Computed tomography (CT) provides high-quality imaging but is costly and exposes patients to relatively high radiation. Cone-beam CT (CBCT) offers faster, lower-dose, and more cost-effective 3D imaging but suffers from scatter, truncation, and noise that limit its clinical utility. In CBCT-to-CT image enhancement, diffusion models achieve strong performance but incur high computational cost. To tackle this problem, we propose Conditional Efficient DDIM (CE-DDIM), an efficient CBCT-to-CT image enhancement framework. CE-DDIM integrates a dual-head conditional U-Net trained with a heteroscedastic negative log-likelihood and total variation regularizer to jointly predict denoising targets and pixel-wise variance. Moreover, a multi-window Hounsfield Unit input is used to improve the reconstruction quality. This design enables high-fidelity reconstructions with 10 × reduction in steps compared to previous methods. On the SynthRAD2023 datasets, CE-DDIM improves both reconstruction fidelity and sampling efficiency over baselines, providing a practical solution for synthetic CT generation.

Clinicians often need to retrieve patient-specific information from electronic health records (EHRs), a task that is time-consuming and error-prone. We present a locally deployable Clinical Contextual Question Answering (CCQA) framework that answers clinical questions directly from EHRs without external data transfer. Open-source large language models (LLMs) ranging from 4B to 70B parameters were benchmarked under fully offline conditions using 1,664 expert-annotated question-answer pairs derived from records of 183 patients. The dataset consisted predominantly of Finnish clinical text. In free-text generation, Llama-3.1-70B achieved 95.3

Many quantities that characterize network elements are defined in an explicit form and calculated directly from the network structure; examples of include several centrality measures like degree, closeness, or betweenness. However, there are also implicitly defined quantitative measures, which are usually calculated iteratively, in a self-consistent manner, like PageRank or countries' fitness / products' complexity relations. The iteration algorithms involve calculations over the entire network; therefore, their convergence properties depend on the structure of the network. Here, we focus on investigating self-consistently defined quantities in bipartite networks of two sets of nodes where the quantities in one set are determined by the quantities in the other set and vice versa. We derive an explicit convergence criterion for iterations of these quantities and describe two different approaches to improve the convergence properties. In the first one, we identify "problematic nodes" that can be removed or merged while in the second one, we introduce a regularization scheme and show how to estimate the regularization parameter.

In complex networks, “circles” are attribute-defined subgraphs whose nodes share common characteristics (e.g., group membership or categories). Nodes that belong to multiple such circles form overlapping regions, but their role in influence spreading processes remains somewhat underexplored. We analyse several networks with circle structures using a probabilistic influence spreading model for processes of simple and complex contagion in them. We quantify the importance of these overlapping nodes using three metrics, i.e., In-Centrality, Out-Centrality, and Betweenness Centrality, which represent the susceptibility, spreading power, and mediating role of nodes, respectively. We find that, at each stage of the spreading process, the overlapping nodes systematically exhibit greater influence than the non-overlapping nodes, even when accounting for structural heterogeneity, i.e., node connectivity. Furthermore, we observe that the criteria used to define circles shape the overlapping effects. When we restrict our analysis to only the largest circles, we find that circles reflect not only node-level attributes but also of topological importance. These findings help clarify the distinction between local attribute-driven circles and global community structures, thus highlighting the strategic importance of overlapping nodes in spreading dynamics. This provides a foundation for future research on overlapping nodes in both circles and communities.

Evolving open systems, in which new entities are continually introduced and those turning unfit go extinct, exhibit a phase transition between a diverging phase, where the system size grows indefinitely, and a finite phase, where it remains bounded. We show that imposing a bipartite interaction topology alone leaves this transition unchanged when the two partitions are introduced with equal initial connectivity. In contrast, when the initial degrees are asymmetric, the robustness of the system is markedly enhanced such that the transition shifts to higher connectivity and the diverging phase persists even when both initial degrees individually exceed the critical point of the corresponding unstructured system. In addition, we find a re-entrant transition, i.e. a return to the diverging phase as asymmetry is increased while the initial degree of one of the partitions is fixed, making it lying entirely outside the original mean-field picture. An extended mean-field analysis identifies the origin of these effects such that in the asymmetric regime, a feedback between the bipartite handshaking constraint and different extinction rates drives the mean degree of emergent network far above the initially assigned connectivity. This degree elevation suppresses extinction probabilities across the community while simultaneously concentrating extinctions among recently introduced, low-degree nodes. The interplay of these two effects constitutes a simple and universal robustness mechanism for evolving systems with asymmetric bipartite structure.

