Data processing: artificial intelligence – Knowledge processing system – Knowledge representation and reasoning technique
Reexamination Certificate
1998-06-12
2001-08-28
Chaki, Kakali (Department: 2122)
Data processing: artificial intelligence
Knowledge processing system
Knowledge representation and reasoning technique
C706S046000, C706S059000
Reexamination Certificate
active
06282531
ABSTRACT:
BACKGROUND OF THE INVENTION
Technical Field
This invention relates generally to the field of computerized knowledge engineering and expert systems and more particularly to computer systems for managing applied knowledge, decision making, risk assessment, and workflow iteratively amongst several dimensions and contexts.
Background
In many areas of endeavor, one or more bodies of knowledge exist about conditions or circumstances which might affect human lives. For example, in health care, a considerable amount of clinical knowledge is available in electronic form about various diseases and the treatments for each. Similarly, in educational systems significant information exists about intelligence and academic performance indicators and ways to encourage achievement. In commercial marketing, demographics data can be gathered and stored electronically about buyers and their preferences, so that marketing can be done and targeted to prospective buyers with similar characteristics. However, a major problem with such systems is that they are usually limited in scope to one or two dimensions and do not take into account contextual information or other bodies of knowledge that might affect the outcome. They also tend to be limited in their ability to provide plans for actions, assign individuals to carry out the actions, and their ability to manage the workflow and follow-up on the actions when multiple tasks and individuals are involved.
For example, in health care systems, in early computer applications, simple databases were created to store information about diseases and differential diagnosis. An attending physician would have to suspect the existence of one or more diseases and check for those in the database. Now automated expert systems embody the experience of one or more diagnostic physicians. Such an expert system helps the attending physician diagnose an illness, often by asking a number of questions of the attending physician about the patient. Such systems help in diagnosing and treating an individual. However, they are still operating only in the dimensions of diagnosis and immediate treatment. U.S. Pat. No. 5,583,758, to McIlroy et al., describes a health care management system which selects treatment guidelines from a data base of diagnosed conditions, and enables a user to propose alternative treatments, which the system will then compare with the system recommended treatment and submit to another health care provider for review and approval.
McIlroy and similar expert systems assume there is one primary disease in question which needs diagnosis or treatment. However, as is often the case for the elderly, multiple diseases may be present in the same individual. One patient might have diabetes, a heart condition, and a respiratory system problem. The systems that allow a user to describe a patient having multiple ailments often require that the same information be entered redundantly about the patient for each disease. Even systems that might take this into account do not consider the patient's life circumstances and tendencies. An outpatient care plan for an elderly male with multiple diseases may be prone to failure if it relies on self-administration of a complex schedule of medications together with frequent trips to a physical therapy facility across town. This is so especially if the man is recently widowed, lives alone, and is poor. Such an individual may not be motivated or able to follow the medication schedule and may not have the resources or assistance needed to get to the therapist's office.
Yet the same outpatient plan might work for a married man of reasonable means whose wife is both healthy and able to drive him to and from the physical therapist's. Systems which focus only on diagnosis and treatment, even for patients with multiple diseases usually do not take these life circumstance factors into account.
Not only do most systems that make use of knowledge bases have limitations in scope, they also tend to be non-iterative or non-recursive. That is, little or no follow-up is done using the data that was used at first. In the above example, most systems do not have a way to check up or record the fact that the care plan was or was not followed and then plan the next step accordingly.
Another defect of many existing systems is that they are designed for only one or two types of participant-actors or task performers. In health care again, a system may be designed for physicians or pharmacists, but not include activities and analysis for nurses, therapists, aides, home caregivers, and administrators. Thus, the system is not able to reflect all that needs to be done or has been done for a given patient. Such a system might record that a doctor has diagnosed a condition or that the pharmacist has prescribed for it, but the system is not designed to ask a caregiver to give the medication or an administrator to arrange for transportation. At best, some systems will enable nurses or administrators to review the patient records.
Until recently, there had been little impetus to develop more comprehensive systems, since many of the computerized bodies of knowledge were used in industries where fees were charged by service. Health care in many countries used to be primarily fee for service. Health Maintenance Organizations (HMO's) and other health care systems and providers have changed the fee structure in many countries to a flat or fixed fee structure, in which resources must be used as efficiently as possible.
Other corporate structures have also changed recently to more resource-conscious methods of working, as well. In school systems, the tools which allow school administrators to identify high achievers or low achievers are usually not tied to systems which manage resources or interventions. As school budgets come under increased pressure, with fixed per pupil per year budgets, interventions must be done with optimum resource usage in mind. In many corporate structures, the focus on resources also includes a greater emphasis on delegation (where allowed) and teamwork. In commercial businesses, many middle management positions have been eliminated in favor of delegating more responsibility to the employees. Nurse practitioners now often do some of the preliminary fact gathering for physicians in HMO's. This change of focus from results-only to results and best quality use of resources means that many existing computerized or information technology (IT) expert systems or knowledge databases do not adequately address resource usage.
Another stumbling block in the development of more comprehensive systems has been the need for accountability and record keeping in many industries. Health care, again, is a good example. To avoid errors and malpractice claims, most systems need to permanently record diagnoses, prescriptions and actions taken, and insure that the electronic records cannot be changed at a later date. Frequently this is done in a patient database that is separate from most other systems or databases. Records in the patient database usually serve only one or two purposes. First, they permanently record the diagnosis made and care given. Second, they may be used in bill preparation.
Expert systems are usually not linked or related to other systems, since the expert systems were developed for one-dimensional use, such as diagnosis, or for computer application software support. As the name implies, an expert system is usually based on the expertise of one individual or type of individual, such as a diagnostic physician. Expert systems for differential diagnoses (with a few exceptions for occupational diseases) usually do not take into account any other aspects of the patient's life circumstances, since they are not likely to be relevant to a proper diagnosis. Similarly, pharmaceutical expert systems focus on matters such as drug interactions and drug toxicity, but not whether the patient is likely to have a spouse (who can help with the medication schedule) or transportation. U.S. Pat. No. 5,563,805, to Arbuckle et al., des
Harris Jeff
Haughton John F.
Merenyi Robert C.
Sherman Alexander
Chaki Kakali
CogniMed, LLC
Khatri Anil
Stretch Maureen
LandOfFree
System for managing applied knowledge and workflow in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System for managing applied knowledge and workflow in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System for managing applied knowledge and workflow in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2514229