, such as monogenic heritable diseases a gene defect.
  • Articular arthrosis can be caused, for example, by age and wear-related joint wear.
  • joint arthrosis can also be the result of a corresponding genetic predisposition that comes into play after a certain age.
  • diagnosis is complicated by the fact that various diagnostic methods for detecting a disease are possible.
  • diagnostic procedures and query standards recommended by the health insurance funds in accordance with a guideline diagnostics specific to the respective disease.
  • diagnosis in the following refers to a finding on a physiological condition or a disease of a patient
  • diagnosis was classically made on the basis of externally recognizable features (symptoms), laboratory values or various diagnostic procedures by a doctor, who provided these data before
  • a significant advantage of the present method according to the invention is that these evaluation steps can be automated and take into account more information than the physician may expect in the short term thus improving the quality of the diagnoses and accelerating the diagnosis.
  • the invention has for its object to enable a user of a medical information system to perform an analysis of patient data regarding the presence of diseases in a reliable, efficient and faster manner.
  • a technical challenge in particular is the high complexity and heterogeneity of the factors that must be taken into account for the risk calculation, as well as the compelling need to be able to quickly execute a large number of complex queries on a large data set of electronic patient records Clinic). While nothing is known about a disease other than a simple correlation and the risk calculation procedure can be correspondingly simple, other highly complex risk calculation methods can exist for other diseases because it is well studied and many studies are available.
  • a practicable diagnostic system must be able to allow for this heterogeneity of risk calculation procedures as well as frequent changes in the calculation methodology. The system must also be able to address the practical problems of diagnosis by the clinician (limited time available, unclear symptoms).
  • the invention relates to a computer-implemented method for medical diagnostic support for patient data of a patient by a data processing system.
  • the data processing system has a graphical user interface.
  • the procedure begins by accessing rules for calculating diagnostic risks of medical diagnoses.
  • the rules and the data objects representing the diagnoses are stored in a database in a manner that accommodates the heterogeneity of knowledge about various diseases and associated symptoms described above.
  • Every diagnosis in this database is stored with a primary medical risk.
  • the primary medical risk for a diagnosis indicates the probability of the presence of this diagnosis in a patient may be accepted if the only information used for this assumption is the general statistical distribution of a disease in the total population.
  • the primary risk of having a disease affecting 10 000 people in a population of 1 million is 0.01 (1%).
  • Age, gender or pre-existing conditions are not taken into account when calculating the primary risk. Rather, the primary risk is used according to the currently available medical knowledge (number of diseases per total population or, if unknown, number of patients within a studied patient group of a medical study).
  • a reference to the literature source from which the value for the primary risk was taken is also stored in the database.
  • a prediction system is not only able to assign a primary risk to every diagnosis.
  • medical diagnosis risks are calculated individually for the patient depending on personal risk factors for a multiplicity of possible diagnoses.
  • rules are applied to the data of the patient.
  • Each rule contains one or more query conditions (age, sex, medical history, etc.).
  • Applying a rule to the data of an electronic health record means checking that all query conditions of a rule for that record are met.
  • the rules are stored in a database in such a way that a large number of possible query conditions in different combinations can be flexibly taken into account.
  • the database scheme used also makes it possible by recording appropriate updates for the medical diagnostic objects and risk calculation methods, so that the inventive method can be easily adapted to the current and ever-changing state of medical knowledge.
  • Applying the rules to the patient data results in the calculation of at least a first medical diagnostic risk for a first medical diagnosis when at least one of the rules is applicable to the patient data.
  • the database contains three rules for calculating a risk for a particular disease K, all three of which contain as a condition a patient age of at least 30 years, then in this example none of the rules is applicable to a 25 year old patient.
  • the output of the first calculated diagnosis risk for the first medical diagnosis is performed together with the first medical diagnosis on the graphical user interface and the output of a user query as to whether an interactive symptom diagnosis and / or a guideline diagnostic should be performed for the first medical diagnosis.
  • Medical guidelines are systematically developed diagnostic and symptom assessment procedures to support physician decision-making. Both symptom diagnostics and guideline diagnostics serve, on the one hand, to specify the first diagnosis risk calculated by application of the rule by interactively specifying further characteristics of the patient. On the other hand, you suggest to the doctor symptoms or guideline criteria for the selection that are stored associated with the first diagnosis. These guideline criteria and symptoms, in turn, may correlate with other diagnoses that are also suggested to the physician for selection.
  • the issue of a symptom user query is made, by which the physician can determine which of the medical symptoms associated with the first medical diagnosis will be used for further analysis of the patient data and have an impact on the previously determined diagnostic risk.
  • the first diagnostic risk calculated in the previous step is modified and specified.
  • the physician will select some or all of the suggested symptoms. Any selection or deselection of a symptom may increase or decrease the first diagnosis risk.
  • the symptom user query can be used by the physician to specify the first diagnostic result based on the application of rules.
  • the second diagnostic risk determined in the symptom diagnosis thus uses the first diagnosis risk as the initial value in order to specify this, depending on the presence or absence of further symptoms. Finally, in a follow-up the step of issuing the second, more precise, diagnostic risk along with the second diagnostic on the graphical user interface.
  • a guideline diagnostic user query is output. If the guideline diagnostics is performed immediately after the first diagnostic risk has been calculated, the first diagnostic risk is the starting point for further clarification of the diagnostic risk. If the guideline diagnostic is performed after the symptom diagnosis, the second diagnosis risk determined in the symptom diagnosis sets the initial value for further clarification of the diagnosis Diagnosis risk.
  • the diagnostic risk calculated in the course of the guideline diagnostics is referred to as the third diagnostic risk, irrespective of the actual sequence of the diagnostic steps. Similarly, the diagnostic risk determined in symptom diagnostics is referred to as a second diagnostic risk.
  • the symptom diagnosis is therefore not a prerequisite for conducting the guideline diagnostics. Rather, both diagnostic methods can be based on each other or individually directly after the calculation of the first diagnosis risk.
  • Symptoms to be used by the physician according to a guideline diagnostic for assessing the existence of a specific diagnosis are referred to below as guideline criteria.
  • the guideline criteria are linked to the diagnostic objects stored in a first database. Conducting a guideline diagnostic diagnostic means that the user is presented with the guideline criteria associated with that diagnosis for selection.
  • the guideline criteria may also include laboratory values of the patient, such as blood glucose, serum creatinine, blood pressure, or similar data.
  • the user usually the physician, selects from the presented set of guideline criteria some that he considers relevant and that should be used to further clarify the previously determined diagnostic risk. By selecting or deselecting the presented guideline criteria by the user, the diagnostic risk calculated in the previous step is further modified and specified.
  • the physician will select some or all of the proposed guideline criteria.
  • the selection or deselection of individual guideline criteria has one Modification of the initial risk value results, as a result of which a third diagnostic risk is returned and displayed.
  • the third diagnostic risk is further refined through the use of disease-specific guideline routines.
  • Guideline routines are diagnostics-specific calculation routines that ultimately result in a modification of the second diagnostic risk value.
  • the guideline routines may, for example, give greater weight to the existence of individual guideline criteria, perform complex Boolean operations (eg, AND, OR, NOR) or arithmetic functions on the selected guideline criteria, and apply the modified diagnostic risk calculated therefrom.
  • the guideline routines on the guideline criteria for calculating the diagnostic risk are heuristics that are based on a combination of several individual factors.
  • the MDRD formula commonly used to diagnose renal dysfunction takes into account not only serum creatinine levels (laboratory findings), but also the patient's age, skin color and gender. So factors that are known from various studies that you have an influence on the presence of kidney dysfunction, or at least correlate with it.
  • ICD codes International Statistical Classification of Diseases and Related Health Problems
  • LEZ performance numbers of the previous illnesses and diagnoses
  • ICD codes represent diagnoses that have already been made in the patient's past according to the patient record. Since the occurrence of some diseases in the past is positively correlated with an increased risk for the occurrence of other diseases, a consideration of this factor in the rules can be helpful in the risk calculation. Also, LEZ codes can aid in the calculation of the diagnostic risk, although they are not always indicative of a particular pre-existing condition.
  • the third diagnostic risk uses the second diagnostic risk as an initial value to specify it, depending on the presence or absence of the guideline criteria assigned to the diagnosis, and on the outcome of the guideline routines. Finally, in a subsequent step, the output of the thus calculated third diagnosis risk is performed together with the third medical diagnosis on the graphical user interface.
