Chornenko M., Petrenko A., Timoshenko Z.

Oles Honchar Dnipropetrovsk National University


Using information technology in the processing of research results is an important and promising. In the modern world the urgent issue of the use and implementation of new information technologies in medicine. This treatment results of medical research helps professionals to put correct diagnosis and save lives.

Timely delivered accurate diagnosis is often easy choice of treatment and increases the likelihood of recovery of the patient. Based on all these considerations, it is natural to try to determine the conditions under which a diagnosis can be made quickly and accurately.In recent years, through the use of modern methods of treatment and diagnosis, based on the latest achievements of science and technology, the possibility of obtaining successful results increased significantly. If during the diagnosis must take into account a large number of which exhibit significant natural variability, it is very effective for describing complex scheme of their influence there is only one way – using appropriate statistical techniques. If the number of factors or the number of categories of data is very large, it is desirable, or even necessary, to use of the computer to the desired results can be obtained in a relatively short time. This approach in no way diminishes accuracy. Instead, it opens up even more space for the manifestation of these qualities, freeing the doctor from having to deal with such problems, which can be expressed in numerical and logical form and hence solve mathematical methods and using computers.

As the cost of integrated diagnostics is very large, it is important to know how to get the most useful information at the first stage of research, especially at the primary health care level. Among the primary methods of medical diagnosis should allocate a general analysis of blood, because his indicators reflect changes that occur throughout the body. Clinical analysis of blood (total) – quantitative and qualitative research elements forming blood.

In modern hospitals for the overall blood use hematology analyzer. Hematology analyzer designed to conduct quantitative and qualitative research of blood in clinical diagnostic laboratories. Automatic hematology analyzer designed for clinical laboratories is a fully automated device. Able to handle tens of samples per hour with high accuracy and reproducibility, and store test results in internal memory and print them on a printer or transmit to a computer for processing measurements.

After receiving the data on the computer processing. Follows for example, consider the inflammatory disease. Clinical practice has been established that inflammatory disease characterized by increased erythrocyte sedimentation rate (ESR), increased white blood cell count, increased the number of monocytes and a decrease in erythrocyte count, hemoglobin level. To automate the statistical processing, as well as for more rapid and quality processing of survey data there were compiled matrix measurements of blood taken by hematological analyzer, of healthy and sick patients.

To determine the most informative indicators used pair group method of data handling (GMDH). GMDH method was proposed by Academician AG Ivakhnenko. GMDH – an original method for solving structural-parametric identification of models or simulations with experimental data under uncertainty. GMDH differs from other methods of building models because that does not require knowledge of the laws of distribution. Output reduced to one scale measurements by their normalization.

Calculate the conditional probability of recognition errors (P0) through the conditional probability of adoption of norm as the norm (P11) and disease as the disease (P22).

Calculate the probability of an erroneous decision for 10 pairs формулаформулаформула,формула,формула,формула,формула,формула,формула,формула.This is implemented to using Mathcad, which allows you to process the data and algorithms required to implement.

Fig. 1. Conditional probabilities by GMDH

Fig. 1. Conditional probabilities by GMDH

P11-conditional probability norm as the norm, P22-conditional probability of acceptance of the disease as the disease, P0-conditional probability of error. The effectiveness judged by the value of the conditional probability of error. The lower the probability of error is better. During the work it was found that the most informative parameters are the erythrocyte sedimentation rate, number of erythrocytes, quantity of hemoglobin and white blood cell count. Informative was the first pair. Informative, these figures are not only in terms of statistical treatment of results, but also in terms of medicine. In further processing it is possible only on the two parameters to conclude that the probable condition of the patient. Using statistical criteria such as Student, Snedekora and others criteria.


Application of the proposed information processing research results is very broad and can be used not only for analysis of blood and not just for inflammatory diseases, but also for other forms of medical diagnostics. Application of statistical criteria in the new information technologies is very important and can qualitatively handle large sample measurement, if realize them on the basis of programs like Mathcad and others, you can not write a formula for each treatment, and only change the data measurement and evaluation criteria. Information-measuring technologies improve work diagnosticians but require material costs, but in our country it is not understood and therefore not enough to finance innovation.

The list of references:

1. Малайчук В. П. Математична дефектоскопія / В. П. Малайчук, А. В. Мозговий. – Дніпропетровськ: Системні технології, 2005. – 179с.

2. Малайчук В. П.  Інформаційно-вимірювальні технології неруйнівного контролю / В. П. Малайчук, А. В. Мозговий. – Дніпропетровськ: РВВ ДНУ, 2001. –237с.

3. Малайчук В. П. Обробка вимірювань і сигналів неруйнівного конт­ролю / В. П. Малайчук, О. М. Петренко. – Дніпропетровськ: РВВ ДНУ,2010. – 138с.

4. Малайчук В. П. Основи теорії ймовірності і матеметатичної ста­тистики / В. П. Малайчук, О. М. Петренко. – Дніпропетровськ: РВВ ДНУ, 2001. – 124с.

5. Научно-технічний портал. – Режим доступу: patents_medicine/medicine_17/medicine_274.shtml

Справочник по клиническим функциональным исследованиям / под ред. А. Гиттера. – 2-е изд., доп. – М.: Медицина, 1966. – 612с.