arXiv subjects

Katarzyna Wac

Publications and source records attributed to Katarzyna Wac.

4 recordsLinked to original sources

arXiv2026

Representation Matters in Longitudinal Affective Computing

Longitudinal, in-the-wild, wearable sensing yields day-level physiology, sleep, activity, and environmental streams, whereas affect and cognition are labeled only episodically (per waves). We recast this cadence mismatch as a temporal representation problem and compare three wave-level mappings from dense histories to sparse labels: levels (within-wave summaries), absolute drift (change across waves), and proportional drift. Using almost a year of data from 82 adults in the Providemus alz study, we model 21 affect and cognition outcomes. Day-scale signals are reduced to compact wave-level descriptors (central tendency, dispersion, and distributional shape) and learned with four regressors under two orthogonal evaluation axes: leave-one-subject-out and leave-one-wave-out. Performance is reported as scaled MAE using both mean and median across folds. Differences emerge: affective states are best predicted by wave-to-wave absolute drift, whereas cognitive performance aligns with within-wave levels, reflecting emotion dynamic theories. Across windowing features, shape descriptors (e.g., minima, kurtosis) carry more signal than simple means/medians. We contribute a representation triad for sparse-label modelling, a wave-level feature schema applicable on-device, and a dual-axis reporting practice that separates cross-participant generalization from temporal robustness. These results convert temporal representation from an implicit preprocessing step into an explicit, testable design choice for real-world affective-computing applications in brain health.

cs.HC↗

arXiv2026

SHANDS: A Multi-View Dataset and Benchmark for Surgical Hand-Gesture and Error Recognition Toward Medical Training

In surgical training for medical students, proficiency development relies on expert-led skill assessment, which is costly, time-limited, difficult to scale, and its expertise remains confined to institutions with available specialists. Automated AI-based assessment offers a viable alternative, but progress is constrained by the lack of datasets containing realistic trainee errors and the multi-view variability needed to train robust computer vision approaches. To address this gap, we present Surgical-Hands (SHands), a large-scale multi-view video dataset for surgical hand-gesture and error recognition for medical training. \textsc{SHands} captures linear incision and suturing using five RGB cameras from complementary viewpoints, performed by 52 participants (20 experts and 32 trainees), each completing three standardized trials per procedure. The videos are annotated at the frame level with 15 gesture primitives and include a validated taxonomy of 8 trainee error types, enabling both gesture recognition and error detection. We further define standardized evaluation protocols for single-view, multi-view, and cross-view generalization, and benchmark state-of-the-art deep learning models on the dataset. SHands is publicly released to support the development of robust and scalable AI systems for surgical training grounded in clinically curated domain knowledge.

cs.CV↗

arXiv2018

Visions and Challenges in Managing and Preserving Data to Measure Quality of Life

Health-related data analysis plays an important role in self-knowledge, disease prevention, diagnosis, and quality of life assessment. With the advent of data-driven solutions, a myriad of apps and Internet of Things (IoT) devices (wearables, home-medical sensors, etc) facilitates data collection and provide cloud storage with a central administration. More recently, blockchain and other distributed ledgers became available as alternative storage options based on decentralised organisation systems. We bring attention to the human data bleeding problem and argue that neither centralised nor decentralised system organisations are a magic bullet for data-driven innovation if individual, community and societal values are ignored. The motivation for this position paper is to elaborate on strategies to protect privacy as well as to encourage data sharing and support open data without requiring a complex access protocol for researchers. Our main contribution is to outline the design of a self-regulated Open Health Archive (OHA) system with focus on quality of life (QoL) data.

cs.CY↗

arXiv2013

Smartphone as a Personal, Pervasive Health Informatics Services Platform: Literature Review

Objectives: The article provides an overview of current trends in personal sensor, signal and imaging informatics, that are based on emerging mobile computing and communications technologies enclosed in a smartphone and enabling the provision of personal, pervasive health informatics services. Methods: The article reviews examples of these trends from the PubMed and Google scholar literature search engines, which, by no means claim to be complete, as the field is evolving and some recent advances may not be documented yet. Results: There exist critical technological advances in the surveyed smartphone technologies, employed in provision and improvement of diagnosis, acute and chronic treatment and rehabilitation health services, as well as in education and training of healthcare practitioners. However, the most emerging trend relates to a routine application of these technologies in a prevention/wellness sector, helping its users in self-care to stay healthy. Conclusions: Smartphone-based personal health informatics services exist, but still have a long way to go to become an everyday, personalized healthcare-provisioning tool in the medical field and in a clinical practice. Key main challenge for their widespread adoption involve lack of user acceptance striving from variable credibility and reliability of applications and solutions as they a) lack evidence-based approach; b) have low levels of medical professional involvement in their design and content; c) are provided in an unreliable way, influencing negatively its usability; and, in some cases, d) being industry-driven, hence exposing bias in information provided, for example towards particular types of treatment or intervention procedures.

cs.CY↗