Method for Cancerous Disease Diagnosis in the Latent Period

1. Mark D.Kats, born in 1937, higher education, applied mathematician, doctor of science, academician of Ukrainian Production Engineering Academy and Ukrainian Academy of Ecological Sciences, corresponding member of International Academy of Computer Sciences and Systems. Donetskaya St. 37/24, Severodonetsk, Lugansk region, 349940 Ukraine, tel.: (380-06452) -3-08-75, E-mail: (Kats M.D.).

2. Name of the project is "DEVELOPMENT OF METHODS FOR CANCEROUS DISEASE DIAGNOSIS IN THE LATENT PERIOD".

3. Application field of the offer.

Medicine. On developing the project, the problem of correct diagnosis of cancerous disease in the latent period (3-6 and more moths before the first clinical symptoms appear) will be solved.

4. Description of the offer.

4.1 Principal possibility of solution of the formulated problem is based on the following axioms:

- cancerous disease already exists and progresses in the latent period. Dynamics of its progress, therefore, must be reflected by corresponding changes in various systems of the human body;

- cancerous disease has different etiology and localization but rather close, similar changes in various systems of the body take place in the latent period;

- yet unknown differential syndromes which can be used for reliable diagnosis of cancerous disease in the latent period regardless of its etiology and future localization are really existing.

4.2 Methodology of search for cancerous disease syndromes in the latent period.

4.2.1 As the modern level of knowledge of the processes taking place in the latent period of cancerous disease is insufficient to make a list of symptoms whose values change regularly during this period, the list may be made on the basis of intuitive ideas of leading specialists in various fields of medical science (oncologists, geneticists, immunologists, endocrinologists, cytologists, molecular biologists, microbiochemists, pathophysiologists-oncologists, experts in field structures, etc.).

4.2.2 In a certain time period (1-3 months, to be specified later) a great number of patients known not to suffer from cancerous disease at the time of the first examination is subject to examination on the basis of symptoms selected as per paragraph 4.2.1. Characteristics of this number of patients (quantity, sexual, age, geographic and other criteria) will be defined more precisely. (For example, quantity will be defined on the basis of cancerous disease statistics available provided that clinical symptoms of cancerous disease will be certainly found out during the examination for 200 to 250 representatives of the patients being studied).

One of criteria for fitness of symptoms being controlled for the solution of the formulated problem can be the presence of regularity in the change of their values from the start of observance till the latent period of cancerous disease terminates.

4.2.3 Experimental data obtained after carrying-out paragraph 4.2.2 is tabulated. Each line of the table contains values of all symptoms recorded for one patient examined during t-th examination and the result of examination (cancerous disease is available or not) in the preset time period a (during t+a-th examination).

4.2.4 Nature of relationship between each of the symptoms selected as per paragraph 4.2.1 and diagnosis is determined and the force of these relations is evaluated on the basis of data as per paragraph 4.2.3 using method of one-dimensional relations recovery. Symptoms which are inessential for the solution of this problem are rejected.

4.2.5 On the basis of the table of experimental data (paragraph 4.2.3) cleared as per data of paragraph 4.2.4, a system mathematical model of relationship between diagnosis (cancer is available or not) and values of symptoms in the preset time period a (for instance, 3 months) in comparison with the first symptoms of a cancerous disease which can be reliably identified by up-to-date medical methods is constructed.

(Mosaic portrait method based on artificial intelligence philosophy is nowadays a sole method of mathematical simulation enabling to obtain an informative mathematical model of a complex object without participation of an expert in a corresponding field only on the basis of initial experimental data collected during observance. Moreover, a posteriori information obtained using the mosaic model contains a large number of new, unknown before syndromes of diseases to be differentiated.)

Mosaic model can also be constructed using a diagnosis verified during t-th examination and values of symptoms recorded during (t-2)th, (t-3)th, etc. examinations which allows to clear up the picture of dynamics of cancerous disease in the latent period and later on to fix the start of the disease at more early time.

The problem of cancerous disease diagnosis model construction can be stated

as the following mathematical formulation.

GIVEN:

Table of experimental data M=XxY (X={Xij}, i=1,m, j=1,n; Y={Yi}, l=1,k) table, each line of which contains information about symptoms values (Xij) recorded during observance period t and verified diagnosis Yi for the i-th patient made out in the preset time period a. (Here m is a number of lines (patients) in table M, n is the number of columns (symptoms) in table M.)

