WO2022200077A1

WO2022200077A1 - System comprising a remote computer facility and a programmer for a medical device and operation method of such system

Info

Publication number: WO2022200077A1
Application number: PCT/EP2022/056309
Authority: WO
Inventors: Jens Mueller; Hendrik DEUTSCHMANN; Thomas Fuss; Miro SELENT
Original assignee: Biotronik Se & Co. Kg
Priority date: 2021-03-23
Filing date: 2022-03-11
Publication date: 2022-09-29

Abstract

The invention is generally directed to a system comprising a remote computer facility (10) and a programmer (3) for a medical device (1). In order to provide holistic information on the health status of a patient, the programmer is configured to bidirectionally communicate with the remote computer facility, wherein the remote computer facility comprises a data service (11), wherein the data service is configured to collect, store and process data of a plurality of patients, wherein data processing is provided using an AI module, wherein the data service is configured to provide information on the health status of a particular patient of the plurality of patients based on the processing of the patient's data, wherein this information is accessible by the programmer on request.

Description

System comprising a remote computer facility and a programmer for a medical device and operation method of such system

The invention is directed to a system comprising a remote computer facility and a programmer for a medical device. Further, the invention is directed to an operation method of such system as well as a computer program product and computer readable data carrier.

A programmer is a programming device which is used to display data provided by a medical device with which a communication connection was previously established. Further a programmer often is enabled to program such medical device, for example in order to activate program updates, or to adapt parameters of the internal algorithms of the medical device to the respective patient's needs.

A medical device is an active or passive medical device, for example, an implant such as a pacemaker (with leads), a BioMonitor, an Implantable Leadless Pacer (ILP), an Implantable Leadless Pressure Sensor (ILPS), an Implantable Cardiac Defibrillator (ICD) or a Shockbox, a device that delivers spinal cord stimulation (SCS), deep brain stimulation (DBS) or neurostimulation. A medical device may deliver one or more therapeutic substances, e.g. a drug pump, contain sensors that collect physiological signals in order to monitor the health status of the patient and/or to deliver a therapy to the patient.

The programmer usually displays the interrogated data of the medical device of a particular patient received by the programmer via a communication link. However, this information only shows a small part of the patient's history and is therefore of limited use for holistic diagnostics and therapy decisions. Accordingly, there is the need to include information from other systems (e.g. electronic health records (EHR), patient wearables, other sensor data, X- Rays, medication plans, medical history protocols) in order to better individualize therapy with regard to the particular patient by adapting the configuration of the algorithm of the medical device to the patient's specific needs by the programmer.

Currently, programmers may receive such additional information necessary for individualized diagnostics/therapy from a remote computer network which is then processed and displayed locally. However, such decentralized processing has significant disadvantages.

The programmer requires significant hardware and software resources thereby causing high costs. - Medical algorithms for programmers worth protecting need to be distributed around the world.

BugFixes must be rolled out decentrally around the world.

It is complicated and time-consuming to adapt calculation and graphing individually to the needs of the programmer's user.

Accordingly, it is desirable to provide a system that overcomes the above disadvantages. Therefore, it is an object of the present invention to provide a system comprising a remote computer facility and a programmer which is improved with regard to data processing and handling. Accordingly, it is an object of the present invention to propose a respective operation method.

The above problem is solved by a system with the features of claim 1 and by a method with the features of claim 8. In particular, the above problem is solved by a system comprising a remote computer facility and a programmer for a medical device, wherein the programmer is configured to bidirectionally communicate with the remote computer facility at least temporarily, wherein the remote computer facility comprises a data service, wherein the data service is configured to collect, store and process data of a plurality of patients, wherein data processing is provided using an AI (artificial intelligence) module, wherein the data service is configured to provide information on health status of a particular patient of the plurality of patients based on the processing of the patient's data, wherein this information is accessible by the programmer on request.

The above inventive system is based on the idea that all data acquisition, processing including filtering and post-processing for the programmer takes place in the data service at the remote computer facility using an AI module. According to the invention, a data service of the remote computer facility having an artificial intelligence module collects interdisciplinary information, processes it according to the patient's diagnostics and therapeutic measures specifically for external programmers and forwards it to the programmer for further processing (inch display).

