FIELD: medicine.
SUBSTANCE: invention refers to medicine, particularly to diagnostic technique for diseases. The technique implies specifying the characteristics, which spatially comprise detectable clusters formed by points of two types; the first type encloses the patients with a confirmed diagnosis of a diagnosable disease; the second type related to the patients having no diagnosable disease; training samples for classifiers consisting of medoid points for each cluster and boundary points separating the clusters are determined. A classification procedure involves using the boundary point classifier according to a special majority variant; the medoid point classifier is used for the points classified by other than the first classifier according to the special majority variant; and for the rest points - the boundary point classifier according to the simple majority variant. After the diagnosis is confirmed, it is decided that the characteristics of the examined patient are included into the training samples, if these characteristics improve the detection accuracy.
EFFECT: higher accuracy for diagnosing the disease.
3 dwg
Title | Year | Author | Number |
---|---|---|---|
MEDICAL IMAGE BINARISATION METHOD | 2013 |
|
RU2533876C2 |
METHOD OF ESTIMATION OF PATIENT HEALTH STATE, EFFECT OF TREATMENT AND ACCUMULATED IRRADIATION DOSE BY BLOOD ANALYSIS DATA | 1998 |
|
RU2135997C1 |
METHOD OF EARLY DIAGNOSTICS OF CHRONIC DISEASES OF A PATIENT BASED ON CLUSTER ANALYSIS OF BIG DATA | 2021 |
|
RU2800315C2 |
AUTOMATED SYSTEM OF DISTRIBUTED COGNITIVE SUPPORT OF MAKING DIAGNOSTIC DECISIONS IN MEDICINE | 2015 |
|
RU2609737C1 |
METHOD FOR NEURAL NETWORK ANALYSIS OF CARDIAL CONDITION | 2011 |
|
RU2461877C1 |
DIFFERENTIAL DIAGNOSIS METHOD FOR MICROBES AND AMINO ACIDS | 2007 |
|
RU2362145C2 |
METHOD FOR GENERATING MATHEMATICAL MODELS OF A PATIENT USING ARTIFICIAL INTELLIGENCE TECHNIQUES | 2017 |
|
RU2720363C2 |
METHOD FOR DIAGNOSING DISEASES AND THEIR FORMS IN COMPUTER-ASSISTED MODE | 2001 |
|
RU2191429C1 |
DIAGNOSTIC TECHNIQUE FOR ACUTE CORONARY SYNDROME | 2020 |
|
RU2733077C1 |
METHOD OF DIAGNOSING PATHOLOGIES OF HEMOSTASIS SYSTEM BY MEANS OF NEURAL NETWORK | 2011 |
|
RU2449281C1 |
Authors
Dates
2015-02-10—Published
2013-01-23—Filed