Українська   Русский
DonNTU   Masters' portal

Abstract

Content

Introduction

Modern diagnostic systems are very diverse and have a wide range of applications. The introduction of computer control systems health care helps to ensure the timeliness and effectiveness of therapeutic and diagnostic measures, increase the quality of decisions, minimize the likelihood of uncontrolled access to patient data and malicious use. In medicine there are more diverse high-tech equipment (diagnostic, laboratory) that are embedded in automated systems processing of medical data. This necessitates the creation of new technologies for data processing, ensuring their authenticity, availability, integrity and confidentiality. Thus, development of effective technology of processing of medical data, allowing to perform a diagnosis of the condition of the patient, it remains an urgent task.

A special role to play in diagnosing specialized computer systems designed for detailed examination certain aspects of the health status of the patient. Interest a system that uses a non-invasive approach to the diagnostic implementation that allows to increase the availability, security and speed of the survey.

1. The relevance of the topic

Currently, such systems are needed to conduct surveys of pulmonary status the respiratory system of the patient. In this case, the source of information about the damage to the respiratory tract, inflammatory processes and the effectiveness of treatment can serve as an analysis of the qualitative and quantitative changes of composition of the exhaled patient air.

2. The goals and objectives of educational services

The aim of the study is to develop a software architecture and further design an automated computer system designed to conduct non-invasive pulmonological diagnostics.

To our tasks include: domain analysis and the physical foundations of chosen method, diagnostics, formalization of the initial data, the formation of functional and non-functional requirements the software product, the analysis of the precedent system and the basic algorithms of data processing.

3. Domain analysis and formalization of the source data

Modern diagnostics of the diseases of the respiratory system involve laboratory and instrumental studies. In the same time a biophysical method of laser correlation spectroscopy (LCS) increases its usage, its indisputable advantages are high sensitivity, speed of obtaining results and non-invasiveness [1]. . Method LKS fairly new, it is based on measuring the spectral characteristics of monochromatic coherent radiation due to the scattering of light when passing through a disperse system of the nanoparticles suspended in a liquid. Interaction radiation particles of this system expands the range of light scattering, the shape of spectral lines characterizes the dispersion the system, with high precision showing the concentration of particles in the size range of 1 to 10,000 nanometers. From the spectrum characteristics the scattered monochromatic coherent radiation can be obtained information about all dynamic processes in the system under study.

Directly measured value in the method LKS is not the spectrum of the scattered light analyzed system, and the range of fluctuations of the photocurrent at the output of the photorecording device. This spectrum is the result of a beating harmonics of the electromagnetic fields with each other, it is concentrated in the low frequency region, that is where it is very it is convenient to analyze the modern powerful radio methods. Measuring the spectrum of fluctuations of the photocurrent and determining its half-width, it is easy to obtain the size of the particles in the system under study. In reality, however, the study size particles (especially biological fluids) are rarely monodisperse. As a rule, the study of polydisperse samples, that is, at the same time in solution are particles of various sizes. Let the spectrum of light scattered by monodisperse particles is a curve of Lorentz:

1

where 1_1 — the half-width of spectrum at half maximum, q — transferred the wave vector of the light scattered by the sample,

2

where n-the refractive index of a medium, λ — wavelength of light, θ — the scattering angle, DT — the coefficient of translation diffusion of scattering particles. Through it you can move on to sizes the scattering particles.

For a polydisperse system spectrum is a sum, and for continuous distributions — the integral of lorenzello with different half-widths G . In this case, the spectrum I (ω):

3

where A(G) — a function of the distribution of particle diffusion coefficients 1_1, and, consequently, in size. The distribution of particle size is thus in the solution the above integral equation with a Lorentz kernel.

The problem of this type is characterized by a strong instability of the solution with respect to small variations experimental data that need to be considered.

For the non-invasive pulmonological examinations as source material effective use of the condensation of moisture exhaled air (KVVV) of the patient, which is obtained using a special device and then placed in the laser spectroscope. LKS KVVV allows, based on the ratio of nanoparticles of a particular hydrodynamic radius, to determine the status of the dy-hateley system and then monitor the effectiveness of treatment and predict its results. Spectra of the studied KVVV appropriate to measure under the same conditions, that is at the same temperature, humidity, etc. If other conditions and cannot be reduced to standard, it is necessary to introduce appropriate amendments.

The distribution function of the particle size KVVV — is the histogram defined on the mesh sizes consisting of 32 points moreover, the mesh sizes are the same for all the considered KVVV. Although visually it is possible to identify certain correlations between disease of the respiratory system and the functions of distribution, to draw a valid conclusion, based on direct considering these LKS is very difficult, because these data are 32 correlated between numbers and analysis of their "mind" is not possible. Therefore, it is advisable to apply an automated approach to design computer system, identifying as the respiratory system of the patient based on his personal medical and KVVV-data [2, 3].

