2019 № 3 Development of algorithm for searching of clinically homogeneous patients from semistructured text data of oncological electronic health record
The growth in the number of patients with malignant neoplasms in Russia significantly increases the load on a specialized network of oncological institutions and oncologists. It is most likely that this trend will continue in the coming years. One of the ways to improve the efficiency of medical activity is the extraction knowledge from medical data arrays, using modern data analysis methods, by clustering patients into groups of clinically homogeneous (similar) patients from electronic health records. The aim of the study is to develop an algorithm for finding clinically homogeneous patients according to the electronic health records of the oncological dispensary, with follow-up possibility of integration into the clinical decision support system (CDSS). The use of such CDSS in practical medicine and in the field of medical education will allow us to analyze both semistructured and unstructured arrays of information, which will require further implementation and improvement of information systems at all levels of
medical care. The homogeneity of patients was determined by machine learning by cosine distance in the space of vector representations of electronic health records. An experiment on 20 randomly selected electronic health records of patients of Krasnodar Regional Oncological Dispensary showed high efficiency of the algorithm in creating clusters of clinically homogeneous patients.
2018 № itm Forecasting the stage of adenomiosis with neural networks
Adenomyosis is a widespread gynecological disease, which is often accompanied by infertility. There are problems with the diagnosis of the disease, since the disease has various clinical manifestations, including often asymp- tomatic course of the disease. Due to the complexity of the diagnosis and according to different sources, its frequency varies from 5% to 70%. An equally difficult problem is determining the stage of the disease. The stage determines the tactics and strategy of treating patients. A sample of 84 patients with adenomyosis, using the Spearman rank correla- tion coefficient, revealed indicators that are interrelated with the stages of the disease. In this work, the application of the heuristic procedure to neural networks for predicting the laboratory-clinical indicators of the adenomyosis stage is considered. A software application has been developed that allows you to predict the stage of adenomyosis without resorting to hysterectomy. The methodological value of the work is that, using the example of a common gynecological disease, it is shown that the use of modern data analysis tools opens up wide possibilities for solving prognostic prob- lems of determining patients’ belonging to certain classes according to the stages or types of the disease. Software applications that automate the procedure for classifying patients can form the basis of various systems of support for making medical decisions.
2016 № 5 Ranking definition method of health insurance companies based on hierarchical classification.
The official site of Federal compulsory medical insurance fund presents 21 activity indicators for the 59 health insurance organizations (HIOs) for 2015. Each indicator represents an activity efficiency criteria of HIO in the certain area as following: its larger or smaller value defines whether the company activity is successful or not. The HIO rating is defined per each of these criteria. Undoubtedly, HIO ranking definition problem on set of all the considered criteria is becoming particularly urgent. The purpose of this work is to apply the metrical approach for HIO activity efficiency assessment, based on the hierarchical classification method that assumes the distance calculation among HIOs as among the points of multidimensional space in the coordinate system corresponding to the criteria of their efficiency. The concept of reference organization that accumulates the best value of all of the HIO criteria is introduced in this paper. The ratings of insurance companies are established in accordance with the distance to the reference organization by the principle – the smaller the distance, the more efficient the company is and the higher its rating is (the place in the HIO list). The method is implemented in the environment of the STATISTICA statistical package. HIO ratings, defined by the proposed method can be used to identify promising and unpromising HIOs in order to plan their upcoming development or structural changes, they can also act as the specific navigator when choosing the health insurance company
2018 № 2 Decision Support System for Choosing Correction Tactics of Internal Carotid Arteries Stenosis
Along with the medical systems development there is an important task on creation of medical decision support systems (DSS), in particular, capable of predicting the probability of postoperative complications. Computer methods of data analysis make it possible to successfully use both classical methods of applied statistics and modern heuristic procedures for identifying latent (hidden) knowledge in patients’ databases with subsequent construction of predictive models. The article describes the DSS, which automates the prediction of complications probability in the surgical treatment of internal carotid arteries stenoses by carotid endarterectomy and angiostentiation methods basing on the clinical parameters of the patient’s condition before treatment and the technological parameters of the operative intervention. The DSS is based on such classification methods as classification trees and neural networks, the training sample for the classification is the database of patients, who have been treated, with the information of complications presence or absence availiable. Studies, preceding the DSS development were implemented in the STATISTICA package environment. The entrance into program is automated. At the user’s require, the patient data, needed for the calculation is imported into the prediction program module of complications probability from the Excel table. Also, at the user’s request, the prediction results can be saved in the source table.
