Simulating federated learning for steatosis detection using ultrasound images.

We aimed to implement four data partitioning strategies evaluated with four federated learning (FL) algorithms and investigate the impact of data distribution on FL model performance in detecting steatosis using B-mode US images. A private dataset (153 patients; 1530 images) and a public dataset (55 patient; 550 images) were included in this retrospective study. The datasets contained patients with metabolic dysfunction-associated fatty liver disease (MAFLD) with biopsy-proven steatosis grades and control individuals without steatosis.

Comparison of Radiologists and Deep Learning for US Grading of Hepatic Steatosis.

Background Screening for nonalcoholic fatty liver disease (NAFLD) is suboptimal due to the subjective interpretation of US images. Purpose To evaluate the agreement and diagnostic performance of radiologists and a deep learning model in grading hepatic steatosis in NAFLD at US, with biopsy as the reference standard. Materials and Methods This retrospective study included patients with NAFLD and control patients without hepatic steatosis who underwent abdominal US and contemporaneous liver biopsy from September 2010 to October 2019.

Identification of growing tuberculosis incidence clusters in a region with a decrease in tuberculosis prevalence in Moscow (2000-2019).

Background: The control of tuberculosis (TB) may benefit from a prospective identification of areas where the incidence may increase in addition to the traditionally identified foci of high incidence. We aimed to identify residential areas with growing tuberculosis incidence rates and assess their significance and stability.

Methods: We analysed the changes in TB incidence rates using case data georeferenced with spatial granularity to apartment buildings in the territory of Moscow from 2000 to 2019.

Mathematical modelling of the progression of active tuberculosis: Insights from fluorography data.

Little is known about the dynamics of the early stages of untreated active pulmonary tuberculosis: unknown are both the rates of progression and the model "scheme". The "parallel" scheme assumes that infectiousness of tuberculosis cases is effectively predefined at the onset of the disease, and the "serial" scheme considers all cases to be non-infectious at the onset, with some of them later becoming infectious. Our aim was to estimate the progression of the early stages of pulmonary tuberculosis using data from a present-day population.

[ORGANIZATION OF EPIDEMIOLOGICAL MONITORING OF TUBERCULOSIS IN THE CITY OF MOSCOW].

Ensuring effective epidemiological surveillance of tuberculosis is essential and one of the paramount tasks for any national tuberculosis control program. The article analyzes the organization of the tuberculosis epidemiological monitoring system at the level of the subject, based on the experience of implementing such a system in the city of Moscow in 1996-2020, defines its tasks and principles of its construction. The systems of epidemiological monitoring of tuberculosis, implemented taking into account the developed principles, provide an analysis of the epidemiological situation in the territory, the results of which can be successfully used for making managerial decisions, program-target planning and evaluating the effectiveness of the measures taken.