4. Functional and non-functional requirements

For the formalization of requirements to developed system was conducted to collect and analyze information on the basis of documents "Vision" formed by the future users of the system. Users of the system should be divided into the following categories:

In addition, some system functions have a specific artist, while others can be performed by different categories users (e.g., registration, authorization, etc.).

Consider the basic functional requirements for the developed system diagnostics. Formally describes the functionality and behavior of the system is shown in Fig. 1.

The system must ensure that the registration (the primary use of the system) and authorization (the use of the system in the presence of the account) user. The system must provide the possibility of making major personal, medical and additional (auxiliary) data about the patient. After the study of the spectral the composition of the condensation of moisture exhaled by the patient air system must record KVVV-examined the data.

The program should provide for the formation of the medical records of the patient for which the following is possible: create a new map with a unique number, viewing, Supplement an already created map data, editing, printing, storing in the database, search the card in the database by various criteria — simple and compound. Each patient if necessary, can be examined any number of times, making all of the survey results in its medical records. As needed ability to group patients to view sample data or task a specific action for the group of patients (e.g., to send diagnostic results to multiple patients). In the case of contact of patient data the system should send the diagnostic data at the specified email address or via SMS. In the absence of necessary information during a database lookup or grouping is to be displayed the corresponding message. It is also necessary to issue an error message when there is no access to the database. The system should not distribute personal information about the patient.

The program should perform the processing and analysis of the integrity of KVVV-patient data, to carry out further spectral analysis KVVV-data, test spectra of conformity to the normal distribution law, calculate the values diagnostic signs in accordance with laid down procedures and algorithms. On the basis of the calculated values diagnostic features identificireba condition of the respiratory system of the patient. At impossibility of automatic identification involved a doctor who associates the patient with one of the existing diagnostic groups or creates the patient's new group. Each group should have its description stored in the appropriate database tables.

Each user has its own limitations of use: technician can enter data about the patient, forming a medical information card to perform the calculation of diagnostic features, as well as the calculation of spectra and checking normality KVVV-patient data. All health care providers can change the customer data, but to update medical records can only a doctor. Also the doctor can correlate patient data with diagnostic group and upgrade diagnostics (recalculate the statistics for the group). The administrator is responsible for forwarding the diagnostic information to the patient if it left their contact details (mobile number, e-mail).

As a software product intended for use by health workers owning computing equipment at the user level, it needs to be lightweight and easy to use, it is convenient to have a friendly interface. The action job should be carried out using menu commands, buttons, icons, main action must match the clues. In addition, it is necessary to develop available help system, allowing you to quickly and easily to master the principles of working with the program.

4

figure 1 — Diagram of use cases (use case diagram) of the design system

5. Analysis of the precedent system and the basic data processing algorithms

Consider the specification of the use cases of the system — the precedents given in the table. For each use case it says here:

Precedents Stakeholders Precondition Postcondition Main successful scenario Alternative
1.Registration All actors Program is running, have the right to register. in accordance Wi. profile User registered All fields are filled in,the user entered in the database check again
2.Authorization Paramedics, administrator User registered User voxel system login Correctly entered the password,the authorization is finished re-enter username and password, after attempting to set complement. the secret question or enter the phone number (if it is stored in the database), which will do the login details
3.Testing the Patient, health workers Equipment in working condition, the medical staff is present the Patient underwent examination Survey completed, the analysis results obtained Re-examination, waiting for results
4.The input of personal data (basic) Laboratory, the patient Patient and the technician was Data entered you Entered all the necessary data Adjustment of data, enter some data (for required)
5.The input of personal data (medical) Laboratory, the patient Patient and the technician was entered master data Medical data entered you Entered all the necessary data Adjustment of data, enter some data (for required)
6.The additional input Laboratory, the patient Patient and the technician was entered basic personal and medical data Additional data entered you Entered all the necessary data Adjustment of data, enter some data (as needed)or skip this item
7.Enter KVVV-data Laboratory, the patient Patient and the technician was entered basic personal, medical and supplementary data KVVV-data entered you Entered all the necessary data Adjustment of data, enter some data (for required)
8.The formation of the medical card Laboratory, the patient you Entered all the necessary data Medical card created All patient data entered in the database, medical record ready Enter the missing data to correct incorrect or obsolete information
9.Shipment of diagnostic data patient Administrator, patient Administrator was the results of the survey and the diagnosis recorded in medical records Data transferred to the patient If the patient indicated your phone number and/or e-mail, the results of diagnostics are coming to him via e-mail and/or SMS Waiting patient
10.The calculation of diagnostic features Laboratory Entered KVVV-data Diagnostic features are computed Signs of the calculated true recalculation of the signs, checking the fidelity of the data entered, re-diagnosis
11.Calculation of the spectra and check the normality of the spectra KVVV-patient data Laboratory Entered KVVV-data Spectra calculated from the test is Spectra calculated successfully, KVVV-data full spectra respectively.norms. law of distribution calculation of results, verification of fidelity of the data entered
12.Modification of the client's data Paramedic Changed some personal data about the patient restated Data was changed correctly stored in database Re-enter changed data
13.Update the medical card Doctor Changed some medical data about the patient restated Data modified successfully saved in database Re-enter changed data
14.Identification of the patient's condition Doctor Entered KVVV-data, diagnostic features and the calculated spectra, tested the normality of spectra Specified diagnostic group Diagnostic group of the patient is identified correctly, the probability of error is small cross-checking of diagnostic procedures, validation of entered data, determination of a patient in a special group, a group definition by the doctor
15.Update statistics for the group Doctor Entered KVVV-data, diagnostic features and the calculated spectra, tested the normality of the spectra KVVV Specified diagnostic group, statistics restated Diagnostic group updated Re-recount statistics
16.Patient search Paramedics Need to find the patient's chart for viewing and/or changing its data the patient was found enter the search criteria of the patient, output information of the maps on the screen the Patient is not entered in the database incorrectly entered the search criteria — is required to re-enter

