Lidé na fakultě
Umer Asgher, MSc., Ph.D.

2025, International Journal of Human-Computer Interaction, ISSN 1044-7318

In digitalized plants, control room operators experience cognitive overload, and literature emphasizes that multimodal physiological integration can better capture operators' cognitive states. In chemical process operations, current methods often overlook cross-modal interactions. This study used a formaldehyde production simulation with 42 participants exposed to failure scenarios, assessing performance by recovery time and plant status. A novel framework for multimodal physiological integration is proposed, modeling high/low levels of eye-based, skin-related, and cardiovascular metrics using Gaussian distributions. Unique combinations of these metrics are formed, and the overlapping coefficient (OVL) is computed to identify consistent physiological combinations across participants. High-OVL combinations appeared in all optimal, 79% of good, and were negligible in the poor class. Successful participants exhibited distinct cognitive strategies, from low-arousal focus to high-arousal compensation. The Bayesian network estimated participants' performance-level probabilities, achieving 91% accuracy and robustness to missing data. The framework supports reflective learning, supervisory support, and adaptive training systems.

Článek v periodiku, který teprve vyjde

2025, Neuroscience Bulletin, 41 (1), p. 46-60), ISSN 1995-8218

This study explored how the human cortical folding pattern composed of convex gyri and concave sulci affected single-subject morphological brain networks, which are becoming an important method for studying the human brain connectome. We found that gyri-gyri networks exhibited higher morphological similarity, lower small-world parameters, and lower long-term test-retest reliability than sulci-sulci networks for cortical thickness- and gyrification index-based networks, while opposite patterns were observed for fractal dimension-based networks. Further behavioral association analysis revealed that gyri-gyri networks and connections between gyral and sulcal regions significantly explained inter-individual variance in Cognition and Motor domains for fractal dimension- and sulcal depth-based networks. Finally, the clinical application showed that only sulci-sulci networks exhibited morphological similarity reductions in major depressive disorder for cortical thickness-, fractal dimension-, and gyrification index-based networks. Taken together, these findings provide novel insights into the constraint of the cortical folding pattern to the network organization of the human brain.

Článek v periodiku excerpovaném SCI Expanded

2025, Journal of Materials Chemistry C, 13 (31), p. 15767-15795), ISSN 2050-7526

Recently, two-dimensional (2D) layered semiconductors have been the subject of promising research work due to their intriguing physical and chemical characteristics. In electronic nano-devices, impact ionization is a viable condition to investigate or probe the level of sensitivity upon the application of external stimuli. However, avalanche field-effect transistors (FETs) have emerged as promising candidates for a wide range of sophisticated applications, especially for sensing traits. In this review, we explore the incorporation of 2D materials into avalanche FETs, highlighting their auspicious properties such as high carrier mobility, variable band gaps, and atomic thickness, which provide significant advantages over typical materials. 2D materials significantly improve the sensitivity, speed, and power efficiency of avalanche FETs. This study also encompasses the advances in photo-, bio- and gas-sensing technologies, emphasizing their implications in contemporary applications such as optoelectronics, imaging, and environmental monitoring. Thus, our review provides a thorough investigation of material attributes, device architecture, and prospective applications by establishing avalanche FETs with 2D materials as the keystone in power and rectifying applications. © 2025 The Royal Society of Chemistry.

