In light of the demographic shift towards an aging population, there is increasing prevalence dementia among elderly. The negative impact on mental health preventing individuals from taking proper care themselves. For requiring hospital care, those receiving home or as a precaution for specific individual, it advantageous to utilize monitoring equipment track their biological parameters ongoing basis. This can minimize risk serious accidents severe hazards. objective present research project...

Background: Infant brain tissue segmentation from MRI data is a critical task in medical imaging, particularly challenging due to the evolving nature of contrasts early months life. The difficulty increases as gray matter (GM) and white (WM) intensities converge, making accurate challenging. This study aims develop an improved U-net-based model enhance precision automatic cerebro-spinal fluid (CSF), GM, WM 10 infant MRIs using iSeg-2017 dataset. Methods: proposed method utilizes U-net...

Diabetes mellitus (DM) is a persistent metabolic disorder associated with the hormone insulin. The two main types of DM are type 1 (T1DM) and 2 (T2DM). Physical activity plays crucial role in therapy diabetes, benefiting both patients. detection, recognition, subsequent classification physical based on intensity integral components treatment. continuous glucose monitoring system (CGMS) signal provides blood (BG) level, combination CGMS heart rate (HR) signals potential targets for detecting...

Non-coordinated physical activity may lead to hypoglycemia, which is a dangerous condition for diabetic people. Decision support systems related type 1 diabetes mellitus (T1DM) still lack the capability of automated therapy modification by recognizing and categorizing activity. Further, this desired adaptive should be achieved without increasing administrative load, already high community. These requirements can satisfied using artificial intelligence-based solutions, signals collected...

In this paper, we compare the performance of user verification systems based on three publicly available mouse dynamics data sets. One these datasets is our new DFL set which contains 21 different users. Two aspects are investigated: effect quantity training performance, and number consecutive actions used for identity predictions. Measurements show that it advisable to use approximately 1000 (60 minutes active movements average), at least 10 prediction.

In recent times, the prevalence of chatbot technology has notably increased, particularly in realm medical assistants. However, there is a noticeable absence chatbots that cater to Hungarian language. Consequently, Hungarian-speaking people currently lack access an automated system capable providing assistance with their health-related inquiries or issues. Our research aims establish competent assistant accessible through both website and mobile app. It crucial highlight project’s objective...

Brain tumors are highly complex, making their detection and classification a significant challenge in modern medical diagnostics. The accurate segmentation of brain from MRI images crucial for effective treatment planning. This study aims to develop an advanced neural network architecture that addresses these challenges.

We present a deep learning model for finding human-understandable connections between input features. Our approach uses parameterized, differentiable activation function, based on the theoretical background of nilpotent fuzzy logic and multi-criteria decision-making (MCDM). The learnable parameter has semantic meaning indicating level compensation neural network determines parameters using gradient descent to find relationships demonstrate utility effectiveness by successfully applying it...

The main drawback of magnetic resonance imaging (MRI) represents the lack a standard intensity scale. All observed numerical values are relative and can only be interpreted together with their context. Before feeding MRI data volumes to supervised learning segmentation procedures, histograms need registered each other, or in other words, they so-called normalization. most popular histogram normalization technique used assist brain is algorithm proposed by Nyuĺ et al 2000, which aligns batch...

In the present study, we investigated effect of different reward functions in insulin regulation using reinforcement learning. An artificial pancreas system is able to deliver into body an automated way. The control algorithm delivery a key player achieving personalized therapy. Neural networks provide approach customize administration by learning patient's habits and administering accordingly. Therefore, conducted experiments with neural based on Our goal was find network-based model...

In the case of diabetes mellitus physical activity does have a high effect on glycemic state patients. This is especially regarding patients with Type 1 mellitus, who need external insulin administration in their daily life. Nevertheless, - as one source stress underrepresented decisions and medical staff available automated glucose regulatory devices. The goal study was to build up simulation framework for data generation assess which machine learning solution can be most accurate...

One of the most challenging area diabetes research is to provide such automated insulin delivery systems – so called artificial pancreas that have robust and adaptive capabilities in a highly sophysticated way. I.e. they are able actions at beginning therapy satisfy requirements patients without knowing users daily lifestyle preferences however on short-term learn these patient specifics increase quality therapy. possible solution closed-loop self-learning features. In present study, we...

Real-world data has a major importance in diabetes related research, especially considering the widespread applications of machine learning algorithms. There are several existing datasets real-world literature; however, they all have their specific formats, applied devices and file structures. The different charachteristics make it cumbersome to use multiple for research purposes. We developed pipeline efficiently storing accessing standardized way. defined JSON format records stored them...

The accelerating rise of the amount daily produced medical image data and high costs human expert training are strongly motivating efforts dedicated to development reliable efficient segmentation interpretation methodology. This paper proposes a fully automatic method provide brain tissues from volumetric multi-spectral magnetic resonance data. proposed provides U-Net neural network based decision making, that is trained with infant MRI records originating iSeg-2017 challenge, which were...

Background/Objectives: Managing blood glucose levels effectively remains a significant challenge for individuals with diabetes. Traditional methods often lack the flexibility needed personalized care. This study explores potential of reinforcement learning-based approaches, which mimic human learning and adapt strategies through ongoing interactions, in creating dynamic management plans. Methods: We developed mathematical model specifically patients type IVP diabetes, validated data from 10...