Attila Biró

Dr. Attila Biró

Physiological Controls Research Center, Óbuda University, Budapest, Hungary
+4
orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 December 2023 in Sensors
Verified
Journal Article
Open access
PDF file
Attila BiróAttila Biró
Attila Biró
Antonio Cuesta-VargasAntonio Cuesta-Vargas
Antonio Cuesta-Vargas
László SzilágyiLászló Szilágyi
László Szilágyi
Background: Optimal sports performance requires a balance between intensive training and adequate rest. IMUs provide objective, quantifiable data to analyze performance dynamics, despite the challenges in quantifying athlete training loads. The ability of AI to analyze complex datasets brings innovation to the monitoring and optimization of athlete training cycles. Traditional techniques rely on subjective assessments to prevent overtraining, which can lead to injury and underperformance. IMUs provide objective, quantitative data on athletes’ physical status during action. AI and machine learning can turn these data into useful insights, enabling data-driven athlete performance management. With IMU-generated multivariate time series data, this paper uses AI to construct a robust model for predicting fatigue and stamina. Materials and Methods: IMUs linked to 19 athletes recorded triaxial acceleration, angular velocity, and magnetic orientation throughout repeated sessions. Standardized training included steady-pace runs and fatigue-inducing techniques. The raw time series data were used to train a supervised ML model based on frequency and time-domain characteristics. The performances of Random Forest, Gradient Boosting Machines, and LSTM networks were compared. A feedback loop adjusted the model in real time based on prediction error and bias estimation. Results: The AI model demonstrated high predictive accuracy for fatigue, showing significant correlations between predicted fatigue levels and observed declines in performance. Stamina predictions enabled individualized training adjustments that were in sync with athletes’ physiological thresholds. Bias correction mechanisms proved effective in minimizing systematic prediction errors. Moreover, real-time adaptations of the model led to enhanced training periodization strategies, reducing the risk of overtraining and improving overall athletic performance. Conclusions: In sports performance analytics, the AI-assisted model using IMU multivariate time series data is effective. Training can be tailored and constantly altered because the model accurately predicts fatigue and stamina. AI models can effectively forecast the beginning of weariness before any physical symptoms appear. This allows for timely interventions to prevent overtraining and potential accidents. The model shows an exceptional ability to customize training programs according to the physiological reactions of each athlete and enhance the overall training effectiveness. In addition, the study demonstrated the model’s efficacy in real-time monitoring performance, improving the decision-making abilities of both coaches and athletes. The approach enables ongoing and thorough data analysis, supporting strategic planning for training and competition, resulting in optimized performance outcomes. These findings highlight the revolutionary capability of AI in sports science, offering a future where data-driven methods greatly enhance athlete training and performance management.

1 December 2021 in JMIR mHealth and uHealth
Verified
Journal Article
Open access
AMAntonio Muro-Culebras
Antonio Muro-Culebras
Adrian Escriche-EscuderAdrian Escriche-Escuder
Adrian Escriche-Escuder
JMJaime Martin-Martin
Jaime Martin-Martin
CRCristina Roldán-Jiménez
Cristina Roldán-Jiménez
Irene De-TorresIrene De-Torres
Irene De-Torres
Antonio Cuesta-VargasAntonio Cuesta-Vargas
Antonio Cuesta-Vargas
Background: There are several mobile health (mHealth) apps in mobile app stores. These apps enter the business-to-customer market with limited controls. Both, apps that users use autonomously and those designed to be recommended by practitioners require an end-user validation to minimize the risk of using apps that are ineffective or harmful. Prior studies have reviewed the most relevant aspects in a tool designed for assessing mHealth app quality, and different options have been developed for this purpose. However, the psychometric properties of the mHealth quality measurement tools, that is, the validity and reliability of the tools for their purpose, also need to be studied. The Consensus-based Standards for the Selection of Health Measurement Instruments (COSMIN) initiative has developed tools for selecting the most suitable measurement instrument for health outcomes, and one of the main fields of study was their psychometric properties. Objective: This study aims to address and psychometrically analyze, following the COSMIN guideline, the quality of the tools that are used to measure the quality of mHealth apps. Methods: From February 1, 2019, to December 31, 2019, 2 reviewers searched PubMed and Embase databases, identifying mHealth app quality measurement tools and all the validation studies associated with each of them. For inclusion, the studies had to be meant to validate a tool designed to assess mHealth apps. Studies that used these tools for the assessment of mHealth apps but did not include any psychometric validation were excluded. The measurement tools were analyzed according to the 10 psychometric properties described in the COSMIN guideline. The dimensions and items analyzed in each tool were also analyzed. Results: The initial search showed 3372 articles. Only 10 finally met the inclusion criteria and were chosen for analysis in this review, analyzing 8 measurement tools. Of these tools, 4 validated ≥5 psychometric properties defined in the COSMIN guideline. Although some of the tools only measure the usability dimension, other tools provide information such as engagement, esthetics, or functionality. Furthermore, 2 measurement tools, Mobile App Rating Scale and mHealth Apps Usability Questionnaire, have a user version, as well as a professional version. Conclusions: The Health Information Technology Usability Evaluation Scale and the Measurement Scales for Perceived Usefulness and Perceived Ease of Use were the most validated tools, but they were very focused on usability. The Mobile App Rating Scale showed a moderate number of validated psychometric properties, measures a significant number of quality dimensions, and has been validated in a large number of mHealth apps, and its use is widespread. It is suggested that the continuation of the validation of this tool in other psychometric properties could provide an appropriate option for evaluating the quality of mHealth apps.

