For a wheelchair user, a standard blue line on a map is often a broken promise. While platforms like OpenStreetMap (OSM) successfully capture where a path is, they frequently fail to convey how it physically feels to travel on it. This information barrier is problematic for wheelchair users. To solve this issue, we present OmniPath, a system that moves from passive mapping to proactive environmental auditing. Our framework fuses the network topology of OSM with the submeter precision of high-density aerial LiDAR (USGS 3DEP) to create a high-fidelity 3D model of the pedestrian environment. Rather than simply routing a user, our agent virtually traverses the network, analyzing the surface in 0.5 meter increments. It rigorously quantifies physical friction points specifically running slope, cross slope, and vertical discontinuities against ADA compliance standards, calculating a weighted severity score to categorize hazards from ``Mild'' to ``Critical.'' To ensure real world reliability, we validated the system against 200 physical ground truth field surveys across the National Mall using stratified random sampling. The framework demonstrated strong diagnostic reliability for high-severity hazards, achieving F1-scores of 0.60 for Severe and 0.58 for critical categories. By automating this micro-scale inspection, OmniPath identifies the ``invisible'' barriers that standard maps miss, effectively transforming a static dataset into accessibility data source that anticipates accessibility challenges before the user ever leaves home.

Delirium occurs in about 80% of cases in the Intensive Care Unit (ICU) and is associated with an extended hospital stay, increased mortality, and other complications. Delirium lacks biomarker-based diagnosis and is frequently treated with antipsychotic drugs (APD), despite numerous studies debating its efficacy. Since randomized controlled trials (RCT) are expensive and time-consuming, we approach the research question of estimating the efficacy and safety outcomes of APD in treating delirium through retrospective cohort analysis. We employed the Causal inference framework to explore the underlying causal model for Delirium patient cohort. We focus on building a structural causal model for delirium in the ICU using large observational data sets linking various delirium-related covariates. We utilized an extensive electronic health records (EHR) dataset (MIMIC-III) to curate delirium data cohort. Our null hypothesis examines any significant differences in outcomes (30-day mortality and ICU length of stay) among delirium patients under different drug-groups (Haloperidol, other drugs, and no drugs). Our causal exploration presents a specialized pipeline through causal model generation, expert knowledge augmentation and average treatment effect estimation. Through our exploratory, machine learning driven, and causal analysis, we estimate and compare effects of antipsychotic drug groups on patients’ survival timeline and ICU length-of-stay.

Causal inference involving Structural causal models (SCMs) provides a principled approach to identifying causation from observational and experimental data in disciplines ranging from economics to medicine. However, to estimate the underlying causal structure, SCMs need to rely on domain knowledge in addition to available data. Clinical research has a vast collection of well-explored hypotheses, experiments, and publications, rich with underused causal information. A key challenge in this context is the absence (or acceptance) of a systematic and methodological framework for encoding priors (background knowledge) into causal models. We propose an abstraction called causal knowledge hierarchy (CKH) for encoding priors into causal models. Our approach is based on the foundation of "levels of evidence" in medicine, with a focus on confidence in causal information. Using CKH, we present a standardized framework for encoding causal priors from various information sources and combining them to derive an SCM. We evaluate our approach on multiple (simulated and real-world) benchmark datasets and demonstrate overall performance compared to the ground truth causal model.

Causal modeling offers a principled foundation for uncovering stable, invariant relationships in time-series data, thereby improving robustness and generalization under distribution shifts. Yet its potential is underutilized in spatiotemporal Earth observation, where models often depend on purely correlational features that fail to transfer across heterogeneous domains. We propose RIC-TSC, a regionally-informed causal time-series classification framework that embeds lag-aware causal discovery directly into sequence modeling, enabling both predictive accuracy and scientific interpretability. Using multi-modal satellite and reanalysis data—including Sentinel-1 microwave backscatter, Sentinel-2 and Landsat-8 optical reflectance, and CARRA meteorological variables—we leverage Joint PCMCI+ (J-PCMCI+) to identify region-specific and invariant predictors of supraglacial lake evolution in Greenland. Causal graphs are estimated globally and per basin, with validated predictors and their time lags supplied to lightweight classifiers. On a balanced benchmark of 1000 manually labeled lakes from two contrasting melt seasons (2018–2019), causal models achieve up to 12.59% higher accuracy than correlation-based baselines under out-of-distribution evaluation. These results show that causal discovery is not only a means of feature selection but also a pathway to generalizable and mechanistically grounded models of dynamic Earth surface processes.

Robust causal discovery in time series datasets depends on reliable benchmark datasets with known ground-truth causal relationships.However, such datasets remain scarce, and existing synthetic alternatives often overlook critical temporal properties inherent in real-world data, including nonstationarity driven by trends and seasonality, irregular sampling intervals, and the presence of unobserved confounders.To address these challenges, we introduce TimeGraph, a comprehensive suite of synthetic time-series benchmark datasets that systematically incorporates both linear and nonlinear dependencies while modeling key temporal characteristics such as trends, seasonal effects, and heterogeneous noise patterns.Each dataset is accompanied by a fully specified causal graph featuring varying densities and diverse noise distributions and is provided in two versions: one including unobserved confounders and one without, thereby offering extensive coverage of real-world complexity while preserving methodological neutrality.We further demonstrate the utility of TimeGraph through systematic evaluations of state-of-the-art causal discovery algorithms including PCMCI+, LPCMCI, and FGES across a diverse array of configurations and metrics.Our experiments reveal significant variations in algorithmic performance under realistic temporal conditions, underscoring the need for robust synthetic benchmarks in the fair and transparent assessment of causal discovery methods.The complete TimeGraph suite, including dataset generation scripts, evaluation metrics, and recommended experimental protocols, is freely available to facilitate reproducible research and foster community-driven advancements in time-series causal discovery.Source code is available at https://github.com/hferdous/TimeGraph.

