Interactive statistical system and method for predicting expert decisions

• US 5,005,143 A
• Filed: 09/25/1989
• Issued: 04/02/1991
• Est. Priority Date: 06/19/1987
• Status: Expired due to Fees

• Current Assignee: University of Pennsylvania
• Original Assignee: University of Pennsylvania
• Inventors: Whittington, Richard, Altschuler, Martin D.
• Primary Examiner(s): Teska, Kevin J.
• Application Number:
  US07/411,911
• Time in Patent Office: 554 Days
• Field of Search: 364/200, 364/900, 364/300, 364/554, 364/513, 364/415, 364/413.02
• US Class Current: 702/181
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G09B 7/04 characterised by modifying the teaching programme in response to a wrong answer, e.g. repeating the question, supplying a further explanation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass
• CPC Class Codes: Y10S 706/924 Medical

Abstract

A system and method for selecting from a set of output actions for combinatory situations such as medical diagnosis defined by a pluality of input parameters. A decision making structure is established by an expert having a plurality of linked nodes. Random values of the input parameters are generated, which includes biasing the random values by a function of a preceding response. The steps are repeated until a sufficient number of responses having a predetermined statistical significance is achieved for each node. After this set up and teaching, the accuracy of the systems'"'"'s predictions may be verified against actual responses by the expert. The decision making structure may be interrogated by entering a situation to be analyzed. The path the expert would take through the decision making structure is predicted by determining the probability of the expert'"'"'s response at each mode in the structure to arrive at an output action. A plurality of structures may be established, one for each individual expert, allowing interrogation of any or all of the decision making structures.

Claims

1. A method for comparing the analyses of a plurality of experts in a particular field, comprising the steps of:

(a) establishing a set of output actions for combinatory situations, said combinatory situations defined by a plurality of input parameters having input values which define a particular case situation;

(b) for each expert,

(1) establishing a decision making structure having a plurality of linked nodes extending from a first node and ending in a terminal node with each node defining a question;

(2) entering the input values of the input parameters defining a case situation to be analyzed;

(3) determining which input values of the particular case situation are significant input values from node to node and predicting which response the expert would make for a question posed at the first node based on the significant input values and then predicting additional responses for questions posed at each subsequent node, said subsequent nodes forming a path based on the prior predicted responses through the decision making structure;

(4) repeating step (3) through the linked nodes until a predicted response is an output action at a terminal node of the decision-making structure; and

(5) determining the probability that the expert would have selected the output action predicted in step (4).

(c) comparing the determinations of which input values are significant input values from step (b) (3) and the output action arrived at in step (b) (4) of an expert'"'"'s decision making structure with that of another expert'"'"'s decision making structure to determine which input values are most significant in arriving at an output action.

2. A method for selecting from a set of output actions designated by an expert for combinatory situations defined by a plurality of input parameters having an incomplete set of input values which define a particular case situation, comprising the steps of:

(a) establishing a sequence of decision making structures beginning with an initial decision making structure, each of said decision making structures having a plurality of linked nodes extending from a first node and ending in a terminal node with each node defining a question, and each succeeding decision making structure designed to both overlay a node of a preceding decision making structure and address more input parameters than the preceding node of the preceding decision making structure;

(b) entering the incomplete set of input values of input parameters defining a particular case situation to be analyzed;

(c) determining which input values of the particular case situation are significant input values from node to node and predicting which response the expert would make for a question posed at the first node based on the significant input values and then predicting additional responses for subsequent questions posed at each subsequent node, said subsequent nodes forming a path based on the prior predicted responses through the decision making structure;

(d) repeating step (c) through the linked nodes until a predicted response is either an output action at a terminal node of one of said decision making structures or leads to a node requiring an input value that is not present;

(e) determining a probability that the expert would have selected the output action predicted in step (d) if the predicted response is an output action at a terminal node of the decision making structure;

(f) transferring processing to the next subsequent decision making structure which addresses more input parameters than the node of the previous decision making structure if the predicted response leads to a node requiring an input value not present; and

(g) repeating steps (c) through (g) through the linked nodes of the current decision making structure and through other decision making structures to which processing is transferred until reaching an output action at a terminal node of a decision making structure.