Adrian Escriche-EscuderAdrian Escriche-Escuder
Adrian Escriche-Escuder
Irene De-TorresIrene De-Torres
Irene De-Torres
CRCristina Roldán-Jiménez
Cristina Roldán-Jiménez
JMJaime Martin-Martin
Jaime Martin-Martin
AMAntonio Muro-Culebras
Antonio Muro-Culebras
Antonio Cuesta-VargasAntonio Cuesta-Vargas
Antonio Cuesta-Vargas
Digital health interventions may improve different behaviours. However, the rapid proliferation of technological solutions often does not allow for a correct assessment of the quality of the tools. This study aims to review and assess the quality of the available mobile applications (apps) related to interventions for low back pain. Two reviewers search the official stores of Android (Play Store) and iOS (App Store) for localisation in Spain and the United Kingdom, in September 2019, searching for apps related to interventions for low back pain. Seventeen apps finally are included. The quality of the apps is measured using the Mobile App Rating Scale (MARS). The scores of each section and the final score of the apps are retrieved and the mean and standard deviation obtained. The average quality ranges between 2.83 and 4.57 (mean 3.82) on a scale from 1 (inadequate) to 5 (excellent). The best scores are found in functionality (4.7), followed by aesthetic content (mean 4.1). Information (2.93) and engagement (3.58) are the worst rated items. Apps generally have good overall quality, especially in terms of functionality and aesthetics. Engagement and information should be improved in most of the apps. Moreover, scientific evidence is necessary to support the use of applied health tools.

8 July 2021 in Frontiers in Digital Health
Verified
Journal Article
Open access
JMJaime Martín-Martín
Jaime Martín-Martín
CRCristina Roldán-Jiménez
Cristina Roldán-Jiménez
IDIrene De-Torres
Irene De-Torres
AMAntonio Muro-Culebras
Antonio Muro-Culebras
AEAdrian Escriche-Escuder
Adrian Escriche-Escuder
Antonio Cuesta-VargasAntonio Cuesta-Vargas
Antonio Cuesta-Vargas
Background: Sedentary behavior (SB) negatively impact health and is highly prevalent in the population. Digital behavior change interventions (DBCIs) have been developed to modify behaviors such as SB by technologies. However, it is unknown which behavior change techniques (BCTs) are most frequently employed in SB as well as the effect associated with DBCIs in this field. The aim of this systematic review was: (a) to evaluate the BCT most frequently employed in digital health including all technologies available and interventions aimed at increasing physical activity (PA), reducing sedentary time, and improving adherence to exercise in the clinical population, and (b) to review the effect associated with DBCIs in this field.Methods: The database used was Medline, as well as Scopus, Scielo, and Google Scholar. For the search strategy, we considered versions of behavior/behavioral, mHealth/eHealth/telemedicine/serious game/gamification. The terms related to PA and SB were included, the criteria for inclusion were randomized clinical trials (RCTs), adults, intervention based on digital media, and outcome variable lifestyle modification; a last 5 years filter was included. Michie's Taxonomy was used to identify BCTs. The study was registered under the number PROSPERO CRD42019138681.Results: Eighteen RCTs were included in the present systematic review, 5 of them healthy adults, and 13 of them with some illness. Studies included 2298 sedentary individuals who were followed up for 5 weeks−3 years. The most used BCTs were goal setting, problem solving, review outcomes/goals, feedback on behavior and outcomes of behavior, self-monitoring of behavior, social support, information about health consequences, and behavior practice/rehearsal. The effect associated with DBCIs showed improvements, among several related to PA and physiologic self-reported and anthropometric outcomes.Conclusion: The BCTs most used in digital health to change outcomes related to SB were goals and planning, feedback and monitoring, social support, natural consequences, repetition, and substitution. Besides these findings, DBCIs are influenced by several factors like the type of intervention, patients' preferences and values, or the number of BCTs employed. More research is needed to determine with precision which DBCIs or BCTs are the most effective to reduce SB in the clinical population.