Článek v periodiku excerpovaném SCI Expanded

2025, BRITISH JOURNAL OF SPORTS MEDICINE, 59 (19), p. 1337-1349), ISSN 0306-3674

To assess the level of agreement among experts on the heel raise exercise parameters that influence midportion and insertional Achilles tendinopathy (AT) rehabilitation outcomes. An international expert panel in AT rehabilitation was invited to complete a three-round Delphi survey. In the first two rounds, experts were asked to review a pregenerated list of exercise parameters (based on the heel raise exercise) and rate their perceived influence on rehabilitation outcome, using a 5-point Likert scale. For each parameter, consensus criteria for major influence on rehabilitation were (a) median≥4, (b) ≥75% of scoring 4 or 5 and (c) IQR≤1. The third round aimed to rank the exercise parameters according to importance (from most to least important) during rehabilitation. 17 experts participated in the entire Delphi process. A total of 16 exercise parameters were assessed, of which 4 (intensity of contraction, total time under tension, number of repetitions and sets, type of contraction) reached consensus as having a major influence on rehabilitation for midportion AT and 3 reached consensus for insertional AT (range of ankle dorsiflexion during the exercise, intensity of contraction, number of repetitions and sets). The rankings of parameters that reached consensus showed that contraction intensity was perceived as the most important variable for midportion AT rehabilitation, while range of ankle dorsiflexion was deemed the most important variable for insertional AT rehabilitation. This study identified key exercise parameters for mid-portion and insertional AT rehabilitation based on expert opinion. This information should assist practitioners in optimising their approach to deliver more effective, patient-specific exercises for AT rehabilitation.

Článek v periodiku excerpovaném SCI Expanded

2025, Diagnostics, 15 (22), ISSN 2075-4418

Abstract: Background: Stress is a critical determinant of mental health, yet conventional monitoring approaches often rely on subjective self-reports or physiological signals that lack real-time responsiveness. Recent advances in large language models (LLMs) offer opportunities for speech-driven, adaptive stress detection, but existing systems are limited to retrospective text analysis, monolingual settings, or detection-only outputs. Methods: We developed a real-time, speech-driven stress detection framework that integrates audio recording, speech-to-text conversion, and linguistic analysis using transformer-based LLMs. The system provides multimodal outputs, delivering recommendations in both text and synthesized speech. Nine LLM variants were evaluated on five benchmark datasets under zero-shot and few-shot learning conditions. Performance was assessed using accuracy, precision, recall, F1-score, and misclassification trends (false-negatives and false-positives). Real-time feasibility was analyzed through latency modeling, and user-centered validation was conducted across cross-domains. Results: Few-shot fine-tuning improved model performance across all datasets, with Large Language Model Meta AI (LLaMA) and Robustly Optimized BERT Pretraining Approach (RoBERTa) achieving the highest F1-scores and reduced false-negatives, particularly for suicide risk detection. Latency analysis revealed a trade-off between responsiveness and accuracy, with delays ranging from ~2 s for smaller models to ~7.6 s for LLaMA-7B on 30 s audio inputs. Multilingual input support and multimodal output enhanced inclusivity. User feedback confirmed strong usability, accessibility, and adoption potential in real-world settings. Conclusions: This study demonstrates that real-time, LLM-powered stress detection is both technically robust and practically feasible. By combining speech-based input, multimodal feedback, and user-centered validation, the framework advances beyond traditional detection only models toward scalable, inclusive, and deployment-ready digital mental health solutions.

Článek v periodiku excerpovaném SCI Expanded

2024, Neural Network World, 34 (3), p. 135-168), ISSN 1210-0552

This study presents a comprehensive multi-objective transportation model aimed at optimizing complex vehicle routing problems, which are nondeterministic polynomial time NP-hard due to spatial, temporal, and capacity constraints. In this study, the multi-objective transportation model integrates decisionmaker preferences with hybrid optimization techniques, including the approximatecombinatorial method, ant colony optimization and evolutionary algorithms. it seeks to minimize transportation costs, time, and emissions while accounting for real-world constraints such as fleet composition, customer demand, and servicelevel agreements. The techniques like multi-criteria decision-making methods are employed to refine the solution set, balancing objectives like cost, time, environmental impact, and service level. The novel optimization model is applied to a fuel distribution case study involving 18 customers and a heterogeneous fleet, where it optimizes vehicle routes to meet delivery requirements efficiently. The multiobjective transportation framework generates multiple feasible solutions, which are further narrowed down using decision-making frameworks to ensure alignment with organizational goals and decision-maker preferences. The integration of quantitative optimization techniques with qualitative decision-making processes makes this model robust and scalable, offering a practical tool for enhancing operational efficiency in transportation systems. This approach effectively addresses real-world logistics challenges, demonstrating significant improvements in route efficiency, cost savings, and environmental sustainability.