The inventive system has major advantages over the state of the art. Below are some examples:

The programmer may be constructed smaller and technically designed "dumber", and thus less expensive.

The programmer may be limited to the relay function (medical device <-> backend) and the display of data for the health care professional (HCP, e.g. clinician).

Medical algorithms of the data service run centrally in a secured environment and are thus safer from cyber security and intellectual property theft. - Bugs may be fixed centrally and do not have to be distributed in the world.

Calculation and graphing may be done individually according to the needs of the programmer's user.

Depending on the diagnostics and therapeutic measures of the patient, the remote computer facility may generate suitable information (histograms, trends, etc.) that are provided to the respective programmer for further processing (display holistic information in Patientmanager/Datamanager). This allows display of a complex depth of health status information of the particular patient without the need for the programmer to have significant hardware and software resources.

The health status information of any patient of the plurality of patients, processed by the data service, may be accessed by any programmer (given a configurable user management grants access), either by data download or remotely embedding views. The information may displayed in a Data Manager/Patient manager located on the programmer.

The remote computer facility and the data service may comprise any hardware or virtual (cloud) computing facility which may include a single computer and a computer network.

The data service is a functionality of the remote computer facility. The data service is configured to collect, store and process data of a plurality of patients. For that, the data service receives any kind of health data or personal data of the plurality of patients. These data comprise not only interdisciplinary information but also other data and information from other external devices which may be connected with the remote computer facility. The data service processes the data (including filtering, pre-processing and post-processing) using at least one processor according to the patient's diagnostics and therapeutic measures wherein data processing is provided using an AI module, wherein the data service is configured to provide information on health status of a particular patient of the plurality of patients based on the processing of the patient's data. In order to provide the above functionality, the common interface to exchange (upload/download) relevant data between the programmer and the remote computer facility is changed in a way that additional data may be transmitted, as well. For collection of data the data service comprises at least one respective interface and/or communication module. For storage of data the data service comprises a data memory. The data of the patients are processed with the aim to provide suitable holistic information for the programmer. Data processing and providing information on health status comprises a data evaluator with abilities to run integrity/sanity checks and create alarms/events, not only based on implant follow-up data but on all patient data from different sources.

The data memory may include any volatile, non-volatile, magnetic, or electrical media, such as a random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), electrically-erasable programmable ROM (EEPROM), flash memory, or any other memory device. The at least one processor of the data service within the remote computer facility may be operated centrally (on-premises) or de-centrally on multiple hardware and software instances (hybrid or cloud-based). The at least one processor of the remote computer facility performs substantial computations, including numerous arithmetic operations and logic operations without human intervention, such as, for example, a personal mobile device (PMD), a desktop computer, a server computer, clusters/warehouse scale computer or embedded system. In order to generate a holistic picture for diagnostics and therapy decisions, a wide variety of health information may be calculated (histograms, trends, alerts, comments, etc.). Some health information are easy to calculate, while others are highly complex and require the AI module. Therefore, at least one processing module of the data processor may have an AI module. The communication module receives various medical and non-medical information from the patient on an event-based or cyclical basis (pull and push mechanisms are possible) and stores this information in the memory module. The goal of the at least one processor is to process the stored information in such a way that the programmer can execute further post-processing and display it without requiring significant resources and time. The processing of data in the data processor may be event-based (e.g. if the programmer requests data from the remote computer facility) or cyclic. The results may be stored permanently in the data memory or sent directly to the programmer without storing in the data memory. The communication of the communication module with the external devices and the programmer may be provided wirelessly and/or the air using electromagnetic waves, for example Bluetooth, WLAN, ZigBee, NFC, Wibree or WiMAX in the radio frequency region, or IrDA or free-space optical communication (FSO) in the infrared or optical frequency region or by wire (electrical and/or optical communication).

