Group website: https://www.utko.fekt.vut.cz/en/analysis-and-processing-audio-signals

We are engaged in research of methods of reconstruction of damaged sound signals and archival sound records, use of deep neural networks for obtaining information from sound signals, research of methods of analysis of sound signals and their use in acoustic, electroacoustic and noise measurements, modeling of electroacoustic and audio systems and research of analysis and synthesis of sound for virtual reality.

Main research activities

• reconstruction of damaged audio signals
• reconstruction of archive audio signals
• compression of audio signals
• obtaining information from music signals
• measurement, analysis and identification of noise sources
• measurements in acoustics and electroacoustics
• analysis and modeling of electroacoustic and audio systems
• sound for virtual and augmented reality

Main research results

• MOKRÝ, O.; RAJMIC, P. Approximal operator with application to audio inpainting. SIGNAL PROCESSING, 2020, vol. 179, no. 1, p. 1-8. ISSN: 0165-1684.
• MOKRÝ, O.; RAJMIC, P. Audio Inpainting: Revisited and Reweighted. IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH AND LANGUAGE PROCESSING, 2020, no. 28, p. 2906-2918. ISSN: 2329-9290.
• ZÁVIŠKA, P.; RAJMIC, P.; OZEROV, A.; RENCKER, L. A Survey and an Extensive Evaluation of Popular Audio Declipping Methods. IEEE Journal of Selected Topics in Signal Processing, 2020, vol. 15, no. 1, p. 5-24. ISSN: 1941-0484.
• ŠTILLOVÁ, K.; KISKA, T.; KORIŤÁKOVÁ, E.; STRÝČEK, O.; MEKYSKA, J.; CHRASTINA, J.; REKTOR, I. Mozart effect in epilepsy: Why is Mozart better than Haydn? Acoustic qualities-based analysis of SEEG. European Journal of Neurology, 2021, vol. 1, no. 1, p. 1-17. ISSN: 1351-5101.
• ZÁVIŠKA, P.; RAJMIC, P.; MOKRÝ, O. Audio declipping performance enhancement via crossfading. SIGNAL PROCESSING, 2021, vol. 192, no. 1, p. 1-5. ISSN: 0165-1684.
• MIKLÁNEK, Š.; IŠTVÁNEK, M.; SMÉKAL, Z.: The MemoVision software. URL: https://github.com/stepanmk/memovision.

Group website: https://www.utko.fekt.vut.cz/en/artificial-intelligence-data-processing-and-analysis

We are researching advanced artificial intelligence methods for big data analysis and acceleration of computations on our high-performance supercomputers. Our technologies are able to offer solutions to a range of problems from advanced visual quality control of products, through security and guarding of premises, predictive maintenance of various types of equipment, including power grids, and prediction of future behavior, including automation of complex processes control.

Main research activities

• Machine learning (research of new algorithms for machine learning and artificial intelligence)
• Big data (parallel and distributed algorithms for processing of large volumes of data)
• Computer vision (image content analysis, object detection and classification, object tracing, segmentation)
• Biomedicine (applications for e-health and telemedicine)
• Biometrics and anonymization (identification and verification of persons based on face recognition, advanced anonymization of faces, extraction of secondary biometric and non-biometric features - age, gender, etc.)
• 3D audio