Human mobility and other social activity patterns influence various aspects of society such as urban planning, traffic predictions, crisis resilience, and epidemic prevention. The behaviour of individuals, like their communication frequencies and movements, are shaped by societal and socio-economic factors. In addition, the differences in the geolocation of people as well as their gender and age cast effects on their activity patterns. In this study we focus on investigating these patterns by using mobile phone data, specifically the call detail records (CDRs), to analyze the social communication and mobility patterns of people. This dataset can provide us insight into the individual and population-level behaviours in rural and urban environments on a daily, weekly and seasonal basis. The results of our analyses show that in the urban areas people have high calling activity but low mobility, while in the rural areas they show the opposite behaviour, i.e. low calling activity combined with high mobility. Overall, there is a decreasing trend in people's mobility through the year even though their calling activity remained consistent except for the holidays during which time the communication frequency drops markedly. We have also observed that there are significant differences in the mobility between the work days and free days. Finally, the age and gender of individuals have also been observed to play a role in the seasonal patterns differently in urban and rural areas.

Accurate segmentation of the primary gross tumor volumes and metastatic lymph nodes in head and neck cancer is crucial for radiotherapy but remains challenging due to high interobserver variability, highlighting a need for an effective auto-segmentation tool. Tumor delineation is used throughout radiotherapy for treatment planning, initially for pre-radiotherapy (pre-RT) MRI scans followed-up by mid-radiotherapy (mid-RT) during the treatment. For the pre-RT task, we propose a dual-stage 3D UNet approach using cascaded neural networks for progressive accuracy refinement. The first-stage models produce an initial binary segmentation, which is then refined with an ensemble of second-stage models for a multiclass segmentation. In Head and Neck Tumor Segmentation for MR-Guided Applications (HNTS-MRG) 2024 Task 1, we utilize a dataset consisting of pre-RT and mid-RT T2-weighted MRI scans. The method is trained using 5-fold cross-validation and evaluated as an ensemble of five coarse models and ten refinement models. Our approach (team FinoxyAI) achieves a mean aggregated Dice similarity coefficient of 0.737 on the test set. Moreover, with this metric, our dual-stage approach highlights consistent improvement in segmentation performance across all folds compared to a single-stage segmentation method.

Traditional measures of closeness and betweenness centrality in networks rely on the shortest paths between nodes. Many standard metrics fail to accurately reflect the physical or probabilistic characteristics of nodal centrality and network flow, often overlooking processes such as cyclic and recurrent spreading. Here, we present new metrics based on our influence spreading model. These probabilistic measures consider all feasible paths, walks, and cycles within the network. We define in-centrality to assess how central a node is as a target of influence, and out-centrality for its role as a source of influence. We compare our metrics with standard ones by analyzing node rankings, using scatter plots, and calculating the Pearson correlation and Spearman's rank correlation coefficients. Our findings show that the betweenness centrality defined by the influence spreading model emphasizes the importance of alternative routes while maintaining similarity to standard betweenness centrality.

Non-invasive mobile wearables like fitness trackers, smart watches and rings allow an easy and less expensive approach to study everyday human behaviour. This alternative approach not only supplements clinical studies, but also provides an opportunity to overcome some of the limitations in them. One of the major challenges faced by them is studying long-term human health and behaviour in realistic settings. Here we have utilised Oura rings to obtain granular data from nineteen healthy participants over the span of one year (October 2023 - September 2024) along with monthly surveys for nine months to track their subjective stress within the duration of the study. We have studied longitudinal sleep and activity patterns of three chronotype groups of participating individuals: morning type (MT), neither type (NT) and evening type (ET). We find that while ET individuals do not seem to lead as healthy life as the MT or NT individuals, they have seemingly improved their habits during the duration of the study. We also show that the Daylight Saving Time changes affect the chronotypes differently. Finally, by utilising mixed effects regression model, we have shown that the stress an individual experiences has a significant correlation with his or her total sleep duration, monthly survey response time, and age.

Members of a society can be characterized by a large number of features, such as gender, age, ethnicity, religion, social status, and shared activities. One of the main tie-forming factors between individuals in human societies is homophily, the tendency of being attracted to similar others. Homophily has been studied mainly in the context of link formation and social dynamics. However, less is known about the role of the multidimensional homophily in forming egocentric communities on Information and Communications Technology (ICT) services. To close this gap, we analyze three ICT datasets, namely, two online social networks and one network deduced from mobile phone calls, in all of which data about individual features are available. We identify communities within egocentric networks and surprisingly find that the larger the community, the more overlap is found between features of its members and the ego. We interpret this finding in terms of the effort needed to manage the communities; the larger diversity requires more effort such that maintaining a large diverse group may exceed the capacity of the members. As the ego reaches out to their alters on an ICT service, we observe that the first alter in each community tends to have a higher feature overlap with the ego than the rest. Moreover, the feature overlap of the ego with all their alters displays a non-monotonic behavior as a function of the ego's degree. We propose a simple mechanism of how people add links in their egocentric networks of alters that reproduces the empirical observations and shows the reason behind non-monotonic tendency of the egocentric feature overlap as a function of the ego's degree.