2017 № 3 From PACS to teleradiology
This article is about the current situation of the distribution and exchange of medical images. Today PACS is familiar to medical facilities and it is widely used system in radiology departments, but the “teleradiology” term still has difficulties with interpretation. When digital medical images are printing on a film and then transfer to other specialists, this way of exchange becomes more and more expensive and inefficient. That’s why the problem of acquisition, dis- tribution and getting access to the results of the diagnostic studies still actual. Another problem is the lack of qualified radiologists in the regions of Russia, due to the uneven population density; competent specialists are concentrated in regional centers. Therefore performing of urgent remote consultation is an actual problem too. In this article have been defined perspectives of advancement of the teleradiological systems, that began to appear due to PACS implementa- tion in radiology departments. The purpose of this scientific research is to develop the methods of teleradiology systems organization for consultations using PACS and telemedicine networks created in the regions. Factors influencing the quality of teleradiology consultations were discussed in article. Prerequisites for the development and implementation of such socially important systems, as a regional teleradiology system were defined
2018 № 4 Development of an algorithm for automated wavelet analysis of clinical oncology dispensary registry data at the regional level
This article describes the issue of overload of the registry on the example of the state budgetary health care institution clinical oncological dispensary No. 1 of the Ministry of health of the Krasnodar territory. Maintaining a schedule of appointments and appointments to the doctor in electronic form plays a key role in healthcare information systems as the most widespread and socially oriented medical service. The aim of the study is to develop an algorithm for automated wavelet analysis of data on the work of the registry. To achieve this goal, there are used the methods of wavelet analysis of information that have been applied in practice and can be reused in other medical institutions. It is possible to create an optimal schedule of work of registrars and as a result reduce the waiting time of patients at the reception, by automating the process of analysis of data on patient’s appointments that contained in the healthcare information system. The reduction in waiting time should increase patient satisfaction and improve the overall impression of the health facility. For each institution, the workload schedule of the registry will be different and may change over time. It follows that the analysis and optimization of work schedules should be carried out regularly and automatically.
2018 № 4 Radiology information system of the Krasnodar region: development of the reference clinical and diagnostic center for breast pathology
In order to increase the availability and quality of specialized oncological care for the population of the Krasnodar Territory, the clinical and diagnostic Center of breast pathology was created. Teleradiology is one of the most efficient way for remote consultations. The article contains a description of the methodology for organizing the work of the Reference Center using telemedicine technologies via Radiology in-formation system of the Krasnodar Territory (RIS). DICOM modalities in diagnostic departments, local PACS in medical facilities are the part of regional RIS and also include 2 mammography units located in interterritorial oncological dispensaries. Every month new mammography exams are delivered to local PACS, which determines the relevance of their remote consultation. To fully operate the Reference Center as key participant in the system of teleradiology consultations, it is necessary to create a complex diagnostic information system in the regional clinical oncology dispensary. This is one of the results of the paper. The use of a unified system increases the efficiency of the information technologies in a remote mammographic screening and also allows standardizing and formalizing activity of a radiologist according to recommendation such as BIRADS, facilitates and speeds up the routine work of the radiologist, improves the quality of radiology interpretation, provides to radiologist effective help, using computeraided diagnostic
2017 № 3 Online method medical-economic control of prescriptions for preferential drug provision
There is a high probability of making mistakes (intentional and unintentional) while filling the electronic form of the recipe for preferential drug provision, that may lead to violations in the prescription of drugs. Expert detection of prescription form filling errors after the drugs had been released to the patient in pharmacy can’t prevent the negative impact on patient’s health. The purpose of the study is to improve the effectiveness of organizing and performing the medical control of drugs prescribing, to simplify procedures for providing citizens with preferential medicines, applying the developed methods and algorithms for real-time verification of the selected recipes. The paper represents the method and the development of decision making support algorithm to be used while performing the medical expertise, also called medical -economic control. The opportunities of unified information system for prescribing and releasing of preferential recipes in the Krasnodar Region are observed in the paper in the frame of implementation of the “electronic prescription” technology and medical expertise of preferential drugs prescribing in real time. The prospects of medical expertise performing for detecting cases of polypharmacy in advance are established in cases of multiple drugs prescription (more than five). The importance of performing medical expertise was defined for all kinds of prescription, not only for preferential drugs, consequently the suggested method and developed algorithm should be applied to medical expertise of all kinds of electronic prescription in real-time