The medical device is any device as indicated above which is configured to be implanted partly or fully within a patient’s body or fixedly mounted at the patient’s body and configured to monitor the health condition of a patient and/or to deliver a therapy to the patient. Particularly, the medical device may be, for example, any implant, a pacemaker, ILP, ICD, Shockbox, device that delivers SCD, a fitness tracker or a BioMonitor. The programmer is a device as described above. With regard to the above invention the programmer comprises a data presenter with a respective UI (user interface) and/or UX (user experience) that is configured to provide the relevant information on the health status of the particular patient. The programmer provides the holistic information on the health status of a particular patient for optimized diagnosis and therapy in a central view which is called Data Manager or Patient Manager. The patient may be chosen using the UI/UX of the programmer. In a first embodiment, the programmer may download the information directly from the remote computer facility and post-process and display it on the device itself (e.g. in the data manager or patient manager). In another embodiment the programmer may also provide a remote view to the remote computer facility similar to a web browser. This means that no post-processing takes place on the programmer itself. In both cases a lot of data has to be transferred to the programmer. As soon as the HCP opens the Patient Manager / Data Manager on the programmer, the data are requested at the data service of the remote computer facility, transferred and displayed. To enable this, the remote network service and the programmer may be connected with a very low latency network technology (e.g. 5G or broadband LAN/WAN). The data request is, for example, a communication signal sent by the programmer to the remote computer facility with data service. This communication signal triggers the transferring of the requested health status information of the particular patient from the data service to the programmer. The data may be protected against tampering in addition to the encryption technologies of the communication networks at the business layer level.

The inventive system optimizes the workflow in the clinic because the programmer provides holistic patient information for optimized diagnosis and therapy, HCPs do not have to collect patient information from various sources manually to get holistic information, HCPs are able to see more than ever data on a programmer which is known best to them. Further, the inventive system mitigates cost pressure in healthcare segment.

In one embodiment, the data service is configured to collect patient's data from the programmer and various external sources such as patient sensors, medical devices, EHR, stored medication plans, wherein the remote computer facility is configured to be connected with these external sources. In one embodiment, the data service is configured to send the information on health status of a particular patient to the programmer on request for display at the programmer as indicated above. For that, the programmer connects to the remote computer facility for data communication.

In one embodiment, the programmer is configured to bidirectionally communicate with a medical device, e.g. an implant, for receiving patient's data and/or device's data from the medical device and for programming the medical device. The bidirectional communication may use the above mentioned communication methods.

As indicated above, in one embodiment, the data service is configured to collect data of the plurality of patients event-based and/or in pre-defmed time intervals. The event-based data collection may be provided if the patient is in the clinic for a follow-up visit or if the patient suffers from an acute attack/seizure of a chronic disease. Alternatively or additionally, the data collection may be provided, for example, weekly or monthly.

In one embodiment, the AI (artificial intelligence) module of the data service is realized with a neural network, for example trained by a deep learning method. The module within the data processor may be realized with a neural network. The neural network is based on a collection of connected nodes. A node that receives a signal processes it and connects other signal nodes to it. Typically, nodes are aggregated into layers. Different layers may perform different transformations on their inputs. Signals travel from an input layer to an output layer. Such a neural network may be trained by processing examples, each of which contains a known input and result, forming probability-weighted associations between the two, which are stored within the data structure of the neural network. Thus, the neural network learns to perform tasks by considering examples / training data. To train the neural network, the deep learning method is used. In one embodiment, the neural network is a feed forward network with several hidden layers or a recurrent neural network with several hidden layers. The neural network may include input signal conditioning and post processing of the results itself. Optionally, the neural network has a model governance layer to make the medical application traceable. The type of data to train the neural network essentially depend on the medical domain (Cardiac Rhythm Management, Neuro Spinal Cord Stimulation, etc.) and the desired medical support function. An example of an AI module is the correct detection of atrial fibrillation (AF) episodes using neural networks via Deep Learning. Anonymized intracardiac or surface ECGs are used as training data. By means of certain characteristics (e.g. R-R interval, etc.) statements about AF are tried to be made. With the above architecture, however, the programmer is able to post-process and display even such complex statements (AF episode was initially calculated by the remote computer facility) in a decentralized manner. Optimally, the remote computer facility, namely its data service, calculates exactly the information, which is really necessary for a diagnosis and provides its as information on health status. For this purpose, it is possible that the data service is configured in such a way that it knows the patient's diagnostics. This may be realized by manual input of the HCP in the remote computer facility or by an external device which is connected to the remote computer facility, or by independent automatic recognition within the processor by integrating external data sources (EFfR, medical history protocols, etc.) This has the advantage that the HCP sees exactly the information on the programmer that is necessary for the HCP. This is especially important if the programmer offers a remote view on the remote computer facility and does not provide any post-processing on the programmer itself. In one embodiment, the data service is configured to collect data of the plurality of patients from different medical domains such as cardiac rhythm management, neurology, ophthalmology, surgery, gynecology, otorhinolaryngology, dermatology, venerology, internal medicine, pediatrics, laboratory medicine, nuclear medicine, psychiatry, psychotherapy, radiology, urology but also from fields such as nutritional science. The data service also stores these data and processes them.

