초록

A plurality of chronic sensors are used to facilitate diagnosis and medical decision making for an individual patient. An expert system evaluates the sensor data, combines the sensor data with stored probability data and provides an output signal for notification or medical intervention.

대표청구항 ▼

1. A system comprising: a plurality of input devices each of which provides an input signal based on a patient state, the plurality of input devices including at least one implantable sensor and at least one external sensor;a knowledge base corresponding to the plurality of input devices;an inferenc

1. A system comprising: a plurality of input devices each of which provides an input signal based on a patient state, the plurality of input devices including at least one implantable sensor and at least one external sensor;a knowledge base corresponding to the plurality of input devices;an inference engine coupled to the plurality of input devices and having access to the knowledge base, wherein the inference engine includes a processor to implement a fuzzy logic algorithm to generate an inference of a patient condition based on application of the algorithm to the knowledge base and at least one input signal provided by said input devices, wherein the knowledge base is adjusted for use in later analysis by using an adaptive feedback signal derived from the inference, and wherein the adaptive feedback signal includes a weighting factor to adjust a weight ascribed to the input signal by the knowledge base; andan output device to generate an output to indicate a therapy based on the inference and to provide the output to a user or a process to provide the therapy. 2. The system of claim 1, wherein the fuzzy logic algorithm applies a plurality of probabilistic if-then rules to the knowledge base and the input signal to generate the output. 3. The system of claim 1, wherein the knowledge base includes treatment options, and wherein the processor is configured to determine a treatment regimen from the knowledge base and the inference of the patient condition. 4. The system of claim 3, wherein the knowledge base includes cost data, and wherein the processor is configured to determine the treatment regimen using the cost data and the inference of the patient condition. 5. The system of claim 3, wherein the processor is configured to select a parameter value for electrical pacing therapy as part of the treatment regimen. 6. The system of claim 1, wherein the implantable sensor includes at least one of a heart sound sensor, a filling pressure sensor, and a transthoracic impedance sensor, and wherein the processor is configured to generate an inference of heart decompensation based on application of the algorithm to the knowledge base and an input signal provided by the implantable sensor. 7. The system of claim 1, wherein the knowledge base includes pharmaceutical data, and wherein the processor is configured to determine a recommended drug therapy from the knowledge base and the inference of the patient condition. 8. The system of claim 7, wherein the processor is configured to detect the presence of a drug side effect based on application of the algorithm to the knowledge base and the input signal. 9. The system of claim 1, including a communication network, wherein the knowledge base is adjusted using adaptive feedback data communicated via the communication network. 10. The system of claim 9, wherein the input devices include an external sensor configured to provide data to the inference engine via the communication network. 11. A method comprising: receiving a plurality of input signals, at least one input signal generated by an implantable sensor and at least one input signal generated by an external device monitoring a patient state;executing a program on a processor to generate an inference of a patient condition, wherein the program implements a fuzzy logic algorithm and applies the algorithm to the plurality of input signals and a knowledge base to generate the inference;adjusting the knowledge base for use in later analysis by using an adaptive feedback signal, the adaptive feedback signal generated based on the inference of the patient condition and including a weighting factor to adjust a weight ascribed to at least one of the input signals by the knowledge base; andproducing an output to indicate a therapy based on the inference, and providing the output indicating the therapy to a user or process to provide the indicated therapy. 12. The method of claim 11, wherein executing a program includes applying a plurality of probabilistic if-then rules to the knowledge base and the input signal to generate the inference. 13. The method of claim 11, wherein producing an output includes: determining a treatment regimen according to the inference of the patient condition from among available treatment options included in the knowledge base. 14. The method of claim 13, wherein determining a treatment regimen includes determining a treatment regimen according to the inference of the patient condition and according to cost data included in the knowledge base. 15. The method of claim 13, wherein determining a treatment regimen includes selecting a parameter value for electrical pacing therapy as part of the treatment regimen. 16. The method of claim 11, wherein receiving a plurality of input signals includes receiving an input signal from at least one of an implantable heart sound sensor, an implantable filling pressure sensor, and an implantable transthoracic impedance sensor, andwherein executing a program includes generating an inference of heart decompensation based on application of the algorithm to the knowledge base and an input signal provided by the implantable sensor. 17. The method of claim 11, wherein determining a treatment regimen includes determining a recommended drug therapy according to the inference of the patient condition and according to pharmaceutical data included in the knowledge base. 18. The method of claim 17, wherein determining a treatment regimen includes: detecting the presence of a drug side effect based on application of the algorithm to the knowledge base and the input signal; andrecommending drug therapy according to the inference of the patient condition and the detected drug side effect. 19. The method of claim 11, wherein adjusting a knowledge base includes adjusting the knowledge base using adaptive feedback data communicated via a communication network. 20. The method of claim 19, wherein receiving the input signal includes providing data from an external sensor via a communication network.