arXiv subjects

Anna Sigridur Islind

Publications and source records attributed to Anna Sigridur Islind.

9 recordsLinked to original sources

arXiv2026

Staging by the Book: Automatic Sleep Stage Classification Using Scoring Rules

Automated sleep staging is commonly approached as a supervised machine learning problem, with deep learning methods dominating recent research. While machine learning models achieve near-human level agreement with human-scored reference sleep stages, their decisions are typically opaque and not designed to follow clinical scoring rules. We propose a transparent alternative: a deterministic, rule-based sleep staging method that explicitly operationalizes the American Academy of Sleep Medicine's (AASM) scoring logic as executable code, coupled with epoch-level natural-language justifications derived from an explanation trace. We evaluate the approach on 50 polysomnography recordings with a 10-scorer majority-vote consensus as reference. Across all recordings, the method agreed with the majority-vote reference in 60.5% of epochs ($κ=0.42$), with substantially higher agreement on a dataset used during development (77.1%, $κ=0.61$). Agreement with the reference was highest for sleep stage N2 (recall 83.5%) and moderate for sleep stage R (recall 68.7%), while Wake and N1 recall were low. Despite lower agreement with the reference than contemporary deep learning models, the method provides deterministic decisions and natural language explanations aligned with AASM scoring rules, making it a complementary tool for auditing, debugging, and governing deep learning-based sleep staging.

eess.SP↗

arXiv2026

Data-Local Autonomous LLM-Guided Neural Architecture Search for Multiclass Multimodal Time-Series Classification

Applying machine learning to sensitive time-series data is often bottlenecked by the iteration loop: Performance depends strongly on preprocessing and architecture, yet training often has to run on-premise under strict data-local constraints. This is a common problem in healthcare and other privacy-constrained domains (e.g., a hospital developing deep learning models on patient EEG). This bottleneck is particularly challenging in multimodal fusion, where sensor modalities must be individually preprocessed and then combined. LLM-guided neural architecture search (NAS) can automate this exploration, but most existing workflows assume cloud execution or access to data-derived artifacts that cannot be exposed. We present a novel data-local, LLM-guided search framework that handles candidate pipelines remotely while executing all training and evaluation locally under a fixed protocol. The controller observes only trial-level summaries, such as pipeline descriptors, metrics, learning-curve statistics, and failure logs, without ever accessing raw samples or intermediate feature representations. Our framework targets multiclass, multimodal learning via one-vs-rest binary experts per class and modality, a lightweight fusion MLP, and joint search over expert architectures and modality-specific preprocessing. We evaluate our method on two regimes: UEA30 (public multivariate time-series classification dataset) and SleepEDFx sleep staging (heterogeneous clinical modalities such as EEG, EOG, and EMG). The results show that the modular baseline model is strong, and the LLM-guided NAS further improves it. Notably, our method finds models that perform within published ranges across most benchmark datasets. Across both settings, our method reduces manual intervention by enabling unattended architecture search while keeping sensitive data on-premise.

cs.LG↗

arXiv2023

Proxy Design: A Method for Involving Proxy Users to Speak on Behalf of Vulnerable or Unreachable Users in Co-Design

Designing digital artifacts is not a linear, straightforward process. This is particularly true when applying a user-centered design approach, or co-design, with users who are unable to participate in the design process. Although the reduced participation of a particular user group may harm the end result, the literature on solving this issue is sparse. In this article, proxy design is outlined as a method for involving a user group as proxy users to speak on behalf of a group that is difficult to reach. We present a design ethnography spanning three years at a cancer rehabilitation clinic, where digital artifacts were designed to be used collaboratively by nurses and patients. The empirical data were analyzed using content analysis and consisted of 20 observation days at the clinic, six proxy design workshops, 21 telephone consultations between patients and nurses, and log data from the digital artifact. We show that simulated consultations, with nurses roleplaying as proxies for patients ignited and initiated the design process and enabled an efficient in-depth understanding of patients. Moreover, we reveal how proxy design as a method further expanded the design. We illustrate: (1) proxy design as a method for initiating design, (2) proxy design as an embedded element in co-design and (3) six design guidelines that should be considered when engaging in proxy design. The main contribution is the conceptualization of proxy design as a method that can ignite and initiate the co-design process when important users are unreachable, vulnerable or unable to represent themselves in the co-design process. Based on the empirical findings from a design ethnography that involved nurses as proxy users speaking on behalf of patients, the article shows that roleplaying in proxy design is a fitting way of initiating the design process, outlining proxy design as an embedded element of co-design.

cs.HC↗

arXiv2023

aSAGA: Automatic Sleep Analysis with Gray Areas

State-of-the-art automatic sleep staging methods have already demonstrated comparable reliability and superior time efficiency to manual sleep staging. However, fully automatic black-box solutions are difficult to adapt into clinical workflow and the interaction between explainable automatic methods and the work of sleep technologists remains underexplored and inadequately conceptualized. Thus, we propose a human-in-the-loop concept for sleep analysis, presenting an automatic sleep staging model (aSAGA), that performs effectively with both clinical polysomnographic recordings and home sleep studies. To validate the model, extensive testing was conducted, employing a preclinical validation approach with three retrospective datasets; open-access, clinical, and research-driven. Furthermore, we validate the utilization of uncertainty mapping to identify ambiguous regions, conceptualized as gray areas, in automatic sleep analysis that warrants manual re-evaluation. The results demonstrate that the automatic sleep analysis achieved a comparable level of agreement with manual analysis across different sleep recording types. Moreover, validation of the gray area concept revealed its potential to enhance sleep staging accuracy and identify areas in the recordings where sleep technologists struggle to reach a consensus. In conclusion, this study introduces and validates a concept from explainable artificial intelligence into sleep medicine and provides the basis for integrating human-in-the-loop automatic sleep staging into clinical workflows, aiming to reduce black-box criticism and the burden associated with manual sleep staging.

