Diagnostic apparatus

• US 20040249777A1
• Filed: 07/21/2004
• Published: 12/09/2004
• Est. Priority Date: 08/15/2001
• Status: Active Grant

• Current Assignee: Metacause Solutions, Ltd.
• Original Assignee: Metacause Solutions, Ltd.
• Inventors: Lewis, Roland W., Ransing, Rajesh S., Ransing, Meghana R.
• Granted Patent: US 7,440,928 B2
• Time in Patent Office: Days
• US Class Current: 706/45
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

Apparatus and method for determining a likely cause or the likelihood of the occurrence of a cause of one or more effects, in which training data relating to previously identified relationships between one or more causes and one or more effects is used to learn the cause and effect relationship. A number of primary and secondary reference points are chosen in the input space created by belief values representing the strength of effect. A Lagrange Interpolation polynomial (or other function representing the cause and effect relationship) and a weight value is associated with each of the said reference point. Weight values associated with primary reference points are considered as independent variables (primary weight values) and other weight values, which are associated with secondary reference points (secondary weight values), depend (preferably, but not necessarily, linearly) on one or more primary weight values. Belief value in the occurrence of likely causes of one or more given effects can be determined using this method or apparatus.

Claims

1. Apparatus for determining a likely cause of one or more effects, the apparatus comprising means for receiving and/or accessing training data relating to previously-identified relationships between one or more causes and one or more effects, means for identifying a trend in said relationships and in the form of shape functions or polynomials which define quantified mappings between said one or more causes and one or more effects, each of said shape functions or polynomials being associated with each of a plurality of respective reference points, at least some of which have a weight value assigned thereto at least one of said weight values being an independent variable (“

primary weight”

) and at least one of said weight values being dependent on one or more of said primary weights (“

secondary weight”

), and means for determining one or more likely causes of one or more given effects using said mappings.

2. Apparatus according to claim 1, comprising a neural network arrangement being capable of providing outputs in the form of quantitative cause and effect relationship values.

4. Apparatus according to claim 1, wherein the trend is defined in terms of respective shape functions which produce a hyper-surface representative of the training data in terms of quantitative cause and effect relationship values.

5. Apparatus according to claim 1, including means for producing a multi-dimensional hyper-surface representing a trend in the training data, the number of dimensions of the hyper-surface being equal to the number of input nodes of a neural network.

6. Apparatus according to claim 5, wherein the neural network includes one or more layers each comprising one or more hidden nodes between said input and output nodes.

7. Apparatus according to claim 5, wherein shape functions are defined at each input and output node of the neural network, each shape function having output values between 0 and 1.

8. Apparatus according to claim 7, wherein the value of the shape function at its respective node is 1, and 0 elsewhere.

9. Apparatus according to claim 5, wherein at least each input and output node in the neural network has a weight value assigned to it.

10. Apparatus according to claim 9, wherein the value of a weight is between −

1 and +1.

11. Apparatus according to claim 9, herein the output of the network is preferably also between −

1 and +1.

12. Apparatus according to claim 1, in which each secondary weight is expressed as a linear combination of two or more primary weights.

13. Apparatus according to claim 4, wherein the hyper-surface is created using one or more multi-dimensional (linear, quadratic, cubic or higher order) polynomial equations.

14. Apparatus according to claim 1, wherein the training data is made up of one or more training files, the or each training file comprising an input vector and its corresponding desired vector.

3. A method of determining a likely cause of one or more effects, the method comprising the steps of receiving and/or accessing quantifying data relating to previously-identified relationships between one or more causes and one or more effects, identifying a trend in said relationships in the form of shape functions or polynomials which define quantified mappings between said one or more causes and one or more effects, each said shape functions or polynomials being associated with each of a plurality of respective reference points, at least some of which have a weight value assigned thereto, at least one of said weight values being an independent variable (“

primary weight”

) and at least one of said weight values being dependent on at least one of said primary weights (“

secondary weight”

), and determining one or more likely causes of one or more given effects using said mappings.

17. Method according to claim 3, wherein the trend is defined in terms of respective shape functions which produce a hyper-surface representative of the training data in terms of quantitative cause and effect relationship values.

18. Method according to claim 17, wherein the hyper-surface is created using one or more multi-dimensional (linear, quadratic, cubic or higher order) polynomial equations.

19. Method according to claim 3, wherein the training data is made up of one or more training files, the or each training file comprising an input vector and its corresponding desired vector.

15. (canceled)

16. (canceled)