Brain Diseases Analysis Laboratory (BDALab)

About the lab

Who We Are: The Brain Diseases Analysis Laboratory (BDALab) is an international multidisciplinary research group focusing on the objective and quantitative analysis of brain diseases.

What We Do: Using the state-of-the-art techniques of biomedical signal processing, machine learning and statistical analysis we provide neuroscientists, psychologists and speech-language pathologists with digital biomarkers facilitating diagnosis, assessment and monitoring of neurodegenerative and neurodevelopmental diseases. We are working in the Health 4.0 concept.

Our Mission: To provide experts with methods enabling objective and quantitative analysis of brain diseases symptoms.

Our Vision: To make brain diseases easier to understand.


Featured research (109)

Parkinson's disease dysgraphia (PDYS), one of the earliest signs of Parkinson's disease (PD), has been researched as a promising biomarker of PD and as the target of a noninvasive and inexpensive approach to monitoring the progress of the disease. However, although several approaches to supportive PDYS diagnosis have been proposed (mainly based on handcrafted features (HF) extracted from online handwriting or the utilization of deep neural networks), it remains unclear which approach provides the highest discrimination power and how these approaches can be transferred between different datasets and languages. This study aims to compare classification performance based on two types of features: features automatically extracted by a pretrained convolutional neural network (CNN) and HF designed by human experts. Both approaches are evaluated on a multilingual dataset collected from 143 PD patients and 151 healthy controls in the Czech Republic, United States, Colombia, and Hungary. The subjects performed the spiral drawing task (SDT; a language-independent task) and the sentence writing task (SWT; a language-dependent task). Models based on logistic regression and gradient boosting were trained in several scenarios, specifically single language (SL), leave one language out (LOLO), and all languages combined (ALC). We found that the HF slightly outperformed the CNN-extracted features in all considered evaluation scenarios for the SWT. In detail, the following balanced accuracy (BACC) scores were achieved: SL—0.65 (HF), 0.58 (CNN); LOLO—0.65 (HF), 0.57 (CNN); and ALC—0.69 (HF), 0.66 (CNN). However, in the case of the SDT, features extracted by a CNN provided competitive results: SL—0.66 (HF), 0.62 (CNN); LOLO—0.56 (HF), 0.54 (CNN); and ALC—0.60 (HF), 0.60 (CNN). In summary, regarding the SWT, the HF outperformed the CNN-extracted features over 6% (mean BACC of 0.66 for HF, and 0.60 for CNN). In the case of the SDT, both feature sets provided almost identical classification performance (mean BACC of 0.60 for HF, and 0.58 for CNN).
Writing is a complex skill. Issues in this process, which are usually associated with developmental dysgraphia (DD), could consistently cause problems in everyday life, like for example, lower self-esteem and poorer academic achievement. That is why the correct diagnosis of DD is crucial for further child development. DD belongs to the category of specific learning disabilities and according to different studies, its prevalence ranges between 0.1 and 30 percent. Diagnosing a child with DD relies, in the first place, on teachers. After that, psychologists, or special educational specialists (in the Czech Republic) commonly use qualitative evaluation of the written process, where the child is observed when he or she is writing. Nevertheless, there are no objective tests or standardized examinations for the assessment of handwriting deficiency either in special educational or psychological practices. In the frame of current research, a new quantitative approach to handwriting proficiency assessment was developed. Digitizing tablets (Wacom Intuos Pro L) with a special inking pen (Wacom Ink Pen) are used to record the online handwriting process and graphomotor skills of children. Administration templates contain simple graphomotor elements and complex figures related to DD symptoms and cognitive (memory and visuospatial) abilities. This new approach to diagnose handwriting issues will be presented in this article.
Background. Hypokinetic dysarthria is a common but difficult-to-treat symptom of Parkinson’s disease (PD). Objectives. We evaluated the long-term effects of multiple-session repetitive transcranial magnetic stimulation on hypokinetic dysarthria in PD. Neural mechanisms of stimulation were assessed by functional MRI. Methods. A randomized parallel-group sham stimulation-controlled design was used. Patients were randomly assigned to ten sessions (2 weeks) of real (1Hz) or sham stimulation over the right superior temporal gyrus. Stimulation effects were evaluated at weeks 2, 6, and 10 after the baseline assessment. Articulation, prosody, and speech intelligibility were quantified by speech therapist using a validated tool (Phonetics score of the Dysarthric Profile). Activations of the speech network regions and intrinsic connectivity were assessed using 3T MRI. Linear mixed models and post-hoc tests were utilized for data analyses. Results. Altogether 33 PD patients completed the study (20 in the real stimulation group and 13 in the sham stimulation group). Linear mixed models revealed significant effects of time (F(3, 88.1) = 22.7, p < 0.001) and time-by-group interactions: F(3, 88.0) = 2.8, p = 0.040) for the Phonetics score. Real as compared to sham stimulation led to activation increases in the orofacial sensorimotor cortex and caudate nucleus and to increased intrinsic connectivity of these regions with the stimulated area. Conclusions. This is the first study to show the long-term treatment effects of non-invasive brain stimulation for hypokinetic dysarthria in PD. Neural mechanisms of the changes are discussed.
Background: We aimed to confirm the "Mozart effect" in epileptic patients using the intracerebral EEG recordings and the hypothesis that the reduction of epileptiform discharges (ED) can be explained by the music's acoustic properties. Methods: Eighteen epilepsy surgery candidates were implanted with depth electrodes in the temporal medial and lateral cortex. Patients listened to the first movement of Mozart's Sonata for Two Pianos K. 448 and to the first movement of Haydn's Symphony No. 94. Musical features from each composition with respect to rhythm, melody, and harmony were analysed. Results: ED in intracerebral EEG were reduced by Mozart's music. Listening to Haydn's music led to reduced ED only in the women; in the men, the ED increased. The acoustic analysis revealed that non-dissonant music with a harmonic spectrum and decreasing tempo with significant high-frequency parts has a reducing effect on ED in men. To reduce ED in women, the music should additionally be, in terms of loudness, gradually less dynamic. Finally, we were able to demonstrate that these acoustic characteristics are more dominant in Mozart's music than in Haydn's music. Conclusions: We confirmed the reduction of intracerebral ED while listening to classical music. An analysis of the musical features revealed that the acoustic characteristics of music are responsible for supressing brain epileptic activity. Based on our study we suggest to study the use of musical pieces with well-defined acoustic properties as an alternative non-invasive method to reduce epileptic activity in patients with epilepsy.

Lab head

Jiri Mekyska
  • Department of Telecommunications
About Jiri Mekyska
  • Jiri Mekyska a principal scientist and the head of the BDALab (Brain Diseases Analysis Laboratory) developing new digital biomarkers enabling to better understand, diagnose and monitor neurodegenerative (e.g. Parkinson’s disease) and neurodevelopmental (e.g. dysgraphia) diseases. I lead a multidisciplinary team of researchers (signal processing engineers, data scientists, neurologists, psychologists, etc.) that moves the research in the field beyond the state of the art.

Members (10)

Zdenek Smekal
  • Brno University of Technology
Zoltán Galáž
  • Brno University of Technology
Vojtech Zvoncak
  • Brno University of Technology
Ján Mucha
  • Brno University of Technology
Tomáš Kiska
  • Brno University of Technology
Pavol Harár
  • University of Vienna
Justyna Skibińska
  • Brno University of Technology
Marek Mikulec
  • Brno University of Technology

Alumni (1)

Peter Drotar
  • Technical University of Kosice - Technicka univerzita v Kosiciach