Additionally, the above problem is particularly solved by an operation method comprising a remote computer facility and a programmer for a medical device, wherein the programmer bidirectionally communicates with the remote computer facility at least temporarily, wherein the remote computer facility comprises a data service, wherein the data service collects, stores and processes data of a plurality of patients, wherein the data service provides information on health status of a particular patient of the plurality of patients based on the processing of the patient's data, wherein these information are accessible by the programmer on request.

In one embodiment, the data service collects patient's data from the programmer and various external sources such as patient sensors, medical devices, EHR, stored medication plans, wherein the remote computer facility is connected with the respective one of these external sources for data connection.

In one embodiment, wherein the data service sends the information on health status of a particular patient to the programmer on request for display at the programmer.

In one embodiment, the programmer bidirectionally communicates with a medical device, e.g. an implant, for receiving patient's data and/or device's data from the medical device and for programming the medical device.

In one embodiment, the data service collects data of the plurality of patients event-based or in pre-defmed time intervals.

In one embodiment, the AI module is realized with a neural network, for example trained by a deep learning method.

In one embodiment, the data service collects data of the plurality of patients from different medical domains.

The above embodiments of the operation method have the same advantages as the above system. Embodiments of the system indicated above may be realized in the operation method analogously. It is referred to the above explanation of the system in this regard.

The above method may, further, be realized as a computer program which comprises instructions which, when executed, cause the system to perform the steps of the above method which is a combination of above and below specified computer instructions and data definitions that enable computer hardware to perform computational or control functions or which is a syntactic unit that conforms to the rules of a particular programming language and that is composed of declarations and statements or instructions needed for a above and below specified function, task, or problem solution. Furthermore, the above defined method may be part of a computer program product comprising instructions which, when executed by a processor, cause the processor to perform the steps of the above defined method. Accordingly, a computer readable data carrier storing such computer program product is disclosed. The present invention will now be described in further detail with reference to the accompanying schematic drawing, wherein Fig. 1 shows an embodiment of the inventive system.

The HCP interrogates a mildly demented male patient with a two-chamber pacemaker 1 as a medical device using a programmer 3 in a follow-up center. The patient has received the pacemaker 1 as an implant because he has 2nd degree AV block. The right ventricle must be appropriately paced by the pacemaker 1. The patient has a pacemaker 1 with home monitoring function, he keeps a diary provided in form of a Patient App 5 and has a smartwatch 7 that regularly sends activity data to the remote computer facility 10. The Patient App diary 5 also connects to the remote computer facility 10 in regular time intervals. The patient could keep or have further/other external sources beside/than the Patient App 5 and/or the smartwatch 7. The HCP focuses on proper pacing of the right ventricle and AV timing (e.g. AV-delay) settings during follow-up. The patient is unable to recall any discomfort in the past few months due to mild dementia. To gain a holistic picture of the patient, the HCP is searching the patient data shown at the UI of the programmer 3 using the patient/data manager, wherein the programmer 3 is connected to the remote computer facility 10. After choosing the right patient, holistic information (inch history) for the patient is requested by the HCP using the programmer 3. Due to this action the remote computer facility 10 is triggered by the programmer 3 to send all relevant health status information on the selected patient. The remote computer facility 10, in particular its data service 11, has determined by means of the intracardiac ECGs, which were transmitted cyclically via Home Monitoring from the pacemaker 1, that with a probability of 87% a deterioration of the AV block from the 2nd degree to the 3rd degree will occur . There are also comments from the patient in the Patient App diary 5 describing discomfort, which are shown to the HCP, as well because diary data were transmitted to the remote computer facility 10 each time the patient provides a new entry in his diary. Further, the HCP may derive from the patient's health status information shown at the programmer 3 that patient's activity decreased massively at times although it is summer outside and the weather was good - this is derived by the HCP from the smartwatch 7 data, which received the remote computer facility 10 daily. Additionally, an AF (atrial fibrillation) episode was reported 2 times by the pacemaker 1 to the remote computer facility 10. However, the AI module of the data service 11 for AF classification found that the probability of an AF episode was only 0.1% and thus not present.