table 1 — Specification of precedents

Consider the processes in the proposed diagnostic system. In Fig. 2 shows a UML activity diagram for process of "View/create a patient account". As can be seen from the chart, after the request to work with the account patient should check authorization staff (technician/doctor). In the absence of authorization, you must enter the required the login/password. If authorization is successful, the staff sets the mode of operation of account — you want to view the existing database record or create a new one. If necessary, create a new record is requested personal, General medical, various ancillary data about the patient, and the results of its survey method LKS — spectra KVVV. All data except basic personal data can be entered later when returning to the account. To work with existing record is entered, its parameters to locate the record in the database. After viewing account information work with her terminated.

5

figure 2-activity Diagram (activity diagram) for the process "View/create account the patient's record."

In Fig. 3 shows a UML diagram for the main process — "the Identification of the patient's condition", allowing to relate the state of the respiratory system of the patient from a specific diagnostic group.

6

figure 3-activity Diagram for the process "Identification of the patient's condition."

To identify the required to search for a patient account in the database. In the absence of records must be issued an appropriate message and exit process. If you have an account and the data contained in it, checked for completeness. Must be filled in all the required fields of entry: as those that allow us to identify the identity of the patient, and those that allow its diagnosis. What follows is an attempt to automatically diagnosis. According to certain rules and procedures determined by the appropriate diagnostic group. In case of success the patient is correlated with the selected group. If a suitable group is not found, the diagnosis is a manual process. The doctor should be in one of two ways:

Identification Results are stored in a database and the process terminates.

In Fig. 4 formalized the feedback process with the patient — sending the diagnostic results.

7

figure 4-activity Diagram for the process "sending a patient of the diagnosis"

In the case when the account of the patient contains their email address and/or mobile number, they are used for feedback with the patient. If the patient is healthy enough to notify remotely. If the data for feedback is absent or the patient is sick, must be present to practice the feedback process.

Insights

Thus, in the framework of the present work was designed at the initial stage of the computer system allowing for automated diagnosis of the condition of the respiratory system of the patient based on the spectral data on the exhaled breath of the patient air.

When writing this abstract master's work is not yet complete.

The final completion: Jun 2017. Full text works and materials on the subject can be obtained from author or his supervisor after the specified date.

References

  1. Бажора Ю.И., Носкин Л.А. Лазерная корреляционная спектроскопия в медицине. – Одеса: «Друк», 2002. – 400 с.
  2. Комлевая Н.О., Комлевой А.Н. Разработка информационной модели диагностирования состояния дыхательной системы // Холодильна техніка і технологія. – 2011. – Вып. 2(130). – С. 75–79.
  3. Комлевая Н.О., Комлевой А.Н. Автоматизация диагностирования состояния дыхательной системы // Тр. XIII Междунар. научно-практ. конф. «Современные информационные и электронные технологии». – Одесса, 2012. – С. 55.
  4. Мацяшек Л.А. Анализ требований и проектирование систем // Разработка информационных систем с использованием UML. – М.:Вильямс, 2002. – 432 с.
  5. Разработка компьютерной системы для исследования заболеваний легких / Н. О. Комлевая, А. Н. Комлевой, К. С. Чернега // Проблеми програмування. - 2014. - № 2-3. - С. 253-262.
  6. Реброва О.Ю. Статистический анализ медицинских данных. – М.: Медиа Сфера, 2006. – 310 с.