Milosz Dudek

Mr. Milosz Dudek

AGH University of Krakow, Kraków, Małopolskie, Poland
orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
10 April 2025 in Sensors
Verified
Journal Article
Open access
PDF file
Milosz DudekMilosz Dudek
Milosz Dudek
DHDaria Hemmerling
Daria Hemmerling
MKMarta Kaczmarska
Marta Kaczmarska
JSJoanna Jolanta Stępień
Joanna Jolanta Stępień
MDMateusz Daniol
Mateusz Daniol
MWMagdalena Wojcik-Pedziwiatr
Magdalena Wojcik-Pedziwiatr
This study explores an innovative approach to early Parkinson’s disease (PD) detection by analyzing speech data collected using a mixed reality (MR) system. A total of 57 Polish participants, including PD patients and healthy controls, performed five speech tasks while using an MR head-mounted display (HMD). Speech data were recorded and analyzed to extract acoustic and linguistic features, which were then evaluated using machine learning models, including logistic regression, support vector machines (SVMs), random forests, AdaBoost, and XGBoost. The XGBoost model achieved the best performance, with an F1-score of 0.90 ± 0.05 in the story-retelling task. Key features such as MFCCs (mel-frequency cepstral coefficients), spectral characteristics, RASTA-filtered auditory spectrum, and local shimmer were identified as significant in detecting PD-related speech alterations. Additionally, state-of-the-art deep learning models (wav2vec2, HuBERT, and WavLM) were fine-tuned for PD detection. HuBERT achieved the highest performance, with an F1-score of 0.94 ± 0.04 in the diadochokinetic task, demonstrating the potential of deep learning to capture complex speech patterns linked to neurodegenerative diseases. This study highlights the effectiveness of combining MR technology for speech data collection with advanced machine learning (ML) and deep learning (DL) techniques, offering a non-invasive and high-precision approach to PD diagnosis. The findings hold promise for broader clinical applications, advancing the diagnostic landscape for neurodegenerative disorders.

6 October 2026
Verified
Book Chapter
Closed access
PDF file
KMKatarzyna Mazur
Katarzyna Mazur
OSOliwia Skórzewska
Oliwia Skórzewska
ZMZuzanna Miodońska
Zuzanna Miodońska
BMBarbara Mika
Barbara Mika
Milosz DudekMilosz Dudek
Milosz Dudek
MKMichal Kręcichwost
Michal Kręcichwost

23 June 2026 in Computer Methods and Programs in Biomedicine
Verified
Journal Article
Closed access
Milosz DudekMilosz Dudek
Milosz Dudek
JSJakub Sikora
Jakub Sikora
JKJustyna Krzywdziak
Justyna Krzywdziak
MZMagda Zbik
Magda Zbik
WSWojciech Szecowka
Wojciech Szecowka
DHDaria Hemmerling
Daria Hemmerling

24 November 2025 in IEEE Pervasive Computing
Verified
Journal Article
Closed access
DHDaria Hemmerling
Daria Hemmerling
PJPaweł Jemioło
Paweł Jemioło
MDMateusz Daniol
Mateusz Daniol
MWMarek Wodzinski
Marek Wodzinski
Milosz DudekMilosz Dudek
Milosz Dudek
MWMagdalena Wójcik-Pędziwiatr
Magdalena Wójcik-Pędziwiatr
This study introduces an innovative multimodal system designed to support the assessment of Parkinson’s disease, atypical Parkinsonian syndromes, and Huntington’s disease by integrating mixed-reality (MR) technology, specifically using augmented reality goggles. The system captures symptoms related to neurodegenerative disorders through detailed sensor data, including hand movements, gait, and gaze patterns. Central to our methodology is an interactive, game-like experience lasting 30 min, comprising 17 tasks that evaluate motor skills, speech, memory, cognitive functions, gait, and gaze. These tasks were developed in collaboration with neurologists and draw upon established clinical paradigms. This engaging experience, combined with traditional clinical tests, provides a comprehensive assessment of both physical and cognitive capabilities. Rather than aiming to provide a diagnostic decision, this study focuses on evaluating the feasibility and usability of the system in a clinical setting. By automating diagnostic tasks and utilizing MR technology, our approach enhances patient comfort and offers a more immersive and efficient assessment process. The system was evaluated with 15 neurodegenerative patients and 32 healthy subjects, demonstrating high scores in both immersion and usability across both groups. This work represents a first step toward integrating MR into clinical workflows for neurodegenerative disease assessment.

1 January 2026 in Sensors
Verified
Journal Article
Open access
PDF file
DHDaria Hemmerling
Daria Hemmerling
Milosz DudekMilosz Dudek
Milosz Dudek
JKJustyna Krzywdziak
Justyna Krzywdziak
MŻMagda Żbik
Magda Żbik
WSWojciech Szecowka
Wojciech Szecowka
MWMagdalena Wójcik-Pędziwiatr
Magdalena Wójcik-Pędziwiatr
Parkinson’s disease (PD) alters both speech and movement, yet most automated assessments still treat these signals separately. We examined whether combining voice with head motion improves discrimination between patients and healthy controls (HC). Synchronous measurements of acoustic and inertial signals were collected using a HoloLens 2 headset. Data were obtained from 165 participants (72 PD/93 HC), following a standardized mixed-reality (MR) protocol. We benchmarked single-modality models against fusion strategies under 5-fold stratified cross-validation. Voice alone was robust (pooled AUC ≈ 0.865), while the inertial channel alone was near chance (AUC ≈ 0.497). Fusion provided a modest but repeatable improvement: gated early-fusion achieved the highest AUC (≈0.875), cross-attention fusion was comparable (≈0.873). Gains were task-dependent. While speech-dominated tasks were already well captured by audio, tasks that embed movement benefited from complementary inertial data. Proposed MR capture proved feasible within a single session and showed that motion acts as a conditional improvement factor rather than a sole predictor. The results outline a practical path to multimodal screening and monitoring for PD, preserving the reliability of acoustic biomarkers while integrating kinematic features when they matter.