System for Formal Selection of Optimal Medical Strategy

System for Formal Selection of Optimal Medical Strategy

1. Mark D. Kats, Dr. Sc., Academician of Ukrainian Process Engineering Academy and Ukrainian Academy of Ecological Science Corresponding Member of International Academy of Computer Sciences and Systems, specialist in applied mathematics. Date of birth: January 12, 1937 Address: Donetskaya St. 37-24, Severodonetsk, Lugansk Reg., 349940, Ukraine. Tel.: (380-6452)-3-08-75, E-mail: root@ixt.lg.ua (Kats M.D.).

2.Name of the project.

"DEVELOPMENT OF THE INTELLECTUAL SYSTEM FOR FORMAL SELECTION OF OPTIMAL STRATEGY AS TO MEDICAL TREATMENT OF DISEASE X TAKING INTO ACCOUNT PATIENT'S INDIVIDUALITIES"

3. Scope of the offer.

Medicine. After development of the project the problem of optimal approach as to medical treatment of decease X for each patient with the help of formalized computer procedure will be solved.

4. Description of the offer.

In cooperation with a medical center having qualified staff specializing in treatment of a particular decease the model of the decease can be constructed basing on experimental data presented by the staff and combining patient's characteristics, data on anamnesis, examination, clinical record, laboratory and instrument analysis, etc. and the treatment being applied with the treatment result. The requirements to experimental information to be presented are given below.

Optimization of the offer will be performed as follows. For each patient the situation describing his characteristics and peculiarities of his disease will be performed; recommendations on treatment will be synthesizer basing on this situation and the model of disease with the help of situational programming with the view of the best outcome of those ones possible when creating the model.

If the treatment providing the best result cannot be applied due to serious state of the patient or for other reasons then the recommendations can be given how to reach the best of the outlets left and so on.

When the effectiveness of the formal selection procedure for optimal approach as to medical treatment or corrections has been clinically proved, the commercial software package " INTELLECTUAL SYSTEM FOR FORMAL SELECTION OF OPTIMAL STRATEGY AS TO MEDICAL TREATMENT OF DISEASE X TAKING INTO ACCOUNT PATIENT'S INDIVIDUALITIES" can be developed, patented and sold.

Decision on prescribing the effective treatment should be based on mathematical model of the disease and formalized methods allowing to synthesizer optimal medical treatment taking into account patient's individualities and clinical state (it can be synthesizer also in terms of minimization of possible negative consequences) .

FORMALIZATION OF PROCEDURE FOR SELECTING OPTIMAL STRATEGY OF THE TREATMENT IS THE RESULT OF STRATEGIC RESEARCHES AIMED AT DEVELOPMENT OF A NEW METHOD OF OPTIMIZATION ALLOWING WITH USE OF MOSAIC MODEL OF THE DISEASE X AND TAKING INTO ACCOUNT PATIENT'S INDIVIDUALITIES TO SYNTHESIZER RECOMMENDATIONS PROVIDING THE BEST RESULT POSSIBLE.

The problem while selecting optimal strategy for treatment of any particular disease can be divided into 2 tasks:

- construction of mathematical model of the disease;

- development of formalized procedure for selecting optimal strategy of the treatment using this model.

Statement of the 1 task:

Given: results of observation and treatment of patients having the disease under investigation, in form of the table with each line containing information of 1 patient:

- characteristics describing patient's individualities (color of eyes, blood group, rhesus factor, etc.);

- characteristics describing the disease: anamnesis, physical examination, laboratory and instrumental methods, clinical record, etc.);

- parameters describing the treatment applied;

- estimation of patient's state after the treatment applied.

It is needed: Construction of mathematical model of the disease combining patient's individualities, disease proceeding , treatment applied and results of this treatment .

Statement of 2 task:

Given: Mathematical model of the disease combining patient's individualities, disease proceeding , treatment applied and results of this treatment .

It is needed: Synthesizer optimal strategy of the treatment basing on situation characterizing patient's individualities and clinical proceeding of the disease with the help of formalized procedure and the model of disease.

Basing on initial table a mathematical model of dependence of treatment results from characteristics of patient’s individualities, disease proceeding , treatment applied and results of this treatment can be constructed using method of mosaic portrait by the following formalized procedures:

- transformation of table of initial experimental data M into table M' at expense of transferring from continual scales (where some symptoms are changed) to discrete scales. Every subrange of each symptom characteristic is assigned its own name (code);

- search of the code combinations which are found in the table M' in the lines belonging to the same gradation of treatment quality and are not found in any line belonging to other gradations.

Each combination of this type is interpreted as a formal, non-contradictory in terms of the given experimental data hypothesis, representing relationship between the input characteristics of the patient and treatment applied and quality of this treatment.