Článek v periodiku excerpovaném SCI Expanded

2024, 2024 New Trends in Aviation Development (NTAD), IEEE (Institute of Electrical and Electronics Engineers)), p. 159-164), ISBN 979-8-3315-2774-7, ISSN 2836-2764

Fear of flying, or aviophobia, is a significant issue affecting a substantial portion of the population, with young adults being particularly vulnerable. Despite the high safety record of aviation, this irrational fear can severely impact an individual’s willingness or ability to travel by air, especially in the context of increasing global air traffic. This study investigates the factors contributing to fear of flying among young commuters aged 18-26 using a stated preference survey. The analysis highlights that fear is primarily driven by emotional responses to situational triggers rather than rational assessments of air travel safety. Key predictors of fear include concerns about terrorism, aviation accidents, and adverse weather conditions, with these factors exerting significant emotional influence despite their statistical rarity. The study also reveals a disconnect between objective safety perceptions and subjective fear, indicating that even passengers who acknowledge the general safety of air travel may still experience anxiety triggered by situational factors.

Stať ve sborníku z prestižní konf.

2024, Frontiers in Robotics and AI, 11, ISSN 2296-9144

Colonoscopy is a reliable diagnostic method to detect colorectal polyps early on and prevent colorectal cancer. The current examination techniques face a significant challenge of high missed rates, resulting in numerous undetected polyps and irregularities. Automated and real-time segmentation methods can help endoscopists to segment the shape and location of polyps from colonoscopy images in order to facilitate clinician’s timely diagnosis and interventions. Different parameters like shapes, small sizes of polyps, and their close resemblance to surrounding tissues make this task challenging. Furthermore, high-definition image quality and reliance on the operator make real-time and accurate endoscopic image segmentation more challenging. Deep learning models utilized for segmenting polyps, designed to capture diverse patterns, are becoming progressively complex. This complexity poses challenges for real-time medical operations. In clinical settings, utilizing automated methods requires the development of accurate, lightweight models with minimal latency, ensuring seamless integration with endoscopic hardware devices. To address these challenges, in this study a novel lightweight and more generalized Enhanced Nanonet model, an improved version of Nanonet using NanonetB for real-time and precise colonoscopy image segmentation, is proposed. The proposed model enhances the performance of Nanonet using Nanonet B on the overall prediction scheme by applying data augmentation, Conditional Random Field (CRF), and Test-Time Augmentation (TTA). Six publicly available datasets are utilized to perform thorough evaluations, assess generalizability, and validate the improvements: Kvasir-SEG, Endotect Challenge 2020, Kvasir-instrument, CVC-ClinicDB, CVC-ColonDB, and CVC-300. Through extensive experimentation, using the Kvasir-SEG dataset, our model achieves a mIoU score of 0.8188 and a Dice coefficient of 0.8060 with only 132,049 parameters and employing minimal computational resources. A thorough cross-dataset evaluation was performed to assess the generalization capability of the proposed Enhanced Nanonet model across various publicly available polyp datasets for potential real-world applications. The result of this study shows that using CRF (Conditional Random Fields) and TTA (Test-Time Augmentation) enhances performance within the same dataset and also across diverse datasets with a model size of just 132,049 parameters. Also, the proposed method indicates improved results in detecting smaller and sessile polyps (flats) that are significant contributors to the high miss rates.

Článek v periodiku excerpovaném databází Scopus

2024, Transportation Research Procedia, Amsterdam, Elsevier B.V.), p. 278-284), ISSN 2352-1465

Vertical take-off and landing unmanned aerial vehicles (VTOL UAVs) are becoming increasingly important in the modern aviation industry. With their growing use, it is essential to find solutions to newly emerging technical problems related to ensuring operational safety. One of these unresolved issues is the accumulation of ice on propeller blades during flight in adverse meteorological conditions. For VTOL UAVs, the propeller blades are the only lifting surfaces, and any disruption of their geometry by an ice layer can lead to the crash of the entire aircraft. A promising solution to this problem, currently being explored experimentally, is the use of hydrophobic coatings, which successfully delay or completely prevent the accumulation of ice. However, hydrophobic coatings currently lack sufficient durability and have a low ability to withstand normal operational wear and tear, causing them to lose their hydrophobic properties quickly. The primary goal of this work is to propose a concept and testing methodology that would allow propeller blades treated with hydrophobic coatings to be exposed to conditions as close as possible to operational wear and tear.

