[Spectrum of food diversity of megapolis' inhabitants obtained by the analysis of buying activity]

The food diversity is a significant indicator of a healthy diet. The aim of the study was to assess the indicators of food diversity in the metropolis (Moscow) based on the analysis of consumer activity in a large retail chain. Material and methods. 201 904 holders of loyalty cards of the retail network were selected from the depersonalized purchase data for 2016-2018 according to the following selection criteria: the presence of a dense segment of purchases (more than 4 weeks, where there was at least 1 purchase in 2 weeks), where the total cost of purchases was at least 4700 rubles and at least 4 different products were purchased. From of these, 111 136 buyers have data available by gender, 43 046 by age. The classification of a set of purchased products and dishes by food groups was carried out, and complex dishes were transformed into a set of their ingredients and the distribution of the unique values of the corresponding food groups was carried out. Counting the number of unique preferences (count base diversity) by the number of positions in each of the 6 food groups (cereals, fish and meat, fat products, dairy products, vegetables, fruits and berries) allowed us to estimate the amount of food diversity. The total number of food categories in all subgroups was also calculated. Results. The low value of food diversity in the category of cereals and cereals was found in 2/3 of buyers, in the category of vegetables and fruits in 60% of buyers, in the category of meat and fish products in half of buyers, in the category of fat-and-oil products in 63.7% and in 44.2% of buyers of dairy products. An acceptable total number of categories of purchased products (more than 20 per week) was achieved only in 14.5% of men and 16.7% of women from our sample. In general, lower indicators of dietary diversity are typical for men and young people (up to 45 years old). Conclusion. Thus, it can be concluded that the importance of food diversity as a factor of health protection is underestimated by the population, especially by young people, which can lead to an increase in the burden on healthcare due to the earlier debut of a wide range of chronic diseases.

[Food diversity analysis based on data of food purchasing in supermarket chain]

Diversity in nutrition is of fundamental importance for health. According to research in recent decades, the diversity of the population's diet has greatly decreased, which carries health risks. The aim of the study was to analyze food diversity in population based on their purchasing activity in large trading network. Material and methods. From the depersonalized data of 1 800 319 loyalty program unique members of the retail network in Moscow, 201 904 buyers were selected according to the following criteria: the presence of a dense segment of purchases (more than 4 weeks, where there was at least 1 purchase in 2 weeks), in which the total cost of purchases was not less than 4700 rubles and at least 4 different food groups were purchased. Data from the cashier's receipts for 12 months (median 124 days) as well as the food labels for extracting the data about ingredients were used. Food diversity was analyzed by using count-base score method in which absolute number of different foods in every of 6 food groups (grains, fish and meat, fats, dairy products, vegetables, fruits and berries) was calculated. Total score as a sum of all scores from all food groups was also calculated. Results. According to the food diversity analysis, 73.9% of buyers purchased 2 or less types of grains. Only 31.4% of buyers purchased more than 4 types of vegetables, only 36.2% purchased more than 2 types of fruits and berries, 41.9% purchased less than 2 types of meat and fish, 61.3% purchased only 1 type of fat, and at least 2 types of dairy products was purchased by 53,3% of buyers. Acceptable rate of food diversity of 20 different types of food per week was achieved only in 11.4% of buyers. Conclusion. Food diversity in buyers of trading network is low, with the lowest scores in buying different grains, vegetables, fruits and berries, meat and fish as well as fats. Better diversity was demonstrated in buying dairy products, as they traditionally are accounted as healthy by consumers.

[Comparison of resting state and task-based functional MRI in preoperative mapping in patients with brain gliomas]

OBJECTIVE

To analyze and compare the results of cerebral cortex mapping with task-based (tb-fMRI) and resting-state functional MRI in patients with glioma of eloquent cortical areas.

MATERIAL AND METHODS

There were 55 patients (24 men and 31 women aged 24 - 74 years, median 39) with glial tumors. In 26 patients, the tumor was located in motor areas. Twenty-nine patients had lesions of Broca and Wernicke's areas. All patients underwent preoperative tb-fMRI and rs-fMRI. Then, resection of tumor was carried out in all cases.