In the class of tasks of discrete programming the principally novel method of suboptimization has been developed (method of situational programming) allowing to synthesizer the necessary control (scope and parameters of the treatment) with the help of the mosaic model of disease and the situation, i.e. characteristics describing patient's individualities such as color of eyes, blood group, rhesus factor, etc. ; disease proceeding such as clinical and examination data, laboratory and instrumental analysis, complaints, etc. Resulting in maximal effectiveness of the treatment course.

5. Advantages compared with analogs.

The problem of treatment optimization taking into account patient's individualities is the second to diagnostics factor by its significance.

However, correct methods of construction of disease models are not available , that is why selection of optimal treatment tactics is rather the art of qualified professionals than a science.

Unfortunately, besides the concept about "treating a patient and not a disease" there are no serious methodological works developing formalized procedures for selection of optimal treatment strategy based on mathematical model of a particular disease.

The offer has no analogs.

6. Potential users.

Potential users of INTELLECTUAL SYSTEMS FOR FORMAL SELECTION OF OPTIMAL STRATEGY AS TO MEDICAL TREATMENT OF DISEASE X TAKING INTO ACCOUNT PATIENT'S INDIVIDUALITIES are medical centers.

7. State of development.

INTELLECTUAL SYSTEMS have been developed, tested and experimentally proved for a number of years with success in studying and optimizing a wide range of complex technical, biological, medical and social systems by methods of mosaic portrait and situational programming.

The computer program is available for mosaic portrait method with the help of which and use of corresponding experimental data a disease model can be constructed.

The computer program is available for situational programming methods with the help of which and use of mosaic model of the disease the recommendations on selection of optimal treatment strategy for a particular patient can be synthesizer.

It is necessary to develop a universal structure of the package (similar to our package of differential diagnostics) into which the model of any disease and the program realizing the method of situational programming can be used as input data.

8. Offer infringement.

Method of mosaic portrait and situational programming are not patented ; main part of the algorithms for their realization providing polynomial dependence of computer time from dimension of the task has not been published; computer programs for their realization are kept by the author.

9. Ownership of the offer belongs to Mark D. Kats.

Kats Mark Davidovich, Dr. Sc., Address: Donetskaya St. 37-24, Severodonetsk, Lugansk Reg., 349940, Ukraine, E-mail: root@ixt.lg.ua (Kats M.D.).

Attachment to the project:

"DEVELOPMENT OF THE INTELLIGENT SYSTEM FOR FORMAL SELECTION OF OPTIMAL STRATEGY AS TO MEDICAL TREATMENT OF DISEASE X TAKING INTO ACCOUNT PATIENT'S INDIVIDUALITIES"

We propose to MEDICAL CENTERS to participate in development of intelligent systems having no analogs in the world and realized using PC and formal procedures:

- to engineer informative mathematical models of particular diseases;

- to synthesize optimal strategy as to medical treatment of particular disease taking into account patient's individualities.

PROGRAM OF ENGINEERING THE OPTIMIZATION SYSTEM FOR MEDICAL TREATMENT.

1. CURRENT STATUS OF THE PROBLEM.

The problem of treatment optimization taking into account patient's individualities is the second to diagnostics factor by its significance.

However, correct methods of construction of disease models are not available , that is why selection of optimal treatment tactics is rather the art of qualified professionals than a science.

Even when the diagnosis is correct the selection of optimal treatment remains a complex problem which is not always successful when an intuitive approach is used.

If we carry the analogy over biological and technical systems, the medical treatment can be compared with control in technical processes. Controlling technological processes without a mathematical model is not only incorrect from the point of view of current scientific paradigm but also indecent.

Specifying the decision for effective medical treatment one should base on mathematical model of the disease under consideration and on formalized methods using parameters which include patient's individualities and clinics, thereby synthesizing the most effective treatment procedure.

2. FORMULATION OF PROBLEM FOR MEDICAL TREATMENT OPTIMIZATION

The problem of formal (objective) selection of optimal strategy for treatment of any particular disease can be divided into 2 tasks:

- construction of mathematical model of the disease;

- development of formalized procedure for selecting optimal strategy of the treatment using this model.

Statement of the 1 task:

Given: results of observation and treatment of patients having the disease under investigation, in form of the table with each line containing information concerning 1 patient:

- characteristics describing patient's individualities (color of eyes, blood group, rhesus factor, etc.);

- characteristics describing the disease: anamnesis, physical examination, laboratory and instrumental methods, clinical record, etc.);

- parameters describing the treatment applied;

- estimation of patient's state after the treatment applied.