  • the physician can confirm the diagnosis, which is consequently stored in the electronic patient record of the patient.
  • the calculation of one or more first diagnostic risks by application of the rules is immediately initiated according to a preferred embodiment of the invention when the doctor or a receptionist opens the electronic patient record.
  • the display of the calculated diagnostic risks can be made later, e.g. only when opening a prescription form by the doctor.
  • This embodiment is particularly advantageous since in everyday practice the electronic patient record is typically first opened by a medical assistant, e.g. Enter laboratory values or administrative data associated with the doctor's visit.
  • the fact that the risk calculation is initiated by opening the electronic file the results are already available to the doctor, which brings a further time savings.
  • the doctor can immediately switch to symptom diagnostics or guideline diagnostics.
  • the diagnoses obtained by applying the rules and further patient-related data are displayed in a pop-up window.
  • a threshold value for the calculated diagnosis risk it is possible to achieve that only really relevant information is displayed.
  • the present invention offers the possibility of a suspected diagnosis check. This feature includes allowing the physician to directly enter a diagnosis into the system as a suspected diagnosis.
  • the suspected diagnosis check differs from the procedure explained above in that rules that are applied to the patient data do not propose the first diagnoses but the physician does so.
  • the doctor selects a suspected diagnosis from a list of possible diagnoses in the first database. In the next step, he can specify his suspicion through the application of symptom diagnostics and / or guideline diagnostics and, if necessary, reject the suspected diagnosis or transfer it to the patient record as secure.
  • patient data is understood to be any type of information which has been recorded with regard to a patient.
  • this also includes electronic image data as well as medical measurement data of any kind.
  • Structured patient data refers to patient data provided according to a previously defined standard or classification. This includes, but is not limited to, the use of ICD codes, Central Pharmacy (PZN) and LEZs in accordance with the Uniform Medical Fee (EBM) Assessment Standard, as well as specific contents of Medical Device Ordinance (KV) forms such as transfers, admissions, disability certificates or similar.
  • the method according to the invention has the advantage that a treating physician is enabled to consider various medical diagnoses in their entirety in a coherent manner. In other words, he can carry out an analysis of patient data in a faster and more efficient way.
  • the method enables a doctor to be automatically alerted to possible medical diagnoses that are not detectable by manually reviewing the patient's data because complex relationships between medical findings need to be considered.
  • a doctor individually diagnosed for the patient medical diagnosis risks and associated diagnoses are displayed. If the physician considers that a potential diagnosis could be of high relevance in the present case of treatment, he can quickly and effectively make a determination by confirming the user prompt as to whether interactive diagnosis of symptoms is to be performed for the first medical diagnosis whether a displayed medical diagnosis is actually relevant or not. In other words, a suspicion for a particular diagnosis can be substantiated or rejected.
  • it is ensured on the whole that the interaction time of the physician with the data processing system is considerably shortened. The same applies analogously to the guideline diagnostics.
  • the user has the option, in the symptom user query of one embodiment of the invention, of selecting various medical symptoms associated with the first medical diagnosis for further analysis of the patient data. After selecting a symptom that it considers appropriate for the patient currently being examined, applying the symptom diagnostic rules associated with that selected symptom is performed at the predetermined diagnostic risk value for a particular diagnosis.
  • the symptom user query is interactive, that is, the physician can consult or exclude individual symptoms that he believes he or she is diagnosing in the diagnosis for diagnosis. This has the advantage that the doctor can interactively determine the influence of each individual symptom on the diagnosis result individually by selecting or deselecting the symptom.
  • the presence of a symptom is not clear (mild headache, mild reddening of the skin, which may also have been caused by clothing, nonspecific discomfort or symptoms that do not fit into the context of other symptoms).
  • the user has according to an embodiment of the invention in an analogous manner for the guideline diagnostics the ability to different guideline criteria, which may include laboratory values that are associated with the previously determined diagnosis to select for further analysis of the patient data.
  • a modification of the previously determined risk value for a specific diagnosis takes place, the strength of the modification depending on the respective guideline criterion.
  • the guideline user query is interactive, meaning that the physician may consult or exclude individual guideline criteria that he believes he can identify to the patient for diagnosis.
  • the previously determined diagnostic risk is modified by the execution of diagnostic-specific guideline routines.
  • the patient data is received from a second database.
  • This second database can be an external database to the data processing system, such as the database of a doctor information system.
  • medical diagnoses are issued only at a predetermined threshold.
  • the medical diagnoses are preferably output sorted according to the calculated risk level. This ensures that a user of the data processing system, i. a treating physician is not unnecessarily confronted with irrelevant medical diagnoses.
  • a threshold is selected for a diagnostic risk of 40% to be displayed to the physician, but this value can be changed by the user.
  • the first, second and third medical diagnosis risks are displayed in the form of a tachograph disc.
  • the diagnostic risk is displayed with colored gradations on the scale of the speedometer disk.
  • the primary risk is displayed as a risk probability in the form of a numerical value together with the speedometer disk.
  • a first operating element is displayed together with the first medical diagnosis risk, the first operating element being designed for a user confirmation, wherein in the case of a user confirmation via the first operating element the first medical diagnosis and / or the medical symptoms with the patient data linked stored in the second database.
  • a second operating element is displayed together with the second medical diagnosis risk, wherein the second operating element is designed for a user confirmation, wherein in the case of a user confirmation via the second operating element, the second diagnosis risk and the second medical diagnosis as a new first diagnosis risk is issued as the new first medical diagnosis on the graphical user interface.
  • the possibility is given of updating the diagnosis specified by the additional entry of symptoms in that overview which was originally generated with the output of the first diagnosis diagnosis for the first medical diagnosis together with the first medical diagnosis. This is particularly relevant in the event that not only a single medical diagnosis provided with a corresponding diagnostic risk was originally displayed, but a set of different diagnoses.
  • the risk is specified for the diagnosis for which the symptom diagnosis was made.
  • symptom diagnostics has the function of identifying other potentially relevant diagnoses that were not included in the list of initial diagnoses. This is done so that the user of symptom diagnostics will be presented with further diagnoses that correlate with the symptoms selected by the user. If the user makes the additional suggested diagnosis If he considers relevant, he can select the diagnoses and thereby add them to the list of the first diagnoses.
  • the symptom diagnostic rules are applied again to the patient data and the medical symptoms hitherto selected by the user. Thereafter, recalculating at least one new second diagnostic risk for a new second medical diagnosis is dynamically performed, followed by an updated output of the recalculated new second diagnostic risk along with the new second medical diagnostic in the graphical user interface. Finally, an updated output of the user query is made, which medical symptoms associated with the new second medical diagnosis should be used for further analysis of the patient data. The physician is thus able to immediately recognize the significance of the specific indication of a single symptom for possible diagnoses with regard to their diagnostic risks.
  • the updated output of the second diagnostic risk is for a re-updated output of the symptom user query, wherein the updated output of the symptom user query indicates which of the medical symptoms associated with the further medical diagnosis previously selected by the user for further analysis patient data is to be used, maintaining previously selected medical symptoms in the updated output of the symptom user query.
  • this further restricts the list of selectable possible medical symptoms or dynamically supplements them with other possible selectable symptoms. This is relevant, for example, if the combined evaluation of patient data and selected symptoms indicate that a possible disease exists, which, however, can only be taken into account for a diagnosis risk calculation if further symptoms not previously indicated are taken into account.
  • the symptom user query takes place in the form of a checkbox list.
  • the first and / or second database is a database external to the data processing system or the first and / or second database is contained in the data processing system.
  • the computer-implemented diagnostic support method is implemented as a plug-in of an interface, which interface can communicate with one of a plurality of physician information systems (AIS). Since the plug-in communicates with various AISs via this interface, its application is not limited to a specific AIS. The plug-in can be used for a variety of AIS.
  • AIS physician information systems
  • At least some of the laboratory values of a patient are entered automatically, e.g. via the connection to a LIMS (Laboratory Information and Management System).
  • the data transmission preferably takes place in accordance with the LOINC (Logical Observation Identifier Names and Codes) system for the encryption and transmission of data from laboratory examinations.
  • LOINC Logical Observation Identifier Names and Codes