Artificial intelligence (AI) achieves high-quality annotations of radiological images, yet often lacks the robustness required in clinical practice. Interactive annotation starts with an AI-generated delineation, allowing radiologists to refine it with feedback, potentially improving precision and reliability. These techniques have been explored in two-dimensional desktop environments, but are not validated by radiologists or integrated with immersive visualization technologies. We used a Virtual Reality (VR) system to determine whether (1) the annotation quality improves when radiologists can edit the AI annotation and (2) whether the extra work done by editing is worthwhile. We evaluated the clinical feasibility of an interactive VR approach to annotate mandibular and mental foramina on segmented 3D mandibular models. Three experienced dentomaxillofacial radiologists reviewed AI-generated annotations and, when needed, refined them at the voxel level in 3D space through click-based interactions until clinical standards were met. Our results indicate that integrating expert feedback within an immersive VR environment enhances annotation accuracy, improves clinical usability, and offers valuable insights for developing medical image analysis systems incorporating radiologist input. This study is the first to compare the quality of original and interactive AI annotation and to use radiologists’ opinions as the measure. More research is needed for generalization.

Graph neural networks (GNNs) have emerged as a state-of-the-art data-driven tool for modeling connectivity data of graph-structured complex networks and integrating information of their nodes and edges in space and time. However, as of yet, the analysis of social networks using the time series of people's mobile connectivity data has not been extensively investigated. In the present study, we investigate four recently proposed snapshot - based temporal GNNs in predicting the phone call and SMS activity between users of a mobile communication network. In addition, we develop a simple non - GNN baseline model using recently proposed EdgeBank method. Our analysis shows that the ROLAND temporal GNN outperforms the baseline model in most cases, whereas the other three GNNs perform on average worse than the baseline. The results show that GNN based approaches hold promise in the analysis of temporal social networks through mobile connectivity data. However, due to the relatively small performance margin between ROLAND and the baseline model, further research is required on specialized GNN architectures for temporal social network analysis.

Radiotherapy is the main treatment modality of oropharyngeal cancer (OPC), in which an accurate segmentation of primary gross tumor volume (GTVt) is essential but also challenging due to significant interobserver variability and the time consumed in manual tumor delineation. For such a challenge an interactive deep learning (DL) based approach offers the advantage of automatic high-performance segmentation with the flexibility for user correction when necessary. In this study, we investigate an interactive DL for GTVt segmentation in OPC by introducing a novel two-stage Interactive Click Refinement (2S-ICR) framework and implementing state-of-the-art algorithms. Using the 2021 HEad and neCK TumOR dataset for development and an external dataset from The University of Texas MD Anderson Cancer Center for evaluation, the 2S-ICR framework achieves a Dice similarity coefficient of 0.722 ± 0.142 without user interaction and 0.858 ± 0.050 after ten interactions, thus outperforming existing methods in both cases.

The 2030 Agenda for Sustainable Development of the United Nations outlines 17 goals for countries of the world to address global challenges in their development. However, the progress of countries towards these goal has been slower than expected and, consequently, there is a need to investigate the reasons behind this fact. In this study, we have used a novel data-driven methodology to analyze time-series data for over 20 years (2000-2022) from 107 countries using unsupervised machine learning (ML) techniques. Our analysis reveals strong positive and negative correlations between certain SDGs (Sustainable Development Goals). Our findings show that progress toward the SDGs is heavily influenced by geographical, cultural and socioeconomic factors, with no country on track to achieve all the goals by 2030. This highlights the need for a region-specific, systemic approach to sustainable development that acknowledges the complex interdependencies between the goals and the variable capacities of countries to reach them. For this our machine learning based approach provides a robust framework for developing efficient and data-informed strategies to promote cooperative and targeted initiatives for sustainable progress.

