Andrzej Kawiak

Dr. Andrzej Kawiak

orcid

Research Focus & Metrics

Representative Publications

View All Publications
24 May 2018 in Frontiers in Neuroinformatics
Verified
Journal Article
Open access
PDF file
GWGrzegorz Marcin Wójcik
Grzegorz Marcin Wójcik
JMJolanta Masiak
Jolanta Masiak
AKAndrzej Kawiak
Andrzej Kawiak
PSPiotr Schneider
Piotr Schneider
LKLukasz Kwasniewicz
Lukasz Kwasniewicz
AGAnna Gajos-Balinska
Anna Gajos-Balinska
The interview is still the main and most important tool in psychiatrist's work. The neuroimaging methods such as CT or MRI are widely used in other fields of medicine, for instance neurology. However, psychiatry lacks effective quantitative methods to support of diagnosis. A novel neuroinformatic approach to help clinical patients by means of electroencephalographic technology in order to build foundations for finding neurophysiological biomarkers of psychiatric disorders is proposed. A cohort of 30 right-handed patients (21 males, 9 females) with psychiatric disorders (mainly with panic and anxiety disorder, Asperger syndrome as well as with phobic anxiety disorders, schizophrenia, bipolar affective disorder, obsessive-compulsive disorder, nonorganic hypersomnia, and moderate depressive episode) were examined using the dense array EEG amplifier in the P300 experiment. The results were compared with the control group of 30 healthy, right-handed male volunteers. The quantitative analysis of cortical activity was conducted using the sLORETA source localization algorithm. The most active Brodmann Areas were pointed out and a new quantitative observable of electrical charge flowing through the selected Brodmann Area is proposed. The precise methodology and research protocol for collecting EEG data as well as the roadmap of future investigations in this area are presented. The essential result of this study is the idea proven by the initial results of our experiments that it is possible to determine quantitatively biomarkers of particular psychiatric disorders in order to support the process of diagnosis and hopefully choose most appropriate medical treatment later.

24 October 2018 in Frontiers in Neuroinformatics
Verified
Journal Article
Open access
PDF file
GWGrzegorz Marcin Wójcik
Grzegorz Marcin Wójcik
JMJolanta Masiak
Jolanta Masiak
AKAndrzej Kawiak
Andrzej Kawiak
LKLukasz Kwasniewicz
Lukasz Kwasniewicz
PSPiotr Schneider
Piotr Schneider
AGAnna Gajos-Balinska
Anna Gajos-Balinska
There are still no good quantitative methods to be applied in psychiatric diagnosis. The interview is still the main and most important tool in the psychiatrist work. This paper presents the results of electroencephalographic research with the subjects of a group of 30 patients with psychiatric disorders compared to the control group of healthy volunteers. All subjects were solving working memory task. The digit-span working memory task test was chosen as one of the most popular tasks given to subjects with cognitive dysfunctions, especially for the patients with panic disorders, depression (including the depressive phase of bipolar disorder), phobias, and schizophrenia. Having such cohort of patients some results for the subjects with insomnia and Asperger syndrome are also presented. The cortical activity of their brains was registered by the dense array EEG amplifier. Source localization using the photogrammetry station and the sLORETA algorithm was then performed in five EEG frequency bands. The most active Brodmann Areas are indicated. Methodology for mapping the brain and research protocol are presented. The first results indicate that the presented technique can be useful in finding psychiatric disorder neurophysiological biomarkers. The first attempts were made to associate hyperactivity of selected Brodmann Areas with particular disorders.