  • all structured medical data of the electronic patient files of a doctor or a clinic are statistically evaluated. This includes the classification of patients and their associated medical data into strata (groups whose representatives are similar in certain characteristics, eg, age, sex, occupation / income, physical activity, diagnoses, etc.). By means of various data mining and inference methods, correlations between these features and the risk for the occurrence of further diagnoses are determined from the straits. Through these methods, statistical relationships can be discovered that are not yet known medicine. The won Correlation data can be used to more precisely and better define the rules for calculating diagnostic risks.
  • strata groups whose representatives are similar in certain characteristics, eg, age, sex, occupation / income, physical activity, diagnoses, etc.
  • the method further comprises the step of preparing the patient data, wherein the application of the rules to the patient data is carried out only for the processed patient data.
  • data processing includes the filtering of structured data from the patient data. This reduces the amount of data to be processed and transferred for each query and significantly speeds up the corresponding query.
  • the patient data is read from a second database and processed, which includes above all the filtering of the structured data from all existing patient data.
  • the prepared patient data is then stored in a third database, whereby the rules are applied to the patient data by accessing the third database.
  • the third database may again be a database external to the data processing system.
  • the third database is a cache memory of the data processing system, so that a query on the relevant patient data can be performed in a very fast manner.
  • this represents a significant advantage, since for corresponding queries the first and third database can be kept relatively small in scope - the amount of data to be transmitted or the number of queries to be performed is drastically reduced.
  • Another technical advantage of loading all structured patient data into the cache memory is that this "memory database" ensures that the patient data will always be in the same structure, even if the structure of the patient data in the second database, for example, depends on the AIS or LIMS used and can be structured differently.
  • At least part of the patient data is displayed in a first display window of the graphical user interface, wherein the first and second diagnosis risk for a first and a second medical diagnosis Diagnosis along with the medical diagnosis on the graphical user interface is output in a popup.
  • the application of the rules to the patient data is carried out automatically after displaying the patient data in the first display window.
  • the method is performed after opening the electronic patient record, the method further comprising the step of receiving new patient data through a user input.
  • the structured data obtained during the medical diagnosis using the method according to the invention can be used to automatically generate medical reports.
  • the system may automatically generate a medical history letter containing the information that a particular patient has arrived in practice on a particular date, the five symptoms in question have been identified in the patient and that a specific diagnosis has been made based on these symptoms.
  • the automated generation of medical reports and other administrative documents significantly increases the efficiency of a doctor's office workflow and avoids errors by manually entering the diagnoses into the medical record.
  • the invention relates to a data processing system having a graphical user interface, wherein the data processing system is designed to carry out the method for medical diagnostic support of a patient.
  • the invention in another aspect, relates to a computer program product having processor executable instructions for performing the method of providing medical diagnostic support to a patient's patient data.
  • the graphical user interface has at least a first and a second display window.
  • the method in this case comprises the step of displaying at least a part of patient data of a patient in the first display window, wherein the displayed patient data are displayed in rows in the first display window.
  • the first display window is designed for line-by-line tracking of the patient data to be displayed by means of a scrollbar.
  • access is made to a first database, this first database containing medical diagnostic objects.
  • the medical diagnostic objects are linked to patient patient data rules and are used to automatically identify individualized diagnostic risks based on the electronic patient record.
  • the first database also contains information on whether the diseases represented by the medical diagnostic objects are usually chronic or on a case-by-case basis.
  • the check is made as to whether at least one of the rules for the patient data is fulfilled. If this is the case, a display element is displayed on the graphical user interface, wherein the display element has at least one of the first diagnostic objects for which the first rule is fulfilled. If the first diagnosis thus determined is recorded in the first database as a possible permanent diagnosis (chronic illness), the output of a user query on the graphical user interface is made as to whether a medical diagnosis associated with the diagnosis object should be adopted as a chronic permanent diagnosis. If the medical diagnosis associated with the diagnostic object is to be adopted as a permanent diagnosis, the permanent diagnosis is displayed in the second display window independently of the position of the scrollbar.
  • the method further comprises storing the permanent diagnosis in the second database, which also contains the patient data, linked to the patient data.
  • diagnostic object is understood to mean any type of information that makes it possible to describe a medical diagnosis, including free-text information, which, for example, names the diagnosis by name or which describes in detail a clinical picture associated with the chronic disease
  • diagnostic objects also include the ICD codes already mentioned above or generally structured information.
  • the graphical user interface also has a third display window, wherein the method, if the medical diagnosis associated with the diagnostic object is to be adopted as a permanent diagnosis, further comprises the following steps: First, it should be noted that in this embodiment, the first Database contains information about which drugs to prescribe in the presence of a diagnosis. Furthermore, the first or a fourth database contains information about which medicaments and associated medicament objects contain which active ingredients. Further, the electronic patient record contains information about which medications have been prescribed to the patient in the past.
  • the first database is searched for drugs to be prescribed in the presence of this diagnosis. Further, these drugs are assigned to drugs (or drug-presenting objects) and the electronic health record is analyzed to see if any medications containing this drug have been prescribed in the past. If this is the case, another display will be made. displayed on the graphical user interface, said further display element has at least one of the drug objects that have been previously prescribed and can also be used for the treatment of a long-term diagnosis of the patient. Thereafter, the output of another user query on the graphical user interface, whether a drug associated with the drug object is to be adopted as a preparation for long-term medication. If the medicament-linked medicament is to be taken over as a preparation for continuous medication, then after a corresponding user confirmation, a permanently visible indication of the permanent medication in the third display window takes place, likewise irrespective of the position of the scrollbar.
  • the medical diagnosis associated with the diagnostic object is to be taken as a permanent diagnosis, that is, if a chronic medical condition is considered safe by the physician, then the next step is a review of whether medications already used to treat the chronic disease according to the patient record, that is the patient data previously prescribed to the patient.
  • the system detects a relevant chronic illness and if drugs or medicines are found in the individual patient file that match these chronic illnesses, the physician is advised to enter the respective preparation in the section "Continuous medication" in the third display window Condition before a diagnosis is suggested to the physician as a permanent diagnosis, the check can still be made as to whether the calculated diagnosis risk exceeds a threshold value, but this query may also be absent in other embodiments of the invention, Again, an expensive and time-consuming search for corresponding medicines is again omitted or agents in the patient data, which in turn enables the physician to perform an analysis of the patient data stored in a corresponding database in a faster and more efficient manner, and also ensures that the doctor, if he or she falsifies has made an individual case diagnosis, although it would actually be apparent from the patient record that it is a chronic disease, receives an indication of the presence of a chronic disease and possibly a permanent medication.
  • the permanent medication confirmed by the doctor can be stored linked to the patient data as a permanent medication.
  • the first and / or second and / or fourth database is a database external to the data processing system, but it is also possible for the first and / or second and / or fourth database in the data processing system itself is included. According to a preferred embodiment, however, the patient data is located on a second database, for example a doctor information system.
  • the first database is identical to the second and fourth databases and is provided, for example, together with the data processing system mentioned above.
  • the second database can be contained in a doctor information system, wherein this doctor information system can carry out the method according to the invention as described above.
  • the doctor information system accesses a web service via a network, which can be called up by a server.
  • This web service provides, for example, a service in the form of a servlet, which makes it possible to apply the method according to the invention to the patient data.
  • the web service may access the first and fourth databases representing external databases regarding the doctor information system, but the web service may be executed either on the doctor information system or server side on a server operated by a medical service provider , The first and fourth databases can be assigned to this server of the medical service provider.
  • the method according to the invention can also be carried out on an external server, the graphical user interface being part of a client, via which patient data are entered and which can be operated by a corresponding doctor.
