Interactive probability based expert system for diagnosis of pacemaker related cardiac problems

• US 5,660,183 A
• Filed: 08/16/1995
• Issued: 08/26/1997
• Est. Priority Date: 08/16/1995
• Status: Expired due to Fees

• Current Assignee: Pacesetter Incorporated (Abbott Laboratories Incorporated)
• Original Assignee: Telectronics Pacing Systems, Inc (Abbott Laboratories Incorporated)
• Inventors: Chiang, Chih-Ming J., Parsonnet, Victor, Nappholz, Tibor, Bernstein, Alan D.
• Primary Examiner(s): Kamm, William E.
• Assistant Examiner(s): EVANISKO, GEORGE ROBERT
• Application Number:
  US08/515,996
• Time in Patent Office: 741 Days
• Field of Search: 607/27, 607/30, 128/695, 128/702, 364/413.06, 364/413.02
• US Class Current: 600/508
• CPC Class Codes: A61N 1/3702 Physiological parameters A61N1/365 takes precedence; evoked response A61N1/371
• CPC Class Codes: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A diagnostic device for providing a prognosis of patient with an implanted cardiac device includes a data bank for storing statistical data relating various symptoms characterizing cardiac problems and associated complications. The device further includes an input device so that a clinician can enter patient specific data including his symptoms. A set of rules stored in a memory are used to provide a diagnosis for the patient based on the statistical data. Preferably the diagnostic device is incorporated in a programmer for a pacemaker or other implantable cardiac device.

Claims

1. A cardiac diagnostic apparatus for diagnosing a patient comprising:

memory means for storing a set of rules, each rule defining a predetermined relation for each of a plurality of cardiac symptoms of patients with a corresponding etiology and a statistical probability relating each of said cardiac symptom to one of said corresponding etiologies;

means for receiving data descriptive of a patient'"'"'s specific cardiac symptoms;

means for identifying rules from said set which are related to said specific cardiac symptoms;

means for automatically generating a prognosis for said specific cardiac symptoms based on said statistical probabilities, said prognosis identifying at least one etiology having a preselected statistical probability; and

means for displaying said prognosis.

2. The apparatus of claim 1 wherein said memory means includes means for storing a plurality of suggested therapies, means for selecting one of said suggested therapies in association with said prognosis and means for displaying said one suggested therapy.

3. A programmer for programming a cardiac implant device, said programmer comprising:

first memory means for storing information related to programming operational parameters of a cardiac implant device;

second memory means for storing a set of diagnostic rules relating a predetermined plurality of cardiac symptoms of cardiac patients with corresponding etiologies, each rule defining a predetermined relation for each of a plurality of cardiac symptoms of patients with a corresponding etiology and a statistical probability relating each of said cardiac symptom to one of said corresponding etiologies;

data input means for inputting patient specific data into said programmer, including at least one cardiac symptom;

data communication means for exchanging data with a cardiac implant device, said data including programming data from and to said first memory means for programming said cardiac implant device;

processing means for processing the patient specific data and for automatically generating a prognosis based on at least said one cardiac symptoms and said rules to generate a prognosis, said prognosis identifying at least one etiology based on said at least one symptom and the related statistical probability; and

output means for generating an output indicative of said prognosis.

4. The programmer of claim 3 wherein said data processing means identifies a plurality of applicable rules, each rule including said one symptom, and analyzes the probability of each applicable rule to generate said prognosis.

5. The programmer of claim 3 wherein said output means includes display means for displaying said prognosis.

6. The programmer of claim 3 further comprising third memory means for storing a statistical data base and an inference machine for deriving said rules from said data base.

7. The programmer of claim 6 wherein said processing means includes means for updating said data base by adding to said data base information received from new patients.

8. The programmer of claim 3 wherein said processing means includes comparing means for comparing the probability associated with a rule to a preset level, wherein said prognosis is generated if said probability exceeds said preset level.

9. The programmer of claim 8 wherein said preset level is programmable by said programmer.

10. A method of generating a prognosis for a patient having cardiac symptoms, comprising the steps:

providing a set of rules relating each cardiac symptom with a corresponding cardiac etiology, each rule being characterized by a corresponding probability derived from predetermined statistical information;

receiving at least one symptom specific to a particular patient;

identifying particular rules of said set related with said one symptom;

analyzing the probabilities associated with said particular rules; and

generating a prognosis based on said analysis to define an etiology causing said one symptom.

11. The method of claim 10 wherein said receiving step includes receiving several symptoms, and wherein during said analysis each rule associated with each symptom is analyzed.