15 June 2020
Verified
Book Chapter
Open access
PDF file
LKLukasz Kwasniewicz
Lukasz Kwasniewicz
GWGrzegorz Marcin Wójcik
Grzegorz Marcin Wójcik
AKAndrzej Kawiak
Andrzej Kawiak
PSPiotr Schneider
Piotr Schneider
AWAdam Wierzbicki
Adam Wierzbicki
In the Internet era, understanding why humans find messages from unknown receivers credible (such as fake news) is an important research topic. Message credibility is an important theoretical aspect of credibility evaluation that relies only on message contents and design. For the first time in the field, we study message credibility by directly measuring brain activity of humans who make credibility evaluations in an experiment that controls message design. Brain activity as measured using EEG is used to investigate areas of the brain involved in message credibility evaluation. We also model and predict human message credibility evaluations using EEG brain activity measurements.

30 August 2019 in Bio-Algorithms and Med-Systems
Verified
Journal Article
Closed access
PDF file
GWGrzegorz Marcin Wójcik
Grzegorz Marcin Wójcik
AKAndrzej Kawiak
Andrzej Kawiak
LKLukasz Kwasniewicz
Lukasz Kwasniewicz
PSPiotr Schneider
Piotr Schneider
JMJolanta Masiak
Jolanta Masiak
The Event-Related Potentials were investigated on a group of 70 participants using the dense array electroencephalographic amplifier with photogrammetry geodesic station. The source localisation was computed for each participant. The activity of brodmann areas (BAs) involved in the brain cortical activity of each participant was measured. Then the mean electric charge flowing through particular areas was calculated. The five different machine learning tools (logistic regression, boosted decision tree, Bayes point machine, classic neural network and averaged perceptron classifier) from the Azure ecosystem were trained, and their accuracy was tested in the task of distinguishing standard and target responses in the experiment. The efficiency of each tool was compared, and it was found out that the best tool was logistic regression and the boosted decision tree in our task. Such an approach can be useful in eliminating somatosensory responses in experimental psychology or even in establishing new communication protocols with mildly mentally disabled subjects.

1 November 2026 in Applied Sciences
Verified
Journal Article
Open access
PDF file
EBEmil Benedykciuk
Emil Benedykciuk
MDMarcin Denkowski
Marcin Denkowski
MJMarek Jezior
Marek Jezior
AKAndrzej Kawiak
Andrzej Kawiak
AKAdam Kobus
Adam Kobus
GWGrzegorz Marcin Wójcik
Grzegorz Marcin Wójcik
Pre-execution feasibility validation can substantially improve LLM-based task planning without proportional inference cost, but only when it is grounded, deterministic, and tightly bounded. We propose Plan-Then-Verify (PTV), a verifier-first control pattern in which a candidate workflow is checked against a documented knowledge base before execution begins. A planner generates a structured task graph, a deterministic verifier validates functions, artifacts, and ordering constraints, and one bounded repair pass applies local edits when possible. Unsupported workflows are blocked through safe abstention rather than hallucinated capability invention. We evaluate PTV on Core-60, a 60-task public benchmark spanning supported, multi-step, repair-sensitive, and unsupported requests. Across local and server-side model panels, PTV substantially improves the reported static conformance score on several model families, including Gemma, Devstral, and DeepSeek, while remaining close to Plan-and-Execute in token cost. This score combines supported-task success with correct abstention on unsupported requests; it is not a real-execution success rate. Paired bootstrap confidence intervals show positive PTV–PAE conformance differences for the positive model families and near-zero effects for model-serving configurations whose structured outputs collapse. Additional schema-constrained reruns show that this boundary condition is not merely a scoring artifact: Qwen and DeepSeek recover usable structured plans and then show clear PTV gains, whereas GPT-OSS and GLM remain fragile under the tested serving stack. The primary benchmark is static and invokes no real tools; an auxiliary sandbox replay executes stored plans against mock tools derived from the same specification. The primary outcomes measure conformance to the frozen knowledge base and feasibility criteria, while the replay provides side-effect-free trace checks. These results suggest that lightweight feasibility validation improves static SOP conformance primarily by amplifying planners capable of producing stable structured workflows, rather than compensating for weak upstream plan representations. Core-60, the knowledge-base specification, implementation, and analysis materials are publicly available.