  • the rules for determining the first diagnosis risk with a time constant for a maximum age linked to the patient data includes at least the date and possibly other time information indicating when a data entry was made into the patient's electronic file, wherein the data entry may constitute the creation of a particular diagnosis, the prescription of a drug, or the performance or billing of a medical examination ,
  • the check as to whether the patient file contains indications of the existence of a diagnosis, in particular of a permanent diagnosis is applied only to the patient data having a younger time stamp than the maximum age.
  • the doctor can set that only those diagnoses, medications and treatments in the patient record play a role in the diagnosis, which have been entered in the file within a predetermined period of time. In addition, it can be avoided that similar diagnosis that occurred several times in the long term in the past, be misinterpreted as the presence of a chronic disease.
  • This predetermined period is initially predetermined by the system, but can also be adapted by the physician in a corresponding manner. In addition, this predefined period is preferably dependent on the type of medical diagnosis, so that the time constant for each query condition is set individually. Nonetheless, it is possible to set a global upper limit as to the age of eligible patient data.
  • the check of whether some of the rules for calculating first diagnostic risks can be applied to the patient data is made automatically after displaying the at least a part of the patient data in the first display window.
  • the method is preferably performed in real time, the method further comprising the step of receiving new patient data through a corresponding user input.
  • this offers the advantage that a treating physician is in principle immediately and directly either after opening the patient file or after entering appropriate patient data in the patient's file reliably informed whether his patient is at risk of having a chronic disease.
  • the check of whether at least one of the first rules for the patient data is fulfilled is carried out in the order of decreasing diagnosis risk of the respective rule.
  • Such a prioritization may also include, for example, only those rules for a specific diagnosis that are associated with the highest diagnostic risk being performed. If a rule executed according to this prioritization applies, the query for further rules for the same diagnosis can be omitted, since a higher risk value is no longer to be expected, even if further rules for this diagnosis apply. This can further reduce the computation time required for risk calculation.
  • the invention relates to the function of prescription medication.
  • the drug objects in the first database are, according to this embodiment, stored with information on package size (number of dosage units present in the package, measured, for example, in milliliters, drops, tablets, or others).
  • each drug object is provided with information about the standard dosage, that is information about how many dosage units per day, week or month usually need to be taken.
  • this additional feature reads the medication items that were previously prescribed to the patient, and stores information about the size of the package and the default dose prescribed with these medication items.
  • the medication prescription aid function can calculate how long the prescribed drug is still sufficient and whether the doctor may need to prescribe another pack.
  • this medication prescription aid primarily relates to permanently prescribed medications.
  • the display of the remaining time until the prescription of another pack is required is preferably displayed in the form of a color-coded scale or dial, where red indicates that the drug must now be re-prescribed, green that the currently prescribed pack is still sufficient and yellow that the new prescription is discretionary of the doctor.
  • the invention relates to a data processing system having a graphical user interface, wherein the data processing system is designed to carry out the method for displaying patient-related diagnoses of chronic diseases.
  • the invention relates to a computer program product with instructions executable by a processor for carrying out the method according to the invention for displaying patient-related diagnoses of chronic diseases.
  • FIG. 1 shows a block diagram of a data processing system according to the invention
  • FIG. 2 is a schematic view of a graphical user interface
  • FIG. 3 shows a flowchart of a method for displaying patient-related diagnoses of chronic diseases
  • FIG. 4 shows a method for medical diagnostic support for patient data of a patient
  • FIG. 5 Steps of a method for medical diagnostic support for patient data of a patient.
  • FIG. 6 shows a database table with rules for calculating the first diagnosis risks
  • FIG. 7 shows a database table for symptom diagnostics
  • FIG. 8 shows a database table for guideline diagnostics
  • FIG. 9 shows a computer-readable storage medium.
  • FIG. 1 shows a block diagram of a data processing system 100 according to the invention.
  • the data processing system 100 comprises a processor 104 and input means 102, such as e.g. a mouse, keyboard, etc. Medical devices may also serve as input means by means of which corresponding medical image and / or measurement data of a patient can be recorded and stored.
  • the data processing system 100 has a memory 116 in which there is a computer-executable code for an application program, for example for carrying out the method according to the invention.
  • the data processing system 100 has a graphical user interface 106, which is output on a corresponding display device 108.
  • This display device 108 may be, for example, an LCD or CRT screen.
  • the data processing system 100 can communicate with databases 122, 132 and 142, for example via the network 118.
  • the interface communicates with the physician information system AlS using a data encryption method, eg a hash method.
  • the databases 122, 132, and 142 may also be part of the data processing system 100 itself.
  • the code for execution by the processor 104 can also be retrieved from a server 144, in which case the code for carrying out the method according to the invention is provided, for example, by means of a web service. The code may be executed either on the server 144 or in the data processing system 100.
  • a treating doctor first opens a patient file.
  • This patient record contains patient data 134 stored in the database 132.
  • the patient data 134 are first transmitted to the data processing system 100 via the network 118.
  • the last-entered patient data are then displayed line by line in the display window 114, this display window having a scroll bar. With this, the doctor is able to scroll through all entries of the patient data by moving the scrollbar.
  • the data processing system 100 and / or its processor 104 will initially proceed to open the patient record in such a way that is accessed via the network 118 to the database 122.
  • This database 122 contains medical diagnostic objects 124.
  • the data processing system 100 may determine if there is a high likelihood of a patient's chronic illness being present.
  • the first database 122 contains information about which of the medical diagnostic objects occur or can occur as long-term diagnoses. If one of the rules 128, which determines the diagnosis risk for the existence of a specific diagnosis based on the patient data 134, is met, and if the thus determined diagnosis object is stored in the first database as a possible permanent diagnosis, the display element, for example a popup, is displayed. on the graphical user interface 106. This pop-up contains further information regarding the possibility of the presence of a chronic disease, and thus in particular information contained in the medical diagnostic object 124.
  • This can be, for example, an ICD code or the name of a corresponding chronic erythrocyte. to be ill.
  • additional information and, where appropriate, also links in the form of hyperlinks to other databases can be specified, by means of which the attending physician can continue to learn in detail about the corresponding chronic disease.
  • the data processing system 100 offers the attending physician the option of taking the corresponding chronic illness into the category "permanent diagnosis" This means that this diagnosis is permanently displayed in the display window 110 of the graphical user interface 106, irrespective of a scrolling movement within the various rows of the patient data in the display window 114. If this is confirmed by the physician, then this permanent display is preferably performed the medical diagnosis, for example in the form of an ICD code, in the display window 110 and also a storage of this display option for the patient in the patient's file in the database 132. In other words, the patient data 134 with the permanent diagnosis "chr onic disease ". The next time the medical records are opened by the attending physician, the data processing system 100 can thus display this permanent diagnosis permanently in the display window 110.
  • the data processing system 100 first of all accesses the database 122 in which information is available which active substances are generally to be presumed in the presence of a specific diagnosis.
  • the database 142 is accessed.
  • the database 142 includes medical drug objects 136 and information about which drugs 138 are contained in which drugs. Access to the database is to identify those drug objects that, according to the drug and drug mapping information, contain the drugs to be prescribed according to the information from the database 122 in the presence of a particular diagnosis.
  • the patient data 134 is analyzed as to whether one or more of the medicaments thus determined been prescribed to the patient in the past.
  • a corresponding user query will be output on the graphical user interface 106.
  • the determined medical drug objects for example in the form of drugs or drug names, possibly represented by PZN numbers, whereupon the attending physician can select from the list made available to him one or more drugs, which he used for long-term medication of the respective patient wants to add the patient record. After selecting one or more drugs, these are then displayed permanently in the display window 112 of the graphical user interface 106.
  • the list of preparations proposed to the physician as a permanent medication is not restricted to those preparations which have already been prescribed, so that the described function can also be used to determine suitable medicaments for the treatment of a chronic disease have not yet been prescribed.
  • the data processing system 100 still further allows a treating physician to make diagnoses in a reliable manner.
  • the data processing system 100 can in turn access the database 122 in order to retrieve there rules 128 with calculation of diagnostic risks for medical diagnostics, wherein the medical diagnoses are also stored linked with medical symptoms 130 in the database 122.
  • this diagnostic risk may be displayed on the graphical user interface 106, again in the form of a pop-up to the physician, for example.