IT IS NECESSARY:

to construct a mathematical model allowing to reliably differentiate a cancerous disease within time a before the first symptoms of the cancerous disease appear on the basis of the table M by means of formalized procedures.

Model for cancerous disease diagnosis in the preclinical (latent) period can be constructed using the mosaic portrait method by realization of the following formalized procedures:

- transformation of table of initial experimental data M to table M' using formalized procedure of split of range of values of each of symptoms into subranges (transfer from continual scales used for measurement of symptoms values to discrete scales of names). In this case a specific code is applied to each subrange of values of each of symptoms;

- search for codes combinations which can be found in table M' only in lines belonging to patients with cancerous disease and not found in any line where this disease is not available.

These combinations are interpreted as differential syndromes of cancerous disease.

Realization of program for cancerous disease diagnosis in the latent period will require substantial expenses (in the first place for examination of large number of patients using many various methods allowing to control various systems of human body) and most probably cooperation with large oncologic centers.

After corresponding clinical verification and correction of the model software package "CANCEROUS DISEASE DIAGNOSIS IN THE LATENT PERIOD'' will be developed. After patenting it will be a final software product suitable for marketing in all countries.

5. Advantages in comparison with analogues.

At present, the methods for cancerous disease in the preclinical (latent) period are not known.

6. Potential users.

Potential users of the package can be all medical institutions involved in diagnosis and treatment of oncological patients.

After corresponding clinical verification and correction of the model a software package "CANCEROUS DISEASE DIAGNOSIS IN THE LATENT PERIOD'' will be developed. After patenting it will be a final software product suitable for marketing in all countries.

7. Finality degree.

AS A RESULT OF STRATEGIC INVESTIGATION WHICH AIM WAS TO SEARCH FOR NEW APPROACHES ENABLING TO SOLVE THE PROBLEM OF CORRECT IDENTIFICATION OF OBJECTS, ON THE BASIS OF INFORMATIVE CHARACTERISTICS BELONGING TO "LARGE" SYSTEMS CLASS, USING EXPERIMENTAL DATA OBTAINED IN THE COURSE OF OBSERVATION (Use of well-known methods of mathematical simulation for solution of the problem leads to serious and mostly insuperable methodical and computing difficulties.) THE MOSAIC PORTRAIT METHOD BASED ON ARTIFICIAL INTELLIGENCE PHILOSOPHY WAS DEVELOPED, EVALUATED AND GAVE GOOD EXPERIMENTAL RESULTS FOR MANY YEARS WHEN SOLVING MANY RESEARCH AND APPLIED PROBLEMS.

THE MOSAIC PORTRAIT METHOD ALLOWS TO CONSTRUCT INFORMATIVE MATHEMATICAL MODELS OF AN OBJECT OF ANY PHYSICAL NATURE CARRYING ESSENTIAL SCOPE OF NEW, UNTRIVIAL, UNKNOWN BEFORE INFORMATION ABOUT RELATIONS BETWEEN ITS INPUT AND OUTPUT VARIABLES USING TABLE OF EXPERIMENTAL DATA WITHOUT PARTICIPATION OF AN EXPERT.

There is a software for mosaic portrait method realization which can be used to construct a model for cancerous disease diagnosis in the latent period if appropriate experimental data is available.

There is a structure of the package into which a model of any differential diagnosis can be inserted.

There are already operating diagnosis software packages for disease progress prognosis ("PROGNOSIS OF MYOCARDIAL INFARCTION COMPLICATIONS AT THE ACUTE PHASE "), for differential diagnosis ("GASTRIC ULCER - GASTRIC CANCER", "IDIOPATHIC HYPERTENSION - NEPHRITIC HYPERTENSION", etc.) for which mathematical models have been developed in collaboration with Military Medical Academy (Saint-Petersburg).

8. Protection of the offer.

As on today the problem of mathematical model construction for reliable diagnosis of cancerous disease in the latent period can be solved only using the mosaic portrait method.

The mosaic portrait method is not protected but the main part of algorithm ensuring polynomiality of computer time and computer time versus dimensionality of the problem, correspondingly, has not been published, and only the author is possessed of the software for its realization.

9. Mark D. Kats has the property right to this offer.

Information supplied by the Author December 1998.  Page last updated: January 19, 2004