Stať ve sborníku z mezinár. konf.

2024, Aerospace, 11 (12), ISSN 2226-4310

The rapid growth in air traffic has led to increasing congestion at airports, creating bottlenecks that disrupt ground operations and compromise the efficiency of air traffic management (ATM). Ensuring the predictability of ground operations is vital for maintaining the sustainability of the ATM sector. Flight efficiency is closely tied to adherence to assigned airport arrival and departure slots, which helps minimize primary delays and prevents cascading reactionary delays. Significant deviations from scheduled arrival times—whether early or late—negatively impact airport operations and air traffic flow, often requiring the imposition of Air Traffic Flow Management (ATFM) regulations to accommodate demand fluctuations. This study leverages a data-driven machine learning approach to enhance the predictability of in-block and landing times. A Bidirectional Long Short-Term Memory (BiLSTM) neural network was trained using a dataset that integrates flight trajectories, meteorological conditions, and airport operations data. The model demonstrated high accuracy in predicting landing time deviations, achieving a Root-Mean-Square Error (RMSE) of 8.71 min and showing consistent performance across various long-haul flight profiles. In contrast, in-block time predictions exhibited greater variability, influenced by limited data on ground-level factors such as taxi-in delays and gate availability. The results highlight the potential of deep learning models to optimize airport resource allocation and improve operational planning. By accurately predicting landing times, this approach supports enhanced runway management and the better alignment of ground handling resources, reducing delays and increasing efficiency in high-traffic airport environments. These findings provide a foundation for developing predictive systems that improve airport operations and air traffic management, with benefits extending to both short- and long-haul flight operations.

Článek v periodiku excerpovaném SCI Expanded

2024, 2024 New Trends in Aviation Development (NTAD), IEEE (Institute of Electrical and Electronics Engineers)), p. 150-158), ISBN 979-8-3315-2774-7, ISSN 2836-2764

Runway slope illusions, a significant challenge in aviation, can distort pilots’ perception during critical phases of flight, particularly approach and landing. This study explores the susceptibility of pilots to these illusions in a controlled simulator environment using virtual reality technology. A systematic review of existing literature was conducted using the PRISMA methodology, revealing a notable gap in experimental studies directly addressing runway slope illusions. Subsequently, an experimental investigation was designed where 15 pilots performed repeated approaches to runways with varying slopes. Key performance metrics such as mean deviation from the ideal glide slope, standard deviation, RMSE, and time spent below the glide slope were analyzed. Although no statistically significant differences were found across the different runway slopes, patterns emerged, highlighting performance variability, especially on upslope runways. The findings underscore the need for further research into the effects of runway slope illusions on pilot performance, particularly under varying flight conditions.

Stať ve sborníku z prestižní konf.

2024, New Trends in Civil Aviation: Proceedings of the 24th International Conference on New Trends in Civil Aviation 2024, Praha, České vysoké učení technické v Praze), p. 53-58), ISBN 978-80-01-07181-6, ISSN 2694-7854

The approach to landing and the landing of an aircraft represent one of the most critical phases of flight. Under certain conditions, a pilot's activity in this phase can be influenced by the visual illusion of the runway aspect ratio. In this context, an experiment was designed to determine the impact of the runway aspect ratio illusion on pilot performance based on its objective evaluation. The experiment was conducted on a flight simulator using a virtual reality headset. Flight data, collected during visual flight rules approaches, were subsequently analyzed. The deviation of the descent angle during visual approaches from the optimal angle for visual aircraft landing was examined. Data analysis using the repeated measures ANOVA confirmed the influence of this illusion on pilots during flight in a simulated environment, as indicated by the root mean square error and maximum absolute error.

Stať ve sborníku z prestižní konf.