  • the diagnosis risk is preferably presented to the physician together with the medical diagnosis.
  • different risks for different medical diagnoses, if present, can be displayed here, preferably according to the risk probability.
  • a user query is displayed on the graphical user interface 106 as to whether an interactive symptom diagnosis is to be performed 610 for this medical diagnosis and whether a guideline diagnosis is carried out additionally or instead of the symptom diagnostics 646. If confirmed by the physician, an output of a symptom user query is made which of the medical symptoms 130 associated with the medical diagnosis should be used for further analysis of the patient data 134. For example, for a diagnosis selected by the physician, various disease symptoms are displayed in the form of a list of checkboxes, each time a checkbox is activated, that is to say the presence of a disease symptom, a dynamic update and recalculation of the diagnosis risk for the corresponding diagnostic result.
  • the diagnostic findings can also be supplemented with even more precise diagnostic findings on the graphical user interface. For example, if a medical diagnosis was initially labeled as a "60% risk for diabetes," the additional specification on the graphical user interface 106 indicates that the risk for "Type I Diabetes is 80%" and the “Risk for Diabetes Type II at 40% ".
  • a treating physician now considers one of the medical diagnoses to be secure, he can confirm this in a corresponding manner and thus store it in the database 132 linked to the patient data 134.
  • FIG. 2 shows a schematic view of a graphical user interface 106.
  • the graphical user interface 106 has display windows 110, 112 and 114.
  • the display window 110 is used for Indication of permanent diagnoses, whereas the display window 112 is designed to display continuous medications.
  • the display window 114 is used for line by line display of patient data, wherein preferably when opening a patient file only the few, most recently made entries are displayed in the patient record. Nevertheless, access to further entries is possible by moving a corresponding element 202 of a scroll bar 200 vertically up and down so that it is possible to scroll through the various entries of the patient file. By clicking on arrows 204, it is also possible to make a leaf in the form of interlaced lines. Further shown in FIG. 2 is a popup 206 in which further information can be provided to a user. Such a popup can be, for example, a display element with diagnostic objects, queries, medication objects, or else diagnostic findings in connection with medical diagnoses, a window for performing interactive symptom diagnostics, or a corresponding query window.
  • FIG. 3 shows a flowchart of a method for displaying patient-related diagnoses of chronic diseases on a graphical user interface of a data processing system.
  • the method begins in step 300 with displaying the patient data in a display window 114, this display window having a scroll bar and only a portion of the patient data displayed in that display window.
  • step 302 reading and applying rules to the patient data occurs, the rules containing query conditions and applied to the available patient data of a patient.
  • the rules 128 are stored linked to medical diagnostic objects 124 in a first database 122.
  • the structure of the rules is shown in detail in FIG.
  • step 304 the check is made as to whether at least one of the rules for the patient data has been fulfilled.
  • step 304 the method continues in step 306 with the display of a display element in the graphical user interface, this display element at least one of the diagnostic objects has For example, an ICD code, which is part of the corresponding diagnostic object for which the rule is met.
  • diagnosis determined is a possibly chronically occurring diagnosis (possible permanent diagnosis)
  • step 308 a user query is issued on the graphical user interface as to whether a medical diagnosis associated with the diagnostic object is to be taken over as a chronic permanent diagnosis. If the medical diagnosis is not accepted as a chronic permanent diagnosis, then the method jumps back to step 304, where it is checked whether another rule for the patient data is fulfilled.
  • the steps 304 to 308 are thus performed cyclically for all the rules.
  • step 310 the indication of the permanent diagnosis, for example in the form of the ICD code, takes place permanently in a second display window irrespective of the position of the scroll bar in the display first display window.
  • step 312 access to the first database 122, in which the medical diagnostic objects are stored with information about which drugs are to be presumed in the presence of a diagnosis.
  • the information about which drugs to administer in the case of a particular diagnosis may alternatively be stored in a fourth database (142). If at least one corresponding active ingredient has been determined, an access (312) to another database of medical drug objects takes place, wherein the medical drug objects are linked to information about contained active ingredients. In this step, all drugs containing at least one of the previously identified active ingredients are determined. In step 314, a check is made on the patient data as to whether the previously determined drugs have already been prescribed to the patient once.
  • this step may also be associated with a review of the time constant of the prescription of the drug, which may be determined from the patient data 134. If the prescription of the drug has been around for a very long time, the drug in this case will be disregarded in 314. Was not the drug yet or if its prescription is too long ago, the method jumps back to step 304, where it is further checked in a cyclical manner with steps 304, 306 and 308 whether at least one of the other rules regarding a chronic illness is fulfilled.
  • step 316 the method continues to display a graphical user interface display element that suggests at least one of the drug objects to be selected by the user, with the proposed drug objects having at least one drug against the patient User-confirmed continuous diagnosis and already prescribed for the patient. Also, only parts of the data associated with a medicine object may be displayed, such as a central pharmaceutical number or a drug description or a drug name.
  • the query in step 318 is to allow a physician to decide whether to use the indicated medication for long-term medication. If it does not want to do so, the process returns to step 304.
  • step 318 is followed by step 320 of displaying the drug in a third graphical user interface window 112 in a persistent, ie independent manner from the position of the scrollbar. After step 320, the method again jumps to step 304.
  • step 300 it is also possible to carry out a data processing of the patient data with an intermediate step 301.
  • those data which are structured are filtered out of the patient data.
  • These structured data are then held in a corresponding memory, for example a cache memory, indicated by the reference numeral 140 in FIG.
  • a first diagnosis risk there is the primary risk for each diagnosis in the system or, after applying the rules, a first diagnosis risk.
  • the risk value which indicates the probability with which a diagnosis is chronically pronounced in the event of its occurrence
  • the risk for the existence of a chronic diagnosis can be predicted even more precisely.
  • it is possible to specify a specific threshold for this risk so that only then diagnoses are proposed as possible permanent diagnoses to the user whose risk for the presence of the chronic form of a diagnosis is above this second threshold.
  • FIG. 4 shows a flow chart of a method for medical diagnostic support for patient data of a patient by means of a data processing system.
  • FIG. 4 a shows the method of calculating the first diagnoses and diagnostic risks by applying rules to the patient data.
  • FIG. 4b shows the further specification of the diagnostic risk of a previously calculated diagnosis, eg a diagnosis, which was calculated in FIG. 4a by means of symptom diagnostics.
  • FIG. 4 c shows the further refinement of the diagnostic risk of a previously calculated diagnosis, eg a diagnosis, which was calculated in FIG. 4 a or 4 b, by means of guideline diagnostics.
  • the method begins in step 400 with reading patient data from a database.
  • step 402 of data preparation is available, wherein either after step 402 or directly after step 400 access to the first database takes place in order to with rules for calculating diagnostic risks for medical diagnoses.
  • step 406 the check is made as to whether at least one of the rules is applicable to the patient data. If this is not the case because, for example, there are too few patient data or because the available patient data are too old, the method ends in step 414. However, if at least one of the rules is applicable in step 406, step 408, in FIG wherein the rules are applied to the patient data, thereby calculating a diagnosis risk for a first medical diagnosis. This first medical diagnosis is output in step 410 along with the first diagnostic risk in the graphical user interface. After step 410, a check is made in step 412 to see if all risks have been calculated for all possible medical diagnoses. If not, the process continues with steps 408 and 410, again followed by step 412.
  • FIG. 4 does not show the additional possibility of limiting an output of diagnostic risks to a corresponding minimum probability, from which corresponding diagnostic risks are first output on the graphical user interface.
  • step 416 the process continues with the issuance of a user query as to whether the diagnosis designated with a certain risk can be adopted as a confirmed diagnosis in the patient data. If this is not the case for any of the calculated diagnostic results, the method ends in step 414. However, it is also possible to directly report one of the displayed diagnostic results, for example at a high probability of diagnosis of more than 90%, either automatically or after confirmation by the attending physician storing the patient data in the corresponding patient database in step 420, after which step 420, the method ends in step 414.
  • step 416 it may be possible to provide the physician at step 418 or 436 the opportunity to perform interactive symptom diagnostics or guideline diagnostics. If the physician does not wish to carry out such an analysis, after step 418/436, the already mentioned step 420 takes place with the saving of the diagnosis as a reliable diagnosis, linked to the patient data in the mask. tientensteinbank. This in turn is followed by step 414 with the completion of the process.
  • step 418 If the physician wishes to perform interactive symptom diagnostics in step 418, the method continues in step 422. However, if the physician wishes to perform interactive guideline diagnostics in step 436, the method continues in step 438.
  • steps 416 and 450 thus serve to give the physician a choice between a) directly adopting one of the diagnostic results as a confirmed diagnosis, b) discarding all diagnostic results, or c) performing an additional interactive symptom diagnostic or guideline diagnostic for one or more of the first diagnostic results.