RESULTS

Comparison of fMRI and rs-fMRI activation maps was assessed by calculating the Dice coefficient for inclusive speech and motor cortex masks and exclusive masks without brainstem, cerebellum, subcortical nuclei. Inclusive Dice coefficient for motor cortex ranged from 0.11 to 0.50, for speech cortex - from 0.006 to 0.240 (p<0.05). In case of exclusive masks, this value ranged from 0.15 to 0.55 for motor cortex and from 0.004 to 0.205 for speech cortex (p<0.05).

CONCLUSION

When comparing the results of cortical mapping in patients with glial tumors, the use of hemispheric exclusive and inclusive masks did not significantly increase activation maps matching. Probably, low degree of correspondence was associated with different genesis of activations, as well as with high variability of speech cortex.

[Machine learning for resting state fMRI-based preoperative mapping: comparison with task-based fMRI and direct cortical stimulation]

OBJECTIVE

To develop a system for preoperative prediction of individual activations of motor and speech areas in patients with brain gliomas using resting state fMRI (rsfMRI), task-based fMRI (tb-fMRI), direct cortical stimulation and machine learning methods.

MATERIAL AND METHODS

Thirty-three patients with gliomas (19 females and 14 males aged 19 - 540) underwent DCS-assisted resection of tumor (19 ones with lesion of motor zones and 14 patients with lesions of speech areas). Awake craniotomy was performed in 14 cases. Preoperative mapping was performed according to special MRI protocol (T1, tb-fMRI, rs-fMRI).

Machine learning system was built on open source data from The Human Connectome Project. MR data of 200 healthy subjects from this database were used for system pre-training. Further, this system was trained on the data of our patients with gliomas.

RESULTS

In DCS, we obtained 332 stimulations including 173 with positive response. According to comparison of functional activations between rs-fMRI and tb-fMRI, there were more positive DCS responses predicted by rs-fMRI (132 vs 112). Non-response stimulation sites (negative) prevailed in tb-fMRI activations (69 vs 44).

CONCLUSION

The developed method with machine learning based on resting state fMRI showed greater sensitivity compared to classical task-based fMRI after verification with DCS: 0.72 versus 0.66 (p<0.05) for identifying the speech zones and 0.79 versus 0.62 (p<0.05) for motor areas.

[Resting-state fMRI in preoperative non-invasive mapping in patients with left hemisphere glioma]

Maximum resection and preservation of neurological function are main principles in surgery of brain tumors, especially glial neoplasms with diffuse growth. Therefore, exact localizing of eloquent brain areas is an important component in surgical planning ensuring optimal resection with minimal postoperative neurological deficit. Functional MRI is used to localize eloquent brain areas adjacent to the tumor. This paper is an initial stage in analysis of resting-state fMRI in assessment of functional changes of neuronal activity caused by brain gliomas of different localization. We report two patients with glial tumors localized within the precentral gyrus of the left hemisphere and near speech area. Considering data of task-based and resting-state fMRI, as well as direct cortical stimulation, we propose a methodology for assessing the overlap of activations obtained by these methods.

Dynamic Causal Modeling of Hippocampal Links within the Human Default Mode Network: Lateralization and Computational Stability of Effective Connections

The purpose of this paper was to study causal relationships between left and right hippocampal regions (LHIP and RHIP, respectively) within the default mode network (DMN) as represented by its key structures: the medial prefrontal cortex (MPFC), posterior cingulate cortex (PCC), and the inferior parietal cortex of left (LIPC) and right (RIPC) hemispheres. Furthermore, we were interested in testing the stability of the connectivity patterns when adding or deleting regions of interest. The functional magnetic resonance imaging (fMRI) data from a group of 30 healthy right-handed subjects in the resting state were collected and a connectivity analysis was performed. To model the effective connectivity, we used the spectral Dynamic Causal Modeling (DCM). Three DCM analyses were completed. Two of them modeled interaction between five nodes that included four DMN key structures in addition to either LHIP or RHIP. The last DCM analysis modeled interactions between four nodes whereby one of the main DMN structures, PCC, was excluded from the analysis. The results of all DCM analyses indicated a high level of stability in the computational method: those parts of the winning models that included the key DMN structures demonstrated causal relations known from recent research. However, we discovered new results as well. First of all, we found a pronounced asymmetry in LHIP and RHIP connections. LHIP demonstrated a high involvement of DMN activity with preponderant information outflow to all other DMN regions. Causal interactions of LHIP were bidirectional only in the case of LIPC. On the contrary, RHIP was primarily affected by inputs from LIPC, RIPC, and LHIP without influencing these or other DMN key structures. For the first time, an inhibitory link was found from MPFC to LIPC, which may indicate the subjects’ effort to maintain a resting state. Functional connectivity data echoed these results, though they also showed links not reflected in the patterns of effective connectivity. We suggest that such lateralized architecture of hippocampal connections may be related to lateralization phenomena in verbal and spatial domains documented in human neurophysiology, neuropsychology, and neurolinguistics.

