Y Balykina

Analysis of standard clustering algorithms for grouping MEDLINE abstracts into evidence-based medicine intervention categories

The paper describes a process of clustering of article abstracts, taken from the largest bibliographic life sciences and biomedical information MEDLINE database into categories that correspond to types of medical interventions - types of patient treatments. Experiments were carried out to evaluate the quality of clustering for the following algorithms: K-means; K-means++; Hierarchical clustering, SIB (Sequential information bottleneck) together with the LSA (Latent Semantic Analysis) methods and MI (Mutual Information) which allow selecting feature vectors. Best results of clustering were achieved by K-means++ together with LSA then 210-dimensional space was chosen: Purity = 0.5719, Entropy = 1.3841, Normalized Entropy = 0.6299.

Optimization Approach for Estimating the Required Amount of Pharmaceuticals in the Russian Federation

OBJECTIVES To propose an algorithm that relates the effectiveness of drugs for a wide range of diseases with the financial capabilities of patients. METHODS Estimates of the volume of pharmaceuticals that are consumed in the Russian Federation by all segments of the population regardless of household income were considered. These were calculated using statistically valid probabilities of the appearance of various diseases, official state data on the structure of expenditures of various strata of the population, and the optimal choice of the most effective medicines with income restrictions taken into account. The main idea was to introduce the utility function of the drug and the cost of treatment. For each disease, its own set of drugs was selected. RESULTS On the basis of the real-world data for several diseases, optimal estimates were calculated using the proposed algorithm. In the process of approbation, some weak points of the algorithm were found, such as the methods of packaging pharmaceuticals and associated cost of a packaging unit. These characteristics should be discussed separately, introducing conventional units of drug volumes. A unit of quantity corresponding to the maximum effect of the drug in question is proposed in the work. CONCLUSIONS The proposed algorithm for estimating the amount of medicines can be successfully used by both pharmaceutical (or dealer) companies and government agencies for objective population provision. The usual sources of such estimates are based either on market surveys or on pharmacy network data. Both ways are very expensive and do not allow predicting mass demand in the future, for example, with an unexpected epidemic or the emergence of new medicines. In addition, the proposed algorithm can be successfully applied to the pricing problem: a variation in price may show a change in the volume of use.

Ambulance resources reallocation in St. Petersburg using imitation modelling approach

The emergency medical system is an important element of the citys infrastructure. A method for solving the problem of ambulances distribution, using simulation tools in combination with the multi-periodicity of the system is considered.

Online optimization algorithms for multi-armed

Nowadays the online learning area is actively developing as a part of machine learning. In this regard, there arises the problem of choosing an algorithm that solves the optimization problem with regard to online data processing. Since currently one of the active areas of online learning is ranking, the comparison of several state of art online optimization algorithms for the multi-armed bandit problem in case of online ranking is presented.

On the question of zero-controllability of nonstationary bilinear systems

Controllable systems are considered. Sufficient conditions for global null-controllability are obtained for a given set of admissible controls. The Lyapunov functions are used for analyzing the presence of an asymptotic stationary point in the system.

Budget allocation planning for multi-sectoral investments

The budget system of any local authorities and government provides the means to decide how much money to spend or how many resources to allocate to government institutions or agencies for the purposes of regional social-economic development. In the paper the Rolling Support Planning Algorithm (RSP Algorithm) is represented, which can be used for budget allocation for a given time horizon based on Key Performance Indicators (KPIs) which are formed by the budget framework policy context. This algorithm allows to solve the budget allocation problem in the case when achievement of target values of KPIs are planned in different future periods.

Mathematical model of ambulance resources allocation with multiperiodicity

Emergency medical system is one of the main elements in city infrastructure. The article contains analysis of existing system of ambulance resource distribution. Paper considers the idea of using multiperiodicity as a tool to increase the efficiency of the Emergency Medical Services. Comparation of different number of sub-periods to evaluate an optimal quantity of them.

Pharmacoeconomic Analysis of the use of Everolimus Compared to Axitinib in Second Line Therapy of Patients with Metastatic Renal Cell Carcinoma.

diversity of the health professionals and the basic scenario. The costliest scenarios were the one implementing HPV DNA testing which did not provide further participation despite a high cost and the one based on P4P incentives towards GP, although it allows high participation rates. ConClusions: Using a comprehensive BIM, we show that full coverage of OS might be the most cost-effective way to implement it, although practical and financial issues might favour other scenarios that may be more balanced regarding the distribution of costs between stakeholders or may be more easily implemented and accepted by health professionals.

Budget Impact Analysis of the Treatment of Chronic Myeloid Leukemia with Tyrosine Kinase Inhibitors – Nilotinib in the First and Second Lines of Therapy

chemotherapy after the index date to the earliest of mean time to HSCT, death, loss to follow-up, or last chemotherapy dose plus 30 days. The primary endpoint was the percent of time in the hospital during the salvage chemotherapy period. Key secondary endpoints were number of hospitalisations and length of hospital stay. Hospitalisations associated with HSCT were excluded. Results are presented as mean (SD) unless indicated. Results: Twenty-two patients were included, with a mean age of 44 (18) years. After the index date, 19 patients died and 8 patients received a HSCT. During the chemotherapy salvage period, patients spent a mean of 56% (95% CI: 46%-69%) of their time in the hospital. There were a mean of 2.2 (1.5) inpatient hospitalisations, 3.2 (6.2) day stays, and 1.6 (3.0) outpatient visits per patient, and the mean length of inpatient hospitalisation was 20.0 (20.0) days. From the index date to death, there were a mean of 2.8 (1.4) inpatient hospitalisations, 4.7 (7.3) day stays, and 4.6 (7.4) outpatient visits per patient and the mean length of inpatient hospitalisation was 19.0 (19.0) days. ConClusions: Adult patients receiving salvage chemotherapy for R/R ALL in Italy spend more than half their time in the hospital. Costs of hospitalisations will be presented.