  • a checklist of symptoms is output, which are linked to the medical diagnosis selected in step 418 in the first database.
  • This may be done, for example, by accessing the first database at step 422, wherein a query of the first database for possible symptoms for a given and selected medical diagnosis is made.
  • those diagnoses that correlate statistically significantly with certain symptoms are stored linked to each other, the link also containing information about the source of literature underlying this linkage.
  • FIG. 800 shows a database table in which several symptoms are stored linked to a specific diagnosis ID 68. These symptoms associated with the diagnosis to be specified are then transmitted to or retrieved from the data processing system and displayed in step 422 to the user in the form of a checklist.
  • the user may now select one or more of the symptoms, or alternatively provide additional details on symptoms, for example in the form of numeric inputs. For example, if a symptom is "high blood pressure," the physician may specify this by adding a corresponding blood pressure value for that symptom.
  • the link between symptoms and correlating diagnoses is thus used as described above to build up the query elements, eg checkboxes, dynamically from the database for symptom diagnostics of a specific symptom. However, the link also serves to find further diagnoses 628 based on the current symptom selection of the user 642, which correlate with the respective symptom selection.
  • an updated calculation of the diagnostic risk for the currently selected diagnosis is dynamically performed by applying the symptom diagnostic rules 800 to the previously determined diagnostic risk.
  • the correlation between the selected symptoms and the diagnoses is, as previously mentioned, literature-based and stored in the first database.
  • the additional diagnoses can be adopted by the user according to a further embodiment of the invention in the list of first diagnoses (suspected diagnoses hypertension and CPOD in Fig. 5-1, for example, after the symptom diagnosis by the suspected diagnosis of renal insufficiency stage II supplemented by selection of the user).
  • the calculation of a second diagnostic and a second diagnostic risk mentioned in 428 is also done by applying the symptom diagnostic rules in table 800 and may include several second diagnoses correlating with the symptom selection with second diagnostic risks, as shown in the display window 510. For the sake of simplicity, only a single second diagnosis is shown in 426 in FIG. 4b and only a single third diagnosis in 442, and FIG. 4c.
  • Figure 630 shows, however, that multiple diagnoses can correlate with the first diagnosis.
  • Step 426 involves the following sub-steps: If the symptom diagnosis was selected to further specify the risk for a patient's stroke of 55% determined by the rules, then in the course of the symptom diagnosis, first of all all symptoms are read out from the table 800 that corresponds to the ID of the patient Diagnosis stroke are stored within a row. A data entry with the diagnosis ID for stroke thus corresponds to a selection element, eg a check box.
  • correlated diagnoses are displayed 628 for all user-selected symptoms, such as shown in 624.
  • FIG. 4b assumes only one further diagnosis, which is referred to as a second diagnosis with a second risk. The risk of the second diagnosis is calculated analogously from the first risk determined by the rules 128 for this second diagnosis, which was additionally modulated by the current symptom selection according to table 800.
  • step 432 After the physician has entered appropriate symptoms in step 424 and one or more second medical diagnoses and diagnostic risks have been displayed in steps 426 and 428, the physician obtains in step 432 the possibility of confirming a diagnosis, which in step 428 is associated with a diagnosis Diagnosis risk was issued. If the physician does not confirm any of the diagnoses in step 432, ie rejects all suggested diagnoses, step 432 is followed by step 416, which returns to the physician the original display window, in which the diagnostic risks calculated in steps 408 to 412 for different Diagnoses are displayed.
  • step 434 the latter now being specified to the physician in step 416 together with the further ones calculated in steps 408 to 412 Diagnoses and their diagnostic risks are provided for selection for storage associated with the patient data, further interactive symptom diagnostics or even complete discarding of all calculated diagnostic risks.
  • an intermediate step 402 may also follow step 400, in which a data processing of the patient data may take place.
  • FIG. 4c Further clarification of the diagnostic risk by the guideline diagnostics FIG. 4c and the calculation of a third diagnostic risk are analogous to the symptom diagnostics shown in FIG. 4b.
  • the display 438 is made of those guideline criteria which are stored linked to the diagnosis selected by the user in the first database. Some or all of these guideline criteria may be selected by the user. Guideline criteria that correlate with the diagnostic objects are also stored in the first database linked to the diagnostic objects.
  • the diagnostic risk is specified and a third diagnosis with associated diagnostic risk is output 444, 644.
  • FIG. 4c assumes a third diagnostic risk.
  • FIG. 5 shows various outputs on a graphical user interface in the event that a medical diagnostic support for patient data of a patient is performed by the data processing system. This is therefore preceded by the fact that a corresponding patient has been selected by the attending physician and thus the patient data has been made available to the data processing system.
  • the data processing system then automatically analyzes the patient data and, as shown in FIG. 4, applies rules to the patient data to calculate at least a first diagnostic risk for a first medical diagnosis.
  • the data processing system Based on the health profile of the patient, ie the patient data stored as structured data in the individual patient record on the physician's computer (including age, sex, ICD diagnoses, prescribed medication, laboratory values, stored findings and symptoms)
  • the data processing system compares the probability or relative frequencies of relevant comorbidities or frequently associated diseases on a transparent guideline and literature basis. It compares these with the already known diagnoses and displays the previously unlisted or recognized diseases in a patient-specific manner, sorted by probability , at.
  • the screen output 500 shows such an output of a diagnosis risk in the form of a "tachometer disc” 602.
  • the tachograph disc can be visualized equally well with probabilities and / or relative frequencies.
  • the tachometer disc consists of a scale with colored graduations, whereby the tachometer disc preferably has red graduation parts, It has a medium probability that it has yellow scales and, with low probability, green scales, which allows a clinician to quickly and easily visually assess the likelihood of a relevant comorbidity, and to specify a likelihood for a diagnosis risk in the center of the tachograph disc, the primary risk is expressed in the form of a percentage probability 604 or a relative percentage frequency 604.
  • the diagnostic risk is represented by the anord 600 in the form of Tachoolin and numerical value shown.
  • the physician is given the option of hiding the display 500 for a certain period of time by actuating the "remember later" button 606 or, by pressing the “do not remember” button, completely masking out the display 500 of the probability of relevant comorbidities.
  • the display element 500 thus serves the purpose of informing the doctor in a clear and general manner about whether or not there is a particular diagnostic risk for a relevant medical diagnosis at all. A clarification as to what this medical diagnosis looks like or whether there are several possible relevant medical diagnoses is not given by the display element 500.
  • the criteria for determining a certain probability are displayed to the physician in a manner that is related to the indication, if desired, as shown in display element 502.
  • This display including the literature and study sources used, forms the basis of the respective diagnostic procedure (application of the rules for determining the first diagnosis risk, symptom diagnostics and guideline diagnostics).
  • the physician reaches the display element 504 by pressing the button 608 "more", on which a summary of the patient's name "Maria Test 74" ( Reference numeral 618) finds possible comorbidities.
  • the possible first diagnosis is presented in the form of a textual description together with the respective ICD 10 code (reference numeral 16), together with the respective first diagnosis risk in the form of a tachograph disc (reference numeral 612).
  • the first diagnosis or diagnoses are referred to in the display 504 and the following displays as the basic risk.
  • the physician is given the option of using the selection elements 620 to determine whether these displayed possible diagnoses individually represent only a suspected diagnosis or a confirmed diagnosis.
  • the diagnosis can be saved individually or all diagnoses can be stored at one go, i. into the patient record.
  • the "show all" button 614 is used to indicate other possible diagnoses whose diagnostic risk is below a predetermined threshold, for example, 40% in the present case Here only possible diagnoses are displayed, which have a diagnostic risk> 40%.
  • the doctor wishes to investigate the respective diagnosis risk, he can call up the indication-related, in each case literature-based symptoms and compare them with findings of the patient or supplement these with a checklist. This is done by the physician clicking on the corresponding "GO" button in column 610 to perform a symptom diagnosis for the respective possible diagnosis, for which the sources of symptom diagnostics are also stored and transparently displayed to the physician, as illustrated with display element 506. Structurally stored findings are recorded and "preselected" in a different color coding. If the doctor moves the mouse over such a marked advertisement, he will be shown a text with his own file source (for example, free text input "consultation of 01.11.2008” or “laboratory value of 15.10.2007”). If the doctor clicks on the corresponding "GO” button in column 646, a guideline diagnostics will be carried out for each possible diagnosis.
  • the physician By pressing the "GO" button in the display element 504, column 610, the physician first arrives at the symptom diagnostics display element 508.
  • the display element 508 has a button 622, via which the physician arrives at the display element 506.
  • the display element 622 has a Checklist 624 with various symptoms (findings) which are symptomatic of the possible diagnosis 616 for which the corresponding "GO" button has been selected in display element 504.
