Computer-based neural network system and method for medical diagnosis and interpretation

• US 5,839,438 A
• Filed: 09/10/1996
• Issued: 11/24/1998
• Est. Priority Date: 09/10/1996
• Status: Expired due to Fees

• Current Assignee: Neuralmed Incorporated
• Original Assignee: Neuralmed Incorporated
• Inventors: Graettinger, Timothy Joseph, DuBose, Paul Alton
• Primary Examiner(s): Kamm, William E.
• Application Number:
  US08/712,986
• Time in Patent Office: 805 Days
• Field of Search: 128/630, 128/904, 128/924
• US Class Current: 600/300
• CPC Class Codes: A61B 5/00 Measuring for diagnostic purposes radiation diagnosis A61B6/00; diagnosis by ultrasonic, sonic or infrasonic waves A61B8/00; Identification of persons
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof
• CPC Class Codes: Y10S 128/904 Telephone telemetry
• CPC Class Codes: Y10S 128/924 using artificial intelligence

Abstract

A neural network system and method for diagnosing patients'"'"' medical conditions provide an efficient aid in identifying and interpreting factors which are significant in the medical diagnosis. The neural network is trained to recognize medical conditions by being provided with input data that is available for a number of patients, and diagnosis made by physicians in each case. Upon completion of the training period the neural network system uses input measurement and interview data to produce a score, or a graded classification, of a patient'"'"'s medical condition that is accompanied with a diagnosis interpretation. The interpretation is a sorted catalogue of individual factors and interactions that influenced the score. The interpretive facility of the present invention is based on comparison with a set of nominal values for each input factor or interaction. It can assist the physician in making a diagnosis of the patient'"'"'s condition and can further provide a "second opinion" that may confirm the physician'"'"'s findings or point to ambiguities that call for a more detailed analysis.

Claims

1. A method for diagnosis of a medical condition comprising the steps of:

providing a plurality of input parameters representing characteristics of the medical condition;

converting said plurality of input parameters into numerical data;

inputting the numerical data to a neural network trained to detect the medical condition and determining a score indicative of the likelihood of the medical condition on the basis of the numerical data;

interpreting the numerical data to provide estimates of the contribution of input parameters to the determined score; and

displaying the determined score and the provided estimates in a human-readable form.

2. The method of claim 1 wherein said plurality of input parameters correspond to a single patient.

3. The method of claim 1 wherein the score determined by the neural network has values ranging between 0 and 1, with scores near 0 indicating low likelihood of the medical condition, and scores near 1 indicating high likelihood of the medical condition.

4. The method of claim 1 wherein the step of interpreting comprises analyzing at least a contribution to the determined score of each of the plurality of input parameters.

5. The method of claim 4 wherein the step of interpreting further comprises analyzing pairwise contributions to the determined score for each pair corresponding to different input parameters of said plurality of input parameters.

6. The method of claim 4 wherein the step of interpreting further comprises analyzing contributions to the determined score for at least each triplet corresponding to different input parameters of said plurality of input parameters.

7. The method of claim 1 wherein in the step of interpreting, the estimates of the contribution of input parameters are computed on the basis of nominal values for each of said plurality of input parameters representing characteristics of the medical condition.

8. The method of claim 1 wherein the step of displaying is substantially coincident with the step of providing a plurality of input parameters.

9. The method of claim 1 further comprising the step of providing a printed record of the determined score and the provided estimates.

10. The method of claim 1 further comprising the step of communicating the determined score and the provided estimates to a remote location.

11. A system for diagnosis and interpretation of a medical condition comprising:

means for providing a plurality of input parameters representing characteristics of the medical condition, the plurality of input parameters being provided as numerical data;

a neural network trained to detect the medical condition for determining a score indicative of the likelihood of the medical condition on the basis of the numerical data;

means for interpreting the numerical data to provide estimates of the contribution of input parameters to the determined score; and

display means for displaying the determined score and the provided estimates in a human-readable form.

12. The system of claim 11 wherein said means for providing comprises input means for entering data and storing such data into a data storage.

13. The system of claim 11 wherein the score determined by the neural network has values ranging between 0 and 1, with scores near 0 indicating low likelihood of the medical condition, and scores near 1 indicating high likelihood of the medical condition.

14. The system of claim 11 wherein the means for interpreting comprises first data processing means for analyzing at least a contribution to the determined score of each of the plurality of input parameters.

15. The system of claim 14 wherein the means for interpreting further comprises a second data processing means for analyzing contributions to the determined score for at least each data pair corresponding to different input parameters of said plurality of input parameters.

16. The system of claim 15 wherein the means for interpreting further comprises means for computing nominal values for each of said plurality of input parameters representing characteristics of the medical condition.

17. The system of claim 16 wherein the first data processing means, the second data processing means and the means for computing nominal values process the numerical data in parallel.

18. The system of claim 11 wherein the means for displaying comprises a computer monitor.

19. The system of claim 11 further comprising a printer for providing a printed record of the determined score and the provided estimates.

20. The system of claim 11 further comprising means for communicating the determined score and the provided estimates to a remote location.

21. A computer-based system to assist the diagnosis of a medical condition, comprising:

a patient record comprising numerical data representing a plurality of input factors associated with characteristics of the medical condition;

a neural network responsive to said patient record and configured to determine a score indicative of the likelihood of the medical condition in the patient record;

a computer interpreter responsive to said patient record for estimating the contribution of said plurality of input factors to the score determined in the neural network; and

a display for displaying the determined score and the estimates provided by the interpreter in a human-readable form to assist the diagnosis of the medical condition.

22. The system of claim 21 further comprising data storage for storing one or more patient records.

23. The system of claim 21 wherein the interpreter comprises first data processing means for analyzing at least a contribution to the determined score of each of said plurality of input parameters.

24. The system of claim 23 wherein the interpreter further comprises a second data processing means for analyzing contributions to the determined score for at least each data pair corresponding to different input parameters of said plurality of input parameters.

25. The system of claim 24 wherein the interpreter further comprises means for computing nominal values for each of said plurality of input parameters on the basis of training the neural network with patient records bearing known association with the medical condition.

26. The system of claim 21 further comprising a printer for providing a printed record of the determined score and the provided estimates.

27. The system of claim 21 further comprising interface means for communicating data to a remote location.

28. The system of claim 27 wherein said interface means is a modem.

29. The system of claim 27 wherein the communication of data to said remote connection is accomplished over packet switched networks, such as the Internet.