Main research results

• BREGER, A.; ORLANDO, J.; HARÁR, P.; DÖRFLER, M.; KLIMSCHA, S.; GRECHENIG, C.; GERENDAS, B.; SCHMIDT-ERFURTH, U.; EHLER, M. On Orthogonal Projections for Dimension Reduction and Applications in Augmented Target Loss Functions for Learning Problems. Journal of Mathematical Imaging and Vision, 2020, vol. 62, no. 3, p. 376-394. ISSN: 1573-7683. (https://link.springer.com/article/10.1007/s10851-019-00902-2)
• KHAN, J.; KAUSHIK, M.; CHAURASIA, A.; DUTTA, M.; BURGET, R. Cardi-Net: A Deep Neural Network for classification of Cardiac disease using Phonocardiogram Signal. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2020, vol. 219, no. 1, p. 1-11. ISSN: 0169-2607. (https://dl.acm.org/doi/10.1016/j.cmpb.2022.106727)
• Baghela, N;Dutta, M. K.;Burget, R. Automatic diagnosis of multiple cardiac diseases from PCG signals using convolutional neural network. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2020, vol. 197, no. 12, p. 1-11. ISSN: 0169-2607. (https://www.sciencedirect.com/science/article/pii/S0169260720315832)
• GÓRRIZ, J.M.; MEKYSKA, J.; et al. Computational approaches to Explainable Artificial Intelligence: Advances in theory, applications and trends. Information Fusion, 2023, vol. 100, no. December 2023, p. 1-37. ISSN: 1872-6305. (https://www.sciencedirect.com/science/article/pii/S1566253523002610)
• JONÁK, M.; MUCHA, J.; JEŽEK, Š.; KOVÁČ, D.; CZÍRIA, K. SPAGRI-AI: Smart precision agriculture dataset of aerial images at different heights for crop and weed detection using super-resolution. AGRICULTURAL SYSTEMS, 2024, vol. 216, no. April 2024, p. 1-11. ISSN: 0308-521X. (https://www.sciencedirect.com/science/article/pii/S0308521X2400026X)
• JEŽEK, Š.; BURGET, R.; JONÁK, M.; KOLAŘÍK, M.; HAVLÍČEK, L.; SKOTÁK, M.: Defect Detection System; Painted metal parts defectoscopy. (used in Konica Minolta systems - https://www.vut.cz/en/rad/results/detail/187567#vysledek-187567)
• MYŠKA, V.; BURGET, R.: ČEPS ARTIC; ČEPS ARTIC, ČEPS ARTIC, software for energy grid monitoring URL: https://cz.energyhub.eu/en.

Leader: prof. Ing. Ivo Provazník, Ph.D.

Group website: https://www.ubmi.fekt.vut.cz/skupina-mapovani-farmakoforu-virtualni-screening

We develop new methods for the numerical processing of large-scale genomic data, emphasizing de novo genome assembly, genotyping of novel bacterial strains, and metataxonomic and metagenomic studies. The group also investigates and structures new medicals via in silico modeling of molecular interactions and by using virtual screening instruments.

Major 4-year outputs:

• Software tools for the numerical processing of large-scale genomic data.
• Software tools to align differently long genomic signals (implementation in Oxford Nanopore sequencers).
• Software tools for the lossless decimation of large-scale genomic signals.
• Software tools to process and analyze genomic, transcriptomic, and epigenetic data.
• Software tools to digitize and process images obtained through the gel electrophoresis of DNA molecules.

Leader: Assoc. Prof. Jana Kolářová

Group website: https://www.ubmi.fekt.vut.cz/skupina-zpracovani-analyzy-zaznamu-ekg

We create smart technologies to facilitate the monitoring and earlier diagnosis of patients in cardiology, neurology, and other branches of clinical medicine. Our aim is to connect the knowledge acquired through experimental research and clinical practice with recent findings in biological signal processing, smart sensors, and artificial intelligence to form fuctional clusters yielding faster and more effective medical care.

Major 4-year outputs:

• Software tools for quality estimation and adaptive filtering in biomedical signals.
• Software tools to enable, by utilizing machine learning methods, automated classification of atrial fibrillation and other cardiac rhythm disorders.
• Software tools allowing automated measurement and analysis of clinical and experimental ECG signals.
• Patent: A device for the optical sensing of electrical activity in a live tissue.

Leader: Vratislav Čmiel, Ph.D.

Group website: https://www.ubmi.fekt.vut.cz/skupina-experimentalnich-mikroskopickych-technik-pro-bunecne-inzenyrstvi

Within molecular biology, we specialize in developing and implementing new microscopic methods to investigate the electrical and structural properties of cells. Our experts perform multiple diverse experiments ranging from the measurement of cell metabolism and metabolic or substance toxicity to forming an artificial vein based on live cells.

Major 4-year outputs:

• A miniature optical detector of albumin for smartphones.
• Patent: An infrared spectroscopy system.
• A methodology for volume imaging and multimodal volume data analyses in cell engineering, supported by advanced image processing algorithms and a scientific confocal microscope allowing spectral measurement.
• A methodology for designing and executing experiments in cell and tissue electrophysiology, using patch clamping devices and microelectrode fields combined with opthogenetic manipulation including visualization, analysis, and statistical data processing.