All the above mentioned necessary health status information inch history were displayed on the programmer 3 which has received them from the remote computer facility 10, in particular its data service 11. The health status information were prepared such by the data service that the graphical representation of the information explicitly draws attention to the previously mentioned findings. The programmer 3 displays this data in its patient/data manager. Based on this information, the HCP refers the patient to an electrophysical laboratory and reconfigures the AV timing parameters of the pacemaker 1 based on the afore mentioned data analysis provided by the remote computer facility 10 and its data service 11. The inventive system may provide further interfaces between relevant external sources (patient sensor data, etc.) and known Home Monitoring Systems. Further, the system may lead to workflow optimization in the clinic.

Claims

1. A system comprising a remote computer facility (10) and a programmer (3) for a medical device (1), wherein the programmer is configured to bidirectionally communicate with the remote computer facility, wherein the remote computer facility comprises a data service (11), wherein the data service is configured to collect, store and process data of a plurality of patients, wherein data processing is provided using an AI module, wherein the data service is configured to provide information on health status of a particular patient of the plurality of patients based on the processing of the patient's data, wherein this information is accessible by the programmer on request.

2. The system of claim 1, wherein the data service (11) is configured to collect patient's data from the programmer and various external sources (5, 7) such as patient sensors, medical devices, EHR, stored medication plans, wherein the remote computer facility is configured to be connected with these external sources.

3. The system of any of the previous claims, wherein the data service (11) is configured to send the information on health status of a particular patient to the programmer on request for display at the programmer.

4. The system of any of the previous claims, wherein the programmer is configured to bidirectionally communicate with a medical device, e.g. an implant, for receiving patient's data and/or device's data from the medical device and for programming the medical device.

5. The system of any of the previous claims, wherein the data service (11) is configured to collect data of the plurality of patients event-based or in pre-defmed time intervals.

6. The system of any of the previous claims, wherein the AI module is realized with a neural network, for example trained by a deep learning method.

7. The system of any of the previous claims, wherein the data service (11) is configured to collect data of the plurality of patients from different medical domains.

8. An operation method of a system comprising a remote computer facility (10) and a programmer (3) for a medical device (1), wherein the programmer bidirectionally communicates with the remote computer facility, wherein the remote computer facility comprises a data service (11), wherein the data service collects, stores and processes data of a plurality of patients, wherein the data service provides information on health status of a particular patient of the plurality of patients based on the processing of the patient's data, wherein these information are accessible by the programmer on request.

9. The method of claim 8, wherein the data service (11) collects patient's data, for example event-based or in pre-defmed time intervals, from the programmer and various external sources (5, 7) such as patient sensors, medical devices, EHR, stored medication plans, wherein the remote computer facility is connected with the respective one of these external sources for data connection.

10. The method of any of the claims 8 to 9, wherein the data service (11) sends the information on health status of a particular patient to the programmer on request for display at the programmer.

11. The method of any of the claims 8 to 10, wherein the programmer bidirectionally communicates with a medical device, e.g. an implant, for receiving patient's data and/or device's data from the medical device and for programming the medical device.

12. The method of any of the claims 8 to 11, wherein the AI module is realized with a neural network, for example trained by a deep learning method.

13. The method of any of the claims 8 to 12, wherein the data service (11) is configured to collect data of the plurality of patients from different medical domains.

14. A computer program product comprising instructions which, when executed by a processor, cause the processor to perform the steps of the method according to any of the claims 8 to 13.

15. Computer readable data carrier storing a computer program product according to claim 14.