[The efficacy of outpatient monitoring of children and adolescents with type 1 diabetes with regular use of professional continuous glucose monitoring].

BACKGROUND: According to research, only 38% of patients reach glycated hemoglobin targets. It is possible to improve the effectiveness of medical care for children with T1D using modern technologies, including continuous glucose monitoring (CGM). AIMS: To evaluate the effectiveness of outpatient monitoring of children and adolescents with T1D with regular use of professional continuous glucose monitoring. METHODS: The inclusion criteria: age 8−12 years; T1D at least 1 year; insulin therapy by multiple injections of insulin; inadequate glycemic control of T1D: НbА1с level of 7.5% and higher and / or children and adolescents with frequent episodes of hypoglycemia (usually 4 times a week) or with a history of severe hypoglycemia; signed informed consent. All patients initially and 12 weeks after inclusion in the study conducted a study of the level of НbА1с, and also performed CGM for 6 days. Based on the results of CGM, glycemia indicators and daily doses of insulin were recorded, treatment was evaluated and corrected, and recommendations for self-monitoring were made. Glucose monitoring was carried 120−144 hours using the blind method iPro2 (Medtronic, USA). RESULTS: In all, 99 children aged 8−18 years were included in the study in all centers. The decrease in the level of НbА1с by the end of the study was 0.72%, while the proportion of patients who reached the target level of НbА1с (defined as <7.5%) was statistically significantly higher at the end of the study (15.5% and 2%, respectively; p<0.05). During the study, patients showed a trend towards a decrease in the average level and variability of glycemia by the end of the study, however, statistical significance was achieved only in relation to the average level of glycemia (p=0.04). Conducted insulin therapy, determined by the average daily doses of long-acting and short-acting insulin, did not statistically significantly change at the end of the study. The frequency of DKA episodes and severe hypoglycemia did not statistically significantly differ from the initial level. CONCLUSIONS: For children with poor glycemic control of T1D, the use of professional CGM is effective in terms of glycemic control and a safe method.

[Remote monitoring and treatment of children and adolescents with type 1 diabetes].

RATIONALE: Continuous subcutaneous insulin infusion (CSII) is an effective method for optimizing glycemic control in children with type 1 diabetes mellitus (DM1). However, the use of CSII does not always result in adequate glycemic control. Telehealth can be applied as one of the methods to improve the effectiveness of treatment. AIMS: To evaluate the use of remote medical support of children and adolescents with DM1 and its influence on glycemic control, quality of life, and incidence of acute complications of DM1. MATERIALS AND METHODS: We conducted a 24-week multi-institutional prospective open-label controlled clinical trial. 180 children and adolescents were included in this study and divided into the following categories: 1) age 8-18 years; 2) DM1 at least 1 year; 3) pump insulin therapy Medtronic Paradigm (Medtronic MiniMed, USA) at least 6 months; 4) self-monitoring of glycemia at least 4 times a day and replacement of the insulin pump infusion system at least once every 3 days; 5) inadequate glycemic control of DM1: the level of glycated hemoglobin (HbA1c) 7.5% or higher. Patients were assigned to a remote consultation group (RC; n=100) or a traditional control group (TC; n=80). All patients were trained on the basic principles of DM1 and CSII, and we measured initial HbA1c, then after 12 and 24 weeks, also registered and analyzed glycemic indicators and daily doses of insulin, evaluated and corrected the treatment. Patients or their parents in the RC group sent pump data via the Internet to the pump insulin therapy center at least once every 2 weeks at home and received treatment recommendations in response. RESULTS: The total number of patients included in the study in all institutions was 180 children at 8-18 years. Patients in both groups did not differ in age, gender, duration of DM1 and CSII, and HbA1c level. The total amount of remote consultations for all institutions was 949. The decrease in the level of HbA1c by the end of the study against the initial one was statistically significantly greater in the RC group: 1.17% compared to 0.59% in the TC group (p<0.05). The proportion of patients who reached the target level of HbA1c (<7.5%) was significantly higher in the RC group (32%) compared to the TC group (12.5%, p<0.05). During the study, the incidence of DKA and severe hypoglycemia in the RC group was statistically significantly lower. CONCLUSIONS: Remote monitoring in children with DM1 resulted in significant improvements in glycemic control (HbA1c, glycemic variability, and hypoglycemic frequency). The accumulation of evidence on the effectiveness and safety of telehealth in DM should contribute to implementing this approach in practical health care.

Automatic Human Sleep Stage Scoring Using Deep Neural Networks.

The classification of sleep stages is the first and an important step in the quantitative analysis of polysomnographic recordings. Sleep stage scoring relies heavily on visual pattern recognition by a human expert and is time consuming and subjective. Thus, there is a need for automatic classification. In this work we developed machine learning algorithms for sleep classification: random forest (RF) classification based on features and artificial neural networks (ANNs) working both with features and raw data. We tested our methods in healthy subjects and in patients. Most algorithms yielded good results comparable to human interrater agreement. Our study revealed that deep neural networks (DNNs) working with raw data performed better than feature-based methods. We also demonstrated that taking the local temporal structure of sleep into account a priori is important. Our results demonstrate the utility of neural network architectures for the classification of sleep.

Crowdsourcing the creation of image segmentation algorithms for connectomics.

To stimulate progress in automating the reconstruction of neural circuits, we organized the first international challenge on 2D segmentation of electron microscopic (EM) images of the brain. Participants submitted boundary maps predicted for a test set of images, and were scored based on their agreement with a consensus of human expert annotations. The winning team had no prior experience with EM images, and employed a convolutional network. This "deep learning" approach has since become accepted as a standard for segmentation of EM images. The challenge has continued to accept submissions, and the best so far has resulted from cooperation between two teams. The challenge has probably saturated, as algorithms cannot progress beyond limits set by ambiguities inherent in 2D scoring and the size of the test dataset. Retrospective evaluation of the challenge scoring system reveals that it was not sufficiently robust to variations in the widths of neurite borders. We propose a solution to this problem, which should be useful for a future 3D segmentation challenge.

Anisotropic ssTEM image segmentation using dense correspondence across sections.

Connectomics based on high resolution ssTEM imagery requires reconstruction of the neuron geometry from histological slides. We present an approach for the automatic membrane segmentation in anisotropic stacks of electron microscopy brain tissue sections. The ambiguities in neuronal segmentation of a section are resolved by using the context from the neighboring sections. We find the global dense correspondence between the sections by SIFT Flow algorithm, evaluate the features of the corresponding pixels and use them to perform the segmentation. Our method is 3.6 and 6.4% more accurate in two different accuracy metrics than the algorithm with no context from other sections.

Stability of the saccadic oculomotor system.

A variety of different types of instability has been found in the saccadic system of humans. Some of the instabilities correspond to clinical conditions, whereas others are inherent in the normal saccadic system. How can these instabilities arise within the mechanism of normal saccadic eye movements? A physiologically-based model of the saccadic system predicts that horizontal saccadic oscillations will occur with excessive mutual inhibition between the left and right burst cells and with underaction of the pause cells. The amplitudes and frequencies of the oscillations had ranges of 0-6 degrees and 6-20 cycles per second, respectively. Application of stability analysis techniques to the model reveals that development of the oscillations can be explained by the Hopf bifurcation mechanism. Future development of this approach will involve classifying pathological instabilities of the saccadic system according to the bifurcation involved in their generation.