Diabetic retinopathy (DR) is a sight-threatening condition caused by diabetes. Screening programmes for DR include eye examinations, where the patient’s fundi are photographed, and the findings, including DR severity, are recorded in the medical report. However, statistical analyses based on DR severity require structured labels that calls for laborious manual annotation process if the report format is unstructured. In this work, we propose a large language model DR-GPT for classification of the DR severity from unstructured medical reports. On a clinical set of medical reports, DR-GPT reaches 0.975 quadratic weighted Cohen’s kappa using truncated Early Treatment Diabetic Retinopathy Study scale. When DR-GPT annotations for unlabeled data are paired with corresponding fundus images, the additional data improves image classifier performance with statistical significance. Our analysis shows that large language models can be applied for unstructured medical report databases to classify diabetic retinopathy with a variety of applications.

Digital twins offer a new and exciting framework that has recently attracted significant interest in fields such as oncology, immunology, and cardiology. The basic idea of a digital twin is to combine simulation and learning to create a virtual model of a physical object. In this paper, we explore how the concept of digital twins can be generalized into a broader, overarching field. From a theoretical standpoint, this generalization is achieved by recognizing that the duality of a digital twin fundamentally connects complexity science with data science, leading to the emergence of complexity data science as a synthesis of the two. We examine the broader implications of this field, including its historical roots, challenges, and opportunities.

BACKGROUND:Radiotherapy is a core treatment modality for oropharyngeal cancer (OPC), where the primary gross tumor volume (GTVp) is manually segmented with high interobserver variability. This calls for reliable and trustworthy automated tools in clinician workflow. Therefore, accurate uncertainty quantification and its downstream utilization is critical. METHODS:Here we propose uncertainty-aware deep learning for OPC GTVp segmentation, and illustrate the utility of uncertainty in multiple applications. We examine two Bayesian deep learning (BDL) models and eight uncertainty measures, and utilize a large multi-institute dataset of 292 PET/CT scans to systematically analyze our approach. RESULTS:We show that our uncertainty-based approach accurately predicts the quality of the deep learning segmentation in 86.6% of cases, identifies low performance cases for semi-automated correction, and visualizes regions of the scans where the segmentations likely fail. CONCLUSIONS:Our BDL-based analysis provides a first-step towards more widespread implementation of uncertainty quantification in OPC GTVp segmentation.

We studied residential clustering and mobility of ethnic minorities using a theoretical framework based on null models of spatial distributions and movements of populations. Using microdata from population registers we compared the patterns of clustering amongst various socioethnic groups living in and around the capital region of Finland. Using the models we were able to connect the factors influencing intraurban migration to the spatial patterns that have been developed over time. We could also demonstrate the interrelationship of the movement and clustering with fertility. The observed clustering seems to be a combined effect of fertility and the tendency to migrate locally. The models also highlight the importance of factors like proximity to the city-centre, average neighbourhood income, and similarity of socioeconomic profiles.

Epidemics, apart from affecting the health of populations, can have large impacts on their social and economic behavior and subsequently feed back to and influence the spreading of the disease. This calls for systematic investigation which factors affect significantly and either beneficially or adversely the disease spreading and regional socio-economics. Based on our recently developed hybrid agent-based socio-economy and epidemic spreading model we perform extensive exploration of its six-dimensional parameter space of the socio-economic part of the model, namely, the attitudes towards the spread of the pandemic, health and the economic situation for both, the population and government agents who impose regulations. We search for significant patterns from the resulting simulated data using basic classification tools, such as self-organizing maps and principal component analysis, and we monitor different quantities of the model output, such as infection rates, the propagation speed of the epidemic, economic activity, government regulations, and the compliance of population on government restrictions. Out of these, the ones describing the epidemic spreading were resulting in the most distinctive clustering of the data, and they were selected as the basis of the remaining analysis. We relate the found clusters to three distinct types of disease spreading: wave-like, chaotic, and transitional spreading patterns. The most important value parameter contributing to phase changes and the speed of the epidemic was found to be the compliance of the population agents towards the government regulations. We conclude that in compliant populations, the infection rates are significantly lower and the infection spreading is slower, while the population agents’ health and economical attitudes show a weaker effect.

In face-to-face communication speech perception isboth visual and auditory.Under very noisy conditions the visual information from the talkers articulation can help to understand hardly audible speech.We have started developing a Finnish audio- visual speech synthesizer.In our current model, we have combined a three-dimen- sional facial model based on work by F. Parke (1982) with a commercial audio text-to-speech synthesizer.The visual speech is based on straightforward letter-to- viseme mapping, in which each letter of a written text corresponds to a viseme.Visual speech is animated by linear interpolation between visemes.Coarticulation has not been modeled.Audio-visual speech synthesizer can be used to prepare well-con- trolled stimuli for speech research and cognitive neuroscience.In addition, various application areas will benefit of high-quality audio-visual speech synthesis includ- ing telecommunication.human