Leader: Ing. Zdeněk Roubal, Ph.D.

The research group of the laboratory has been involved in monitoring environmental variables for a long period. Especially in caves, to monitor parameters important for speleotherapy. In addition to this main focus, questions of human influence on the cave environment and the examination of phenomena between the earth’s surface and the underground spaces are also studied. This is also very topical in terms of groundwater quality. Researchers of the group can design and built a necessary unique custom instrumentation for above mentioned purposes.

Main research activities:

• Development of a methodology for the measurement of light air ions
• Correlation between aerosols and light air ions
• Monitoring cave parameters that affect the success of speleotherapy

• SZABÓ, Z.; ROUBAL, Z.; KADLEC, R.; PRACNÝ, P.; LANG, M.: Cave drip meter with data logger. Amateur cave. (functional sample)
• ROUBAL, Z.; SZABÓ, Z.; KADLEC, R.: Soil surface CO2 meter Amateur cave. (functional sample)
• TJ04000064, Evaluation of the influence of surface conditions on the cave environment, https://starfos.tacr.cz/en/projekty/TJ04000064

Main research results – publications

• FAIMON, J.; BALDÍK, V.; ŠTELCL, J.; VŠIANSKÝ D.; REZ J.; PRACNÝ, P.; NOVOTNÝ, R.; LANG, M.; ROUBAL, Z.; SZABÓ Z.; HADACZ, R. Corrosion of calcite speleothems in epigenic caves of Moravian Karst (Czech Republic). Environ Earth Sci, 2024, vol. 83, no. 184. https://doi.org/10.1007/s12665-024-11449-w
• ROUBAL, Z.; BARTUŠEK, K.; SZABÓ, Z.; DREXLER, P.; ÜBERHUBEROVÁ, J. Measuring Light Air Ions in a Speleotherapeutic Cave. Measurement Science Review, 2017, vol. 17, no. 1, p. 27-36. ISSN: 1335-8871. https://doi.org/10.1515/msr-2017-0004
• ROUBAL, Z.; GESCHEIDTOVÁ, E.; BARTUŠEK, K.; SZABÓ, Z.; STEINBAUER, M.; ÜBERHUBEROVÁ, J.; LAJČÍKOVÁ, A. Evaluating the Parameters of a Systematic Long-Term Measurement of the Concentration and Mobility of Air Ions in the Environment inside Cisarska Cave. Atmosphere, 2021, vol. 12, no. 12, pp. 1-31. ISSN: 2073-4433. https://doi.org/10.3390/atmos12121615

Leader: Assoc. Prof. Radim Kolář

Group website: https://www.ubmi.fekt.vut.cz/skupina-zpracovani-obrazu-v-mikroskopii

Within medical image processing, we concentrate on the following tasks and domains: the development and application of software instruments used for the CT, PET, and MRI imaging modalities; monitoring and classification of objects in a videosequence; diagnosing retinal diseases; and monitoring of vital functions with various types of video camera. Our aim is to develop techniques leading towards easier and more effective use of information technologies within the identification and diagnosis of diverse diseases.

Major 4-year outputs:

• Patent: An opthalmological, retina and iris sensing device with an expert eye diagnosis system.
• Utility model: A device for the sensing and recognition of the iris and retina.
• Software tools to support glaucoma diagnosis via advanced analysis of retina images.
• Software tools for merging 2D image data obtained from a fundus camera.
• Software tools measuring bone density in spinal CT images.
• Software tools to identify pathological changes in the human spine.
• Software tools for automated analysis of CT images of the brain.
• Software tools to perform automated analysis of image data acquired via 3D subtraction angiography of the lower extremities.

Leader: Assoc. Prof. Jaromír Hubálek

Group website: http://www.umel.feec.vutbr.cz/labsensnano/

The laboratory experts examine the applicability of nanotechnologies in both general sensorics and the development of sensors of various quantities. The corresponding work involves the designing of not only microsensors to detect gases via nanoparticles but also nanostructured and functionalized electrodes to be employed in electrochemical sensors and biosensors. Other investigated subjects prominently include advanced techniques for the diagnostics and subsequent analysis of substances that find use in medicine.