  • a diagnosis suggestion 626 is displayed in the display element 508 for the selected symptoms together with a corresponding match in the form of a newly calculated diagnosis risk as a tachograph disc.
  • the display element 510 With each further selection of one of the check elements, an update of the diagnosis proposal and corresponding match takes place, which in turn results in a probability-sorted arrangement 628 of diagnostic risks.
  • the first diagnosis proposal made is also specified more precisely by selecting further findings.
  • display element 508 only a determination was possible that, if necessary renal insufficiency stage 1 or stage II is present, whereas in display element 510 a new calculation of diagnostic risks for various medical diagnoses could be made on the basis of a more precise specification of the present findings, whereby renal insufficiency stage III or stage IV are now eligible as additional diagnoses comes.
  • the probabilities in the form of the tachograph discs 628 have been shown calculated in the display element 510 in a more precise manner.
  • the attending physician may supplement the necessary symptoms / findings - by questioning the patient, examining or adding information already known in the checklist.
  • the diagnosis suggestions appear together with ICD 10 codes, i. in plain text and coding which, according to the literature cited, i. appropriate symptom diagnostic rules, correlate with the findings described.
  • the display changes dynamically, i. the degree of filling of the previously described tachograph disc and insertion of the plausible ICD diagnoses, depending on further findings or degree of correlation.
  • the respective diagnostic proposal can be taken directly into the central overview, the process can be performed with one or more diagnoses.
  • Such a precise central overview is shown with display element 512.
  • the display element 512 again shows the name of the patient 618 and the possible diagnoses 616.
  • the central overview now allows the physician to display all comorbidity probabilities (button 614), and discard appropriate diagnoses (click on the cross 639, and if necessary reactivate later) or to save all the displays (click on Element 634). All diagnoses can also be discarded at once (click on item 636) or it All diagnostics and displays can be taken over by clicking on element 638. In the latter case, the possible diagnoses and symptoms are not transferred to the patient database, but the system only remembers the view 512 so that the doctor can later restore this view identically.
  • a further alternative is to let the preselection "suspect" 620 persist until a threshold probability, which is preferably very high (over 90%), is exceeded, at which point the choice automatically changes to "secured".
  • the physician can have the respective proposed guideline diagnostics displayed and run through to conclusively secure the diagnosis.
  • a corresponding display window is given by the display element 514.
  • a display window 516 is opened, in which the corresponding guideline diagnostics with corresponding literature sources for the necessary or recommended diagnostics and their interpretation are named.
  • correlating indications are displayed and again provided with a degree of graphic correlation. The most plausible diagnosis (or another) can be taken directly into the overview and in sequence into the file.
  • FIGS. 6, 7 and 8 show in simplified form the database tables on which the individual risk calculation methods are based in accordance with a preferred embodiment of the invention.
  • Database table 700 in FIG. 6 contains the rules 128 which are used to calculate the first diagnosis risks immediately upon opening the patient record.
  • Each rule has an ID (column 702), a value that indicates how much the primary risk changes if the rule can be applied to a patient (Column 716), and a diagnosis associated with the rule, which has a Diagnostic ID ( Column 716) is identified in table 700.
  • the table contains additional columns that contain conditions for applying the rule, ie eg the medication the patient has taken so far (column 704), ICD codes (column 706), LEZ codes (column 708), the age (column 710) and the sex (column 712) of the patient.
  • the database table 700 given corresponds to a preferred embodiment of the invention, further embodiments with additional or occasionally deviating features are possible.
  • Not every feature requires a data value to be present in a rule for example, Rule 1988 has no value for an ICD Code.
  • a specific diagnosis eg the diagnosis with the ID 23, can be assigned several rules (rule IDs 1987-1989). If a rule can be applied to a patient, it will modify the patient's primary risk of having a particular diagnosis. For example, if Rule ID 1987 applies to a patient, it increases their ID 23 risk by 15.23%.
  • the diagnostic risks may also be given relative values, eg "x 1, 2.” Such values should be understood to mean that the risk of diagnosis, when the rule is applicable, is calculated by multiplying the primary risk by the risk factor Rule-related diagnosis by factor 1, 2. A rule is applicable if all conditions in each column are met, so Rule 1987 is then applicable and modifies the level of diagnosis risk for ID 32 diagnosis if the patient is male is between 35 and 45 years of age, if the patient's electronic patient record has been previously marked with the ICD Code 706 and the LEZ Code 54. Whether the patient is taking certain medications is not considered in this rule.
  • the rules for each diagnosis are applied to the patient data in order of the magnitude of their effect on the primary risk. As soon as a diagnosis is made, the application of the rules for this diagnosis is aborted. This has the background that if the rules are ordered by the magnitude of the value in column 716 and, for example, rule 990 is true for the diagnosis 23, then there is no advantage in enforcing the 1987 and 1988 rules since they are lower Effect on the primary risk.
  • FIG. 7 shows a detail of a simplified database table according to a preferred embodiment of the invention, which is used for symptom diagnostics.
  • Table 800 which contains symptom diagnostic rules to further clarify the diagnostic risk
  • one or more symptoms via their Symptom IDs 802 are associated with a diagnosis via their diagnostic ID.
  • Diagnostic ID 68 is associated with multiple symptoms (ID 1321-1323) in the example shown. If the diagnostic method according to the invention has determined a diagnosis risk for a particular illness after opening a patient file, eg a risk of 60% for diagnosis with ID 68, and the user has selected performing an interactive symptom diagnosis, then the user is initially provided with a selection presented with the first diagnosis associated symptoms.
  • the descriptions of all the symptoms associated with ID 68 diagnostic according to Table 800 would be suggested to the user for selection.
  • the entries (rows) of the table 800 thus correspond in each case to a graphical selection option by the physician on a display.
  • the option is implemented in the form of a check box. That is, the user would be presented with the description 804 of the symptoms 1321-1323 in the form of checkbox elements of a graphical user interface if he had previously selected to perform interactive symptom diagnostics in order to further clarify the diagnosis risk for the diagnosis 68.
  • the first diagnostic risk which is the initial value for the patient Symptom diagnostics, modified.
  • the result of this modification is a second, more precise diagnostic risk. For example, selecting the symptom with the ID 1322 increases the first diagnostic risk by 12.9%. By selecting the symptom with the ID 1324, on the other hand, the first diagnosis risk is multiplied by the factor 1, 22. Symptom diagnostic rules thus serve to further specify the previously determined diagnostic risk by including the presence of certain symptoms.
  • FIG. 8 shows a detail of a simplified database table according to a preferred embodiment of the invention, which is used for guideline diagnostics.
  • Table 900 identifies one or more symptoms and laboratory findings with guideline criteria IDs 902 as diagnostic criteria.
  • Diagnosis ID 68 is thus associated with guideline criterion IDs 1421-1423.
  • the user would see on a graphical user interface those guideline criteria 1421-1423 which, according to database table 900, are linked to one another are.
  • the precision of the diagnosis can be further improved, and the new diagnostic risk determined is returned as the third diagnostic risk.
  • guideline diagnostics include the ability to formulate guideline routines (values of column 904 for entries ID 1426-1430) specifically for a diagnosis whose risk is to be specified.
  • These guideline routines may include, for example, complex Boolean or arithmetic functions that are applied to the data provided by the user by selecting relevant guideline elements on a graphical user interface. For example, a guideline routine could interrogate the presence of two particular guideline criteria in the presence of a given laboratory value and, if the query conditions were applied, modify the diagnostic risk for which the guideline diagnostic is performed until then.
  • the laboratory value could be applied to the diagnostic risk as a multiplication factor if the risk level correlates directly with the laboratory value.
  • the guideline routine thus checks whether the required guideline criteria constellation is given and causes a speaking modification of the previously known diagnostic risk according to a calculation routine contained in the code of these guidelines routines and therefore does not appear in the database table.
  • the fact that the guideline routine can be adapted specifically for each diagnosis without having to change the database schema results in a high degree of complexity for calculating the diagnosis risk.
  • Table 900 contains guidelines criteria that serve to further clarify the previously determined diagnostic risk by including the presence of certain guideline criteria.
  • the guideline diagnostic 900 also includes diagnostic-specific guideline routines.
  • FIG. 9 uses JavaScript code to run the guideline routines in a user's browser.
  • other embodiments of the invention may use any other programming language for the implementation of the guideline routines.
  • Another possibility is to propose guideline substances according to organ systems, if a guideline diagnostics, as described in display element 514 of FIG. 5, is carried out.
  • a guideline diagnostics as described in display element 514 of FIG. 5
  • the function of displaying a time range of drug packages is extended to the extent that - when presenting chronic diseases - is accessible - in NetflixVerowski of recommended drugs - despite a given indication on the basis of literature - ordered by organ systems, the doctor recommended lead substances can be displayed to ensure adequate patient care.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Business, Economics & Management (AREA)
  • General Business, Economics & Management (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