cs.LG↗

arXiv2022

State of the Art of Audio- and Video-Based Solutions for AAL

The report illustrates the state of the art of the most successful AAL applications and functions based on audio and video data, namely (i) lifelogging and self-monitoring, (ii) remote monitoring of vital signs, (iii) emotional state recognition, (iv) food intake monitoring, activity and behaviour recognition, (v) activity and personal assistance, (vi) gesture recognition, (vii) fall detection and prevention, (viii) mobility assessment and frailty recognition, and (ix) cognitive and motor rehabilitation. For these application scenarios, the report illustrates the state of play in terms of scientific advances, available products and research project. The open challenges are also highlighted.

cs.CY↗

arXiv2021

Nordic vaccination strategies face/off via age range comparative analysis on key indicators of COVID-19 severity and healthcare stress level

In this paper we analyze the impact of vaccinations on spread of the COVID-19 virus for different age groups. More specifically we examine the deployment of vaccines in the Nordic countries in a comparative analysis where we analyze factors such as healthcare stress level and severity of disease through new infections, hospitalizations, intensive care unit (ICU) occupancy and deaths. Moreover, we analyze the impact of the various vaccine types, vaccination rate on the spread of the virus in each age group for Denmark, Finland, Iceland, Norway and Sweden from the start of the vaccination period in December 2020 until the end of August 2021. We perform a three-fold analysis: i) frequency analysis of infections and vaccine rates by age groups, ii) rolling correlations between vaccination strategies, severity of COVID-19 and healthcare stress level and; iii) we also employ the epidemic Renormalization Group (eRG) framework. The latter is used to mathematically model wave structures, as well as the impact of vaccinations on wave dynamics. We further compare the Nordic countries with England. Our main results are the quantification of the impact of the vaccination campaigns on age groups epidemiological data, across countries with high vaccine uptake. The data clearly shows that vaccines markedly reduce the number of new cases and the risk of serious illness.

q-bio.PE↗

arXiv2021

Effects of the COVID-19 Pandemic on Learning and Teaching: a Case Study from Higher Education

In December 2019, the first case of SARS-CoV-2 infection was identified in Wuhan, China. Since that day, COVID-19 has spread worldwide, affecting 153 million people. Education, as many other sectors, has managed to adapt to the requirements and barriers implied by the impossibility to teach students face-to-face as it was done before. Yet, little is known about the implications of emergency remote teaching (ERT) during the pandemic. This study describes and analyzes the impact of the pandemic on the study patterns of higher education students. The analysis was performed by the integration of three main components: (1) interaction with the learning management system (LMS), (2) Assignment submission rate, and (3) Teachers' perspective. Several variables were created to analyze the study patterns, clicks on different LMS components, usage during the day, week and part of the term, the time span of interaction with the LMS, and grade categories. The results showed significant differences in study patterns depending on the year of study, and the variables reflecting the effect of teachers' changes in the course structure are identified. This study outlines the first insights of higher education's new normality, providing important implications for supporting teachers in creating academic material that adequately addresses students' particular needs depending on their year of study, changes in study pattern, and distribution of time and activity through the term.

cs.CY↗

arXiv2021

You better watch out: US COVID-19 wave dynamics versus vaccination strategy

We employ the epidemic Renormalization Group (eRG) framework to understand, reproduce and predict the COVID-19 pandemic diffusion across the US. The human mobility across different geographical US divisions is modelled via open source flight data alongside the impact of social distancing for each such division. We analyse the impact of the vaccination strategy on the current pandemic wave dynamics in the US. We observe that the ongoing vaccination campaign will not impact the current pandemic wave and therefore strict social distancing measures must still be enacted. To curb the current and the next waves our results indisputably show that vaccinations alone are not enough and strict social distancing measures are required until sufficient immunity is achieved. Our results are essential for a successful vaccination strategy in the US.

physics.soc-ph↗

arXiv2020

Changes in mobility patterns in Europe during the COVID-19 pandemic: Novel insights using open source data

The COVID-19 pandemic has changed the way we act, interact and move around in the world. The pandemic triggered a worldwide health crisis that has been tackled using a variety of strategies across Europe. Whereas some countries have taken strict measures, others have avoided lock-downs altogether. In this paper, we report on findings obtained by combining data from different publicly available sources in order to shed light on the changes in mobility patterns in Europe during the pandemic. Using that data, we show that mobility patterns have changed in different counties depending on the strategies they adopted during the pandemic. Our data shows that the majority of European citizens walked less during the lock-downs, and that, even though flights were less frequent, driving increased drastically. In this paper, we focus on data for a number of countries, for which we have also developed a dashboard that can be used by other researchers for further analyses. Our work shows the importance of granularity in open source data and how such data can be used to shed light on the effects of the pandemic.

cs.CY↗