It is needed: Construction of mathematical model of the disease combining patient's individualities, disease proceeding , the treatment applied with results of this treatment .

Statement of 2 task:

Given: Mathematical model of the disease combining patient's individualities, disease proceeding , treatment applied with results of this treatment .

It is needed: Select optimal strategy of the treatment basing on situation characterizing patient's individualities and clinical proceeding of the disease with the help of formalized procedure and the model of disease.

3. METHODS SOLVING THE PROBLEM OF FORMAL SELECTION OF OPTIMAL MEDICAL TREATMENT STRATEGY FOR A PATIENT.

To solve the problem of mathematical model construction the method of mosaic portrait can be used.

If the table of experimental material has been obtained basing on investigation results, with each line containing the symptom characteristics and estimation of treatment quality for 1 patient , then the method of mosaic portrait is realized by the following formalized procedures:

- Transformation of table of initial experimental data M into table M' by means of formalized procedure of dividing the range of symptom characteristics into subranges ( transferring from continual scales where symptoms characteristics are measured to discrete scales of nominations. Every subrange of each symptom characteristic is assigned its own name (code);

- Search of the code combinations which are found in the table M' in the lines belonging to the same gradation of treatment quality and are not found in any line belonging to other gradations.

To synthesize the optimal set of medical procedures the novel method has been developed basing on mosaic model of the disease (method of situational programming) allowing to synthesize the necessary control (scope and parameters of the treatment) with the help of the mosaic model of disease and the situation, i.e. characteristics describing patient's individualities and disease proceeding thereby resulting in maximal effectiveness of the treatment course.

4. MAIN TASKS AND THE FINAL OBJECTIVE OF THE WORK.

The main task of the work is to develop an effective intelligent system not available in the world practice which is realized using PC and allowing by means of formalized procedures and mosaic model of the disease to synthesize optimal strategy of medical treatment for a patient.

After the intelligent system for medical treatment optimization has been developed and patented it will be executed as a PC software package for commercialization.

The final objective of the work is commercialization of the PC software package to medical centers.

5. PROGRAM OF WORKS

Development of models for optimal treatment strategy can be carried out jointly with medical centers of different profiles - scientific, research, training , diagnostic or hospitals which are:

- specializing in studying (treatment) of particular diseases;

- having at disposal qualified staff and modern diagnostic equipment to correctly estimate results of the treatment;

- having possibility to examine up to 100 patients having the same disease within required time period.

5.1. DETAILED DEFINITION OF PROBLEM. COLLECTING EXPERIMENTAL INFORMATION REQUIRED.

5.1.1. Choice of disease for investigation.

5.1.2. Working out the list of input variables characterizing the disease under investigation.

5.1.3. Working out the list of input variables characterizing patient's constant individualities which do not change during his life: eye color, blood group, rhesus factor.

5.1.4. Working out the list of procedures applied to a patient during medical treatment.

5.1.5. Input of estimation of treatment quality (e.g. 1- decline, 2- no changes, 3- improvement, 4 - significant improvement).

5.1.6. Development of formalized anamnesis comprising the list of variables as per items 5.1.2.-5.1.4.

5.1.7. Determination of minimal required and sufficient scope of experiments for constructing mathematical model of the disease under investigation taking into account patient's individualities.

5.1.8. Special examination and treatment of patients:

- recording patient's parameters according to the form as per item 5.1.6. and results of the treatment according to gradations specified in item 5.1.5.

When examination is finished, the table is to be developed with each line containing information on the symptoms' characteristics (see item 5.1.6.) and estimation of treatment quality for 1 patient (see item 5.1.5.).

5.2. DEVELOPMENT OF INTELLIGENT COMPUTER SYSTEM COMPRISES THE FOLLOWING STAGES:

5.2.1. Construction of mathematical model of the disease under investigation.

5.2.2. Development of computer program synthesizing optimal treatment strategy based on the disease model and the situation describing patient's individualities and conditions of disease proceeding.

5.2.3. Comparison of statistic data on effectiveness of traditional and synthesized formalized treatment.

5.2.4. Development of intelligent medical system, i.e. development of program package providing the solution of the problems defined.

5.3. PATENTING OF INTELLIGENT MEDICAL SYSTEM.

The models obtained by means of mosaic portrait and situational programming contain a large scope of new and previously unknown in medical science information allowing patenting of the system.

6. COMMERCIALIZATION OF MEDICAL SYSTEM

6.1. Advertising activities:

- development of demonstrational diskettes;

- publications in different medical magazines including foreign ones; reports and presentations of demonstrational versions at scientific conferences, dispatching advertising materials to medical centers, etc.

6.2. Commercialization of the system.

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