30 March 2023 in Sensors
Verified
Journal Article
Open access
PDF file
Attila BiróAttila Biró
Attila Biró
SSSándor Miklós Szilágyi
Sándor Miklós Szilágyi
László SzilágyiLászló Szilágyi
László Szilágyi
JMJaime Martin-Martin
Jaime Martin-Martin
Antonio Cuesta-VargasAntonio Cuesta-Vargas
Antonio Cuesta-Vargas
Background: One of the most critical topics in sports safety today is the reduction in injury risks through controlled fatigue using non-invasive athlete monitoring. Due to the risk of injuries, it is prohibited to use accelerometer-based smart trackers, activity measurement bracelets, and smart watches for recording health parameters during performance sports activities. This study analyzes the synergy feasibility of medical radar sensors and tri-axial acceleration sensor data to predict physical activity key performance indexes in performance sports by using machine learning (ML). The novelty of this method is that it uses a 24 GHz Doppler radar sensor to detect vital signs such as the heartbeat and breathing without touching the person and to predict the intensity of physical activity, combined with the acceleration data from 3D accelerometers. Methods: This study is based on the data collected from professional athletes and freely available datasets created for research purposes. A combination of sensor data management was used: a medical radar sensor with no-contact remote sensing to measure the heart rate (HR) and 3D acceleration to measure the velocity of the activity. Various advanced ML methods and models were employed on the top of sensors to analyze the vital parameters and predict the health activity key performance indexes. three-axial acceleration, heart rate data, age, as well as activity level variances. Results: The ML models recognized the physical activity intensity and estimated the energy expenditure on a realistic level. Leave-one-out (LOO) cross-validation (CV), as well as out-of-sample testing (OST) methods, have been used to evaluate the level of accuracy in activity intensity prediction. The energy expenditure prediction with three-axial accelerometer sensors by using linear regression provided 97–99% accuracy on selected sports (cycling, running, and soccer). The ML-based RPE results using medical radar sensors on a time-series heart rate (HR) dataset varied between 90 and 96% accuracy. The expected level of accuracy was examined with different models. The average accuracy for all the models (RPE and METs) and setups was higher than 90%. Conclusions: The ML models that classify the rating of the perceived exertion and the metabolic equivalent of tasks perform consistently.

1 December 2023 in Acta Universitatis Sapientiae, Informatica
Verified
Journal Article
Open access
PDF file
Attila BiróAttila Biró
Attila Biró
Antonio Cuesta-VargasAntonio Cuesta-Vargas
Antonio Cuesta-Vargas
László SzilágyiLászló Szilágyi
László Szilágyi
The paper examines the potential of artificial intelligence (AI) in parsing text and conducting sentiment analysis to identify early markers of mental health and neurodegenerative disorders. Through the analysis of textual data, we investigate whether AI can provide a noninvasive, continuous, and objective complement to traditional diagnostic practices. Background: the early detection of mental health (such as depression, anxiety, psychotic disorders, Alzheimer’s disease and dementia) and neurodegenerative disorders (like Parkinson’s disease) remains a critical challenge in clinical practice. Traditional diagnostic methods rely on clinical evaluations that may be subjective and episodic. Recent advancements in AI and natural language processing (NLP) have opened new avenues for precognitive health assessments, suggesting that variations in language and expressed sentiments in written text can serve as potential biomarkers for these conditions. Materials and Methods: the research used a dataset comprising various forms of textual data, including anonymized social media interactions, transcripts from patient interviews, and electronic health records. NLP algorithms were deployed to parse the text, and machine learning models were trained to identify language patterns and sentiment changes. The study also incorporated a sentiment analysis to gauge emotional expression, a key component of mental health diagnostics. Results: the AI models were able to identify language use patterns and sentiment shifts that correlated with clinically validated instances of mental health symptoms and neurodegenerative conditions. Notably, the models detected an increased use of negative a ect words, a higher frequency of first-person singular pronouns, and a decrease in future tense in individuals with depression. For neurode-generative conditions, there was a notable decline in language complexity and semantic coherence over time. Conclusions: the implemented pipeline of AI-parsed text and sentiment analysis appears to be a promising tool for the early detection and ongoing monitoring of mental health and neurodegenerative disorders. However, these methods are supplementary and cannot replace the nuanced clinical evaluation process. Future research must refine the AI algorithms to account for linguistic diversity and context, while also addressing ethical considerations regarding data use and privacy. The integration of AI tools in clinical settings necessitates a multidisciplinary approach, ensuring that technological advancements align with patient-centered care and ethical standards.