Major 4-year outputs:

• A technique to allow the processing of a signal from a bolometer (bolometer array) and an electronic system to perform the task: the relevant patent has been put into practice abroad.
• A miniature bolometer membrane with increased absorption and a procedure to form a bolometer absorption layer: the relevant patent has been put into practice abroad.
• A 100-pixel MEMS with bolometers.

Group website: https://www.utko.fekt.vut.cz/en/processing-biomedical-signals

We are researching new approaches to advanced analysis of a wide range of diseases and disorders, such as Parkinson's disease, Lewy body dementia, epilepsy, developmental dysgraphia, oncological diseases, cardiovascular diseases, etc. This research combines methods of signal processing (image, speech, accelerometric, etc.), machine learning, statistics and the Internet of Things.

Main research activities

• Medical image processing
• Reconstruction of magnetic resonance imaging (MRI) images and sequences, perfusion analysis of oncological diseases
• Processing of speech, writing, and other time series, including prediction
• Segmentation and measurement of geometric parameters in image sequences
• Quantitative analysis in pathology
• Supportive diagnosis and monitoring of disorders/diseases
• Data science
• Health 4.0

Main research results

• PULIDO, M.L.B; HERNANDEZ, J.B.A.; BALLESTER, M.A.F.; GONZALEZ, C.M.T.; MEKYSKA, J.; SMÉKAL, Z. Alzheimer's disease and automatic speech analysis: A review. EXPERT SYSTEMS WITH APPLICATIONS, 2020, vol. 150, no. 1, p. 1-19. ISSN: 0957-4174. (https://www.sciencedirect.com/science/article/pii/S0957417420300397)
• KLOBUŠIAKOVÁ, P.; MEKYSKA, J.; BRABENEC, L.; GALÁŽ, Z.; ZVONČÁK, V.; MUCHA, J.; RAPCSAK, S.; REKTOROVÁ, I. Articulatory network reorganization in Parkinson's disease as assessed by multimodal MRI and acoustic measures. PARKINSONISM & RELATED DISORDERS, 2021, vol. 84, no. 1, p. 122-128. ISSN: 1353-8020. (https://www.sciencedirect.com/science/article/pii/S1353802021000535)
• BRABENEC, L.; KLOBUŠIAKOVÁ, P.; MEKYSKA, J.; REKTOROVÁ, I. Shannon entropy: A novel parameter for quantifying pentagon copying performance in non-demented Parkinson's disease patients. PARKINSONISM & RELATED DISORDERS, 2022, vol. 94, no. 1, p. 45-48. ISSN: 1353-8020. (https://www.sciencedirect.com/science/article/pii/S1353802021004442)
• MEZINA, A.; GENZOR, S.; BURGET, R.; MYŠKA, V.; MIZERA, J.; OMETOV, A. Corticosteroid Treatment Prediction using Chest X-ray and Clinical Data. Computational and Structural Biotechnology Journal, 2023, vol. 24, no. 12, p. 53-65. ISSN: 2001-0370. (https://www.sciencedirect.com/science/article/pii/S2001037023004713)
• MIZERA, J.; GENZOR, S.; SOVA, M.; STANKE, L.; BURGET, R.; JAKUBEC, P.; VYKOPAL, M.; POBEHA, P.; ZAPLETALOVÁ, J. The effectiveness of glucocorticoid treatment in post-COVID-19 pulmonary involvement. Pneumonia, 2024, vol. 16, no. 1, p. 1-10. ISSN: 2200-6133. (https://pubmed.ncbi.nlm.nih.gov/38311783/)
• KOVÁČ, D.; MEKYSKA, J.: Articulatory decay analysis; Vocal Tract Resonances Analysis Software. online. URL: https://github.com/BDALab/Articulatory_decay. (software)
• GALÁŽ, Z.; MUCHA, J.; ZVONČÁK, V.; MEKYSKA, J.: HandwritingFeatures; Handwriting Features. online. URL: https://github.com/BDALab/handwriting-features. (software)