L'invention concerne un procédé mis en oeuvre par ordinateur pour apporter une aide au diagnostic médical relativement à des données (134) d'un patient par l'intermédiaire d'un système de traitement de données (100), ce système de traitement de données présentant une interface graphique utilisateur (106). On utilise une base de données contenant des règles pour calculer des risques de diagnostics.
PCT/EP2009/065333 2008-11-19 2009-11-17 Procédé mis en oeuvre par ordinateur pour l'aide au diagnostic médical WO2010057891A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP09752195A EP2347358A1 (fr) 2008-11-19 2009-11-17 Procédé mis en oeuvre par ordinateur pour l'aide au diagnostic médical
US13/128,185 US8548827B2 (en) 2008-11-19 2009-11-17 Computer-implemented method for medical diagnosis support

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP08169432.5 2008-11-19
EP08169432A EP2192510A1 (fr) 2008-11-19 2008-11-19 Procédé destiné au soutien du diagnostic médical

Publications (1)

Publication Number Publication Date
WO2010057891A1 true WO2010057891A1 (fr) 2010-05-27

Family

ID=40521984

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2009/065333 WO2010057891A1 (fr) 2008-11-19 2009-11-17 Procédé mis en oeuvre par ordinateur pour l'aide au diagnostic médical

Country Status (3)

Country Link
US (1) US8548827B2 (fr)
EP (2) EP2192510A1 (fr)
WO (1) WO2010057891A1 (fr)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130054264A1 (en) * 2011-03-04 2013-02-28 Sterling Point Research, Llc Systems and methods for optimizing medical care through data monitoring and feedback treatment
US11676730B2 (en) 2011-12-16 2023-06-13 Etiometry Inc. System and methods for transitioning patient care from signal based monitoring to risk based monitoring
US10643750B2 (en) * 2013-03-15 2020-05-05 Humana Inc. System and method for determining veracity of patient diagnoses within one or more electronic health records
US20150149940A1 (en) * 2013-11-27 2015-05-28 General Electric Company Medical Test Result Presentation
US9396534B2 (en) * 2014-03-31 2016-07-19 Toshiba Medical Systems Corporation Medical image processing apparatus and medical image processing system
US10586618B2 (en) * 2014-05-07 2020-03-10 Lifetrack Medical Systems Private Ltd. Characterizing states of subject
EP3223177A1 (fr) * 2016-03-24 2017-09-27 Fujitsu Limited Système et procédé permettant d'aider au diagnostic d'un patient
US10007490B1 (en) * 2016-12-14 2018-06-26 Facebook, Inc. Automatic generation of native E-commerce applications for mobile devices
US20190156947A1 (en) * 2017-11-22 2019-05-23 Vital Images, Inc. Automated information collection and evaluation of clinical data
CN109817300B (zh) * 2019-01-18 2020-10-02 杭州逸曜信息技术有限公司 一种基于人工智能的用药规则生成方法
CN114846559A (zh) * 2019-12-19 2022-08-02 株式会社神龙 模拟系统和程序
US11832960B2 (en) 2020-06-12 2023-12-05 Bart M Downing Automated health review system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000057264A1 (fr) * 1999-03-22 2000-09-28 Weitz Sandra R Systeme de gestion en pratique medicale
US6149585A (en) * 1998-10-28 2000-11-21 Sage Health Management Solutions, Inc. Diagnostic enhancement method and apparatus
WO2002009580A1 (fr) * 2000-08-02 2002-02-07 Healthshore Inc. Systeme en ligne d'evaluation et de traitement medicaux, procede et portail
US20030158468A1 (en) * 2000-02-14 2003-08-21 Iliff Edwin C. Automated diagnostic system and method including multiple diagnostic modes
US20060135859A1 (en) * 2004-10-22 2006-06-22 Iliff Edwin C Matrix interface for medical diagnostic and treatment advice system and method
US20080177578A1 (en) * 2000-03-10 2008-07-24 Zakim David S System and method for obtaining, processing and evaluating patient information for diagnosing disease and selecting treatment

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030212579A1 (en) * 2002-05-08 2003-11-13 Brown Stephen J. Remote health management system
EP1156440A2 (fr) * 2000-04-19 2001-11-21 Siemens Aktiengesellschaft Méthode et dispositif pour déterminer de facon automatisé les risques de santé pour un patient
CA2325205A1 (fr) * 2000-11-02 2002-05-02 The Sullivan Group Module informatise de gestion des risques pour diagnostic medical

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6149585A (en) * 1998-10-28 2000-11-21 Sage Health Management Solutions, Inc. Diagnostic enhancement method and apparatus
WO2000057264A1 (fr) * 1999-03-22 2000-09-28 Weitz Sandra R Systeme de gestion en pratique medicale
US20030158468A1 (en) * 2000-02-14 2003-08-21 Iliff Edwin C. Automated diagnostic system and method including multiple diagnostic modes
US20080177578A1 (en) * 2000-03-10 2008-07-24 Zakim David S System and method for obtaining, processing and evaluating patient information for diagnosing disease and selecting treatment
WO2002009580A1 (fr) * 2000-08-02 2002-02-07 Healthshore Inc. Systeme en ligne d'evaluation et de traitement medicaux, procede et portail
US20060135859A1 (en) * 2004-10-22 2006-06-22 Iliff Edwin C Matrix interface for medical diagnostic and treatment advice system and method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
ANONYMOUS: "IBM WebSphere Applet Designer Guide passage", IBM WEBSPHERE APPLET DESIGNER GUIDE, XX, XX, 1 August 2000 (2000-08-01), pages 95 - 113, XP002268915 *
JANSSEN B ET AL: "Guidelines based on decision support software; Quality management in neurological outpatient schizophrenia treatment", DER NERVENARZT ; ORGAN DER DEUTSCHEN GESELLSCHAFT FÜR PSYCHIATRIE, PSYCHOTHERAPIE UND NERVENHEILKUNDE ORGAN DER DEUTSCHEN GESELLSCHAFT FÜR NEUROLOGIE, SPRINGER, BERLIN, DE, vol. 77, no. 5, 1 May 2006 (2006-05-01), pages 567 - 575, XP019415119, ISSN: 1433-0407 *
SHORTLIFFE E H, CIMINO J J: "Biomedical Informatics: Computer Applications in Health Care and Biomedicine", 25 May 2006, SPRINGER, USA, ISBN: 978-0-387-28986-1, XP002570368 *

Also Published As

Publication number Publication date
US20120130743A1 (en) 2012-05-24
EP2347358A1 (fr) 2011-07-27
EP2192510A1 (fr) 2010-06-02
US8548827B2 (en) 2013-10-01

Similar Documents

Publication Publication Date Title
EP2359280A1 (fr) Procédé mis en oeuvre par ordinateur pour afficher des diagnostics de maladies chroniques spécifiques d'un patient
WO2010057891A1 (fr) Procédé mis en oeuvre par ordinateur pour l'aide au diagnostic médical
US7991485B2 (en) System and method for obtaining, processing and evaluating patient information for diagnosing disease and selecting treatment
US20020147615A1 (en) Physician decision support system with rapid diagnostic code identification
DE102005012628A1 (de) Verarbeitungssystem für klinische Daten
DE102008010683A1 (de) Verfahren und Systeme zur Einbringung der klinischen Anzeige von und Suche nach medizinischen Krankenaktendaten aus einer Vielzahl von Informationssystemen
DE102004013650A1 (de) System und Verfahren zur Verarbeitung von Information betreffend Labortests und Ergebnisse
US20020147614A1 (en) Physician decision support system with improved diagnostic code capture
US20020123906A1 (en) Chronic pain patient risk stratification system
DE10316298A1 (de) Verfahren und Anordnung zur automatischen Aufbereitung und Auswertung medizinischer Daten
DE10240216A1 (de) Verfahren und Datenbank zum Auffinden von medizinischen Studien
Mellon et al. Interventions for improving medication adherence in solid organ transplant recipients
DE102021123842A1 (de) Bewertung einer verringerung eines krankheitsrisikos und behandlungsentscheidung
US20220405680A1 (en) Automated Healthcare Provider Quality Reporting System (PQRS)
DE102004013651A1 (de) Medizinisches Datensatzklassifizierungssystem
US20030046305A1 (en) Method and system for assisting a professional person to advise a client
EP3469500B1 (fr) Procédé et système d'assistance à la prescription d'une médication pour un patient
EP2236076B1 (fr) Procédé et système de détermination de la différence entres des valeurs pré- et postprandiales
EP2370910A1 (fr) Procédé d'obtention sensible au contexte d'informations relatives au patient
EP1282062A2 (fr) Procédé de médiation pour un fournisseur de services de santé à distance
Stead et al. Demand-oriented medical records: toward a physician work station
Borden Postmarketing surveillance of drugs
Lavril et al. ARTEL: An expert system in hypertension for the general practitioner
EP2056220A1 (fr) Procédé de fourniture sensible au contexte d'informations relatives au patient
Gerland Can Health Locus of Control and patient-physician relationship predict a specific cluster of non-adherence?: an observational study by patients' self-report in general medical practices $ h

Legal Events

Date Code Title Description
DPE2 Request for preliminary examination filed before expiration of 19th month from priority date (pct application filed from 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09752195

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2009752195

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 13128185

Country of ref document: US