[The Role of Backward Connections in Generation of Brain Electrical Responces Elicited by Simple Visual Stimuli]

The role of backward connections (from higher to lower hierarchical layers) in the generation of evoked brain esponses. to simple visual stimuli was explored using Dynamic Causal Modelling approach (DCM), and the contribution of backward connections to evoked responses was quantified as a function of post-stimulus time. The early components of visual evoked related potential (ERP) reflect exogenous (sensory) inputs and can be explained by connections from the lower to higher layers'(forward connections). The ERP components after 200 ms reflect endogenous activity and require the inclusion of both forward and backward connections in the model. Our results complement the work of other authors who showed earlier using DCM that the backward connections occur at the late latencies in auditory oddball experiment.

Effective Connectivity within the Default Mode Network: Dynamic Causal Modeling of Resting-State fMRI Data

The Default Mode Network (DMN) is a brain system that mediates internal modes of cognitive activity, showing higher neural activation when one is at rest. Nowadays, there is a lot of interest in assessing functional interactions between its key regions, but in the majority of studies only association of Blood-oxygen-level dependent (BOLD) activation patterns is measured, so it is impossible to identify causal influences. There are some studies of causal interactions (i.e., effective connectivity), however often with inconsistent results. The aim of the current work is to find a stable pattern of connectivity between four DMN key regions: the medial prefrontal cortex (mPFC), the posterior cingulate cortex (PCC), left and right intraparietal cortex (LIPC and RIPC). For this purpose functional magnetic resonance imaging (fMRI) data from 30 healthy subjects (1000 time points from each one) was acquired and spectral dynamic causal modeling (DCM) on a resting-state fMRI data was performed. The endogenous brain fluctuations were explicitly modeled by Discrete Cosine Set at the low frequency band of 0.0078-0.1 Hz. The best model at the group level is the one where connections from both bilateral IPC to mPFC and PCC are significant and symmetrical in strength (p < 0.05). Connections between mPFC and PCC are bidirectional, significant in the group and weaker than connections originating from bilateral IPC. In general, all connections from LIPC/RIPC to other DMN regions are much stronger. One can assume that these regions have a driving role within the DMN. Our results replicate some data from earlier works on effective connectivity within the DMN as well as provide new insights on internal DMN relationships and brain's functioning at resting state.

[Dynamic causal modeling of brain electrical responses elicited by simple stimuli in visual oddball paradigm]

Dynamic Causal Modeling (DCM) is a technique designed to assess the effective connectivity in the brain, i.e. the influence one neuronal system exerts over another. The central idea behind DCM is to treat the brain as a deterministic nonlinear dynamical system that is subject to inputs, and produces outputs. DCM for EEG uses neural mass model to explain source activity and to build a forward model that predicts scalp-recorded response, based on a particular underlying network structure. Further analysis is done by selecting, using the Bayesian inference, among the competing hypotheses (models) the one that is best to explain the data. We used DCM approach to find a plausible model for ERPs recorded for standard and deviant stimuli in visual oddball task, and to evaluate the reproducibility of this model over a set of individual recordings. The model that best explained the data and gave reproducible results was the one that allowed the changes in strength of forward connections. These results are compatible with the DCM for auditory oddball experiment by other authors.