Automated medical decision making utilizing bayesian network...

Details Automated medical decision making utilizing bayesian network... Automated medical decision making utilizing bayesian network...

2000-04-07

2004-02-03

Starks, Jr., Wilbert L. (Department: 2121)

Data processing: artificial intelligence

Neural network

C706S045000, C706S924000, C600S300000

Reexamination Certificate

active

06687685

ABSTRACT:

I. FIELD OF THE INVENTION
This invention relates generally to the field of medicine, and more particularly to methods of self-diagnosis and evaluation, particularly in connection with automated medical decision-making, including online triage.
II. RELATED ART
Spiraling medical care costs in this country have become a well-known problem, and particularly with respect to emergency care. Studies show that approximately half of emergency room patients are not in need of urgent care.
1
Expense of treatment could be significantly reduced if those patients were assisted in determining whether they are suffering from an emergent condition prior to going to the emergency room.
1
U.S. Department of Health and Human Services, “National Hospital Ambulatory Medical Care Survey: 1992 Emergency Department Summary,” DHHS Publication No. (PHS) 96-1786 (March 1997).
On the other hand, a significant number of patients unduly delay their visit to the emergency room after the onset of symptoms, wasting critical hours. For example, it is estimated that over 40% of heart attack patients wait more than four hours before seeking medical help, increasing the risk of damage to the heart. In addition, the cost and complications relating to providing medical care to such patients soars.
Additionally, many patients who are suffering from an emergent condition and who go to emergency rooms are there because they have not properly managed their chronic conditions, such as diabetes, asthma, hypertension, or heart disease. In addition, expense in treating such patients goes well beyond the emergency room. It is believed that such patients are the largest overall contributors to overall medical expense, much of which could be reduced if the disease were better monitored and controlled.
There has therefore arisen a need in the medical industry for a means for enabling people who are not medically trained to ascertain their condition and evaluate whether it requires emergency care, or what other level of care is appropriate. There has further arisen a need for new system for disease management, and more particularly to a method of facilitating self-monitoring and management of chronic health conditions.
Various prior art methodologies have been developed for achieving the foregoing objectives. One such prior art technique is known as telephone triage.
2
With telephone triage, a patient calls a number which is answered by a health care professional, who may be a doctor or a nurse. The health care professional then walks the patient through a series or pre-formulated questions, and then makes a medical judgment as to whether the patient should seek emergency care.
2
See, e.g., Wheeler, Sheila Quilter, “Telephone Triage Protocols,” (1998).
Conventional telephone triage, while reasonably effective, has certain disadvantages. First, a trained health care professional is required to answer the phone, adding to the cost of the system. When the system is too busy, the patient will experience delays in what may be a critical time. When the system is not busy, the time of the health care professional manning the line is wasted, resulting in increased expense. In addition, human judgment is necessarily interposed in the process, which may result in dissimilar and potentially inaccurate advice being given to patients. In addition, the system is imperfectly dynamic, and oftentimes questions are asked which are not the most significant or meaningful.
An improvement of traditional telephone triage is described in U.S. Pat. No. 5,764,923 to Tallman et al. The Tallman patent discloses what is known in the art as “computer-assisted medical decision making.” It discloses a medical network management system wherein patients access a team of health care professionals over the telephone. The health care professionals assess patient conditions with the assistance of an automated set of assessment algorithms. While the invention disclosed in Tallman may provide certain efficiencies, the system still requires the presence of a team of health care professionals, and ultimately relies on the individual medical judgment of the person answering the phone. In addition, Tallman discloses no specific systematic methodology for developing the algorithms upon which it relies.
Another prior art system for self-evaluation and disease management is described in U.S. Pat. No. 5,827,180 to Goodman. The Goodman patent describes a health network wherein information pertaining to a patient is shared between a health care provider, a treatment facility and a patient. The information may pertain to treatment instructions specific to each patient. The system requires a treatment plan specific to each patient.
While the system described in Goodman may assist in disease management, it suffers from some drawbacks. The system in Goodman must be established for treatment of a patient on a patient-specific basis. Algorithms are developed specific to treatment of that patient, not to the disease. A person who has not established a presence on the Goodman system could not obtain any assistance in terms of self-diagnosis, treatment or disease management. Moreover, Goodman contemplates communication to the patient with a pager, which is less convenient and more intrusive than online systems.
Online medical services are also in the prior art. For example, Dr. Schueler's Health Informatics, Inc., operates a website. At that website, a user can gain medical information, and from such information, the user may attempt to determine the nature of his or her condition. That website does not, however, offer online triage or utilize the disposition level of the present invention.
Attempts to model medical knowledge in automated format have typically taken the form of so-called “rule-based” systems. Rule-based systems are normally configured as a series (often very complex) of if x, then y type rules. While formidable rule-based systems have been constructed, they have their shortcomings. One of the primary shortcomings is the tremendous pace at which such systems grow when an additional layer of variables is added. Another is their difficulty in addressing uncertainty.
There has therefore arisen a need for a reliable, statistically sound and convenient method for automated triage and disease management, particularly in an online environment, without the direct involvement of health care professionals. There has further arisen a need for a consistent, systematic methodology for evaluating the statistic significance of responses in connection with such an automated triage system for assisting users in determining the nature of their condition and facilitating disease management.


REFERENCES:
patent: 4839822 (1989-06-01), Dormond et al.
patent: 4872122 (1989-10-01), Altschuler et al.
patent: 5005143 (1991-04-01), Altschuler et al.
patent: 5133046 (1992-07-01), Kaplan
patent: 5307263 (1994-04-01), Brown
patent: 5331550 (1994-07-01), Stafford et al.
patent: 5471382 (1995-11-01), Tallman et al.
patent: 5594638 (1997-01-01), Iliff
patent: 5660176 (1997-08-01), Iliff
patent: 5680866 (1997-10-01), Kangas et al.
patent: 5715374 (1998-02-01), Heckerman et al.
patent: 5764923 (1998-06-01), Tallman et al.
patent: 5800347 (1998-09-01), Skates et al.
patent: 5802256 (1998-09-01), Heckerman et al.
patent: 5827180 (1998-10-01), Goodman
patent: 5922079 (1999-07-01), Booth et al.
patent: 5930803 (1999-07-01), Beker et al.
patent: 5950632 (1999-09-01), Reber et al.
patent: 5951469 (1999-09-01), Yamaura
patent: 5964700 (1999-10-01), Tallman et al.
patent: 6014626 (2000-01-01), Cohen
patent: 6067466 (2000-05-01), Selker et al.
patent: 6076083 (2000-06-01), Baker
patent: 6081786 (2000-06-01), Barry et al.
patent: 6083173 (2000-07-01), Grant et al.
patent: 6090044 (2000-07-01), Bishop et al.
patent: 6154736 (2000-11-01), Chickering et al.
patent: 6188988 (2001-02-01), Barry et al.
patent: 6456622 (2002-09-01), Skaanning et al.
patent: 6556977 (2003-04-01), Lapointe et al.
patent: 6601055 (2003-07-01), Roberts
patent: 2001/0023419 (2001-09-01), Lapointe et al.
patent: 2002/0044296 (2002-

Affiliated with

Also associated with

Rating

Automated medical decision making utilizing bayesian network... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Automated medical decision making utilizing bayesian network..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automated medical decision making utilizing bayesian network... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3345529