Accessible navigation for wheelchair users remains a significant challenge due to obstacles such as uneven sidewalks, steep inclines, and the lack of access ramps. Existing routing systems primarily focus on minimizing travel time or distance, often failing to consider mobility constraints. This paper presents MyPath, an end-to-end accessible navigation system that integrates user-reported data, machine learning, and programmatic analysis to optimize routing for wheelchair users. Additionally, MyPath provides detailed, step-by-step navigation instructions. By unifying these elements within a mobile application, My-Path provides a real-time, personalized, and scalable solution for accessible routing. Our contributions include (1) a novel hybrid routing framework integrating real-world feedback and algorithmic approach, (2) automated detection of environmental barriers, and (3) Step by step detailed navigation. The proposed system has broad implications for urban accessibility, smart mobility, and inclusive navigation solutions.

Wheelchair and mobility aid users often face challenges in navigating the built environment due to uneven sidewalks, temporary barriers, steep inclines, and narrow lanes. To assist these users, accessible routing systems have been introduced that generate wheelchair-accessible paths to facilitate navigation in unfamiliar environments. In general, accessible routing systems rely on surface and path characteristics like surface type, incline, width, etc., and crowd-sourced information about barriers to provide the optimal route. Emerging routing systems even provide personalized routing to users that are catered to the user's specific needs and requirements. However, these types of systems collect crowd-sourced personal/identifiable information which introduces privacy and data heterogeneity concerns that are not addressed by them or elsewhere in the concerned domain. To address these two issues specifically, we propose the novel FedAccess system for accessible routing that utilizes the federated learning paradigm for surface recognition using vibration data. The surface-induced vibrations are captured through smartphone-embedded motion sensors (accelerometers and gyroscopes) from 23 manual wheelchair users during their regular navigation. We have covered 10 distinct surfaces from the USA. As a result, the distribution of the data is naturally non-IID. Empirical evaluation shows that the FedAccess system can protect user data and identity while dealing with non-IID data and still recognize heterogeneous surfaces with higher accuracy than the state-of-the-art.

The Greenland Ice Sheet (GrIS) has emerged as a significant contributor to global sea level rise, primarily due to increased meltwater runoff. Supraglacial lakes, which form on the ice sheet surface during the summer months, can impact ice sheet dynamics and mass loss; thus, better understanding these lakes' seasonal evolution and dynamics is an important task. This study presents a computation-ally efficient time series classification approach that uses Gaussian Mixture Models (GMMs) of the Reconstructed Phase Spaces (RPSs) to identify supraglacial lakes based on their seasonal evolution: 1) those that refreeze at the end of the melt season, 2) those that drain during the melt season, and 3) those that become buried, remaining liquid insulated a few meters beneath the surface. Our approach uses time series data from the Sentinel-l and Sentinel-2 satellites, which utilize microwave and visible radiation, respectively. Evaluated on a GrIS-wide dataset, the RPS-GMM model, trained on a single representative sample per class, achieves 85.46% accuracy with Sentinel-l data alone and 89.70% with combined Sentinel-l and Sentinel-2 data. This performance significantly surpasses existing machine learning and deep learning models which require a large training data. The results demonstrate the robustness of the RPS-GMM model in capturing the complex temporal dynamics of supraglaciallakes with minimal training data.

Conventional temporal causal discovery (CD) methods suffer from high dimensionality, fail to identify lagged causal relationships, and often ignore dynamics in relations. In this study, we present a novel constraint-based CD approach for autocorrelated and non-stationary time series data (eCDANs) capable of detecting lagged and contemporaneous causal relationships along with temporal changes. eCDANs addresses high dimensionality by optimizing the conditioning sets while conducting conditional independence (CI) tests and identifies the changes in causal relations by introducing a surrogate variable to represent time dependency. Experiments on synthetic and real-world data show that eCDANs can identify time influence and outperform the baselines.

Learning High-Resolution representations is essential for semantic segmentation. Convolutional neural network (CNN) architectures with downstream and upstream propagation flow are popular for segmentation in medical diagnosis. However, due to performing spatial downsampling and upsampling in multiple stages, information loss is inexorable. On the contrary, connecting layers densely on high spatial resolution is computationally expensive. In this work, we devise a Loose Dense Connection Strategy to connect neurons in subsequent layers with reduced parameters. On top of that, using a m-way Tree structure for feature propagation we propose Receptive Field Chain Network (RFC-Net) that learns high-resolution global features on a compressed computational space. Our experiments demonstrates that RFC Net achieves state-of-the-art performance on Kvasir and CVC-ClinicDB benchmarks for Polyp segmentation. Our code is publicly available at github.com/sourajitcs/RFC-NetAAAI23.

The warming of the Arctic, also known as Arctic amplification, is led by several atmospheric and oceanic drivers. However, the details of its underlying thermodynamic causes are still unknown. Inferring the causal effects of atmospheric processes on sea ice melt using fixed treatment effect strategies leads to unrealistic counterfactual estimations. Such methods are also prone to bias due to time-varying confoundedness. Further, the complex non-linearity in Earth science data makes it infeasible to perform causal inference using existing marginal structural techniques. In order to tackle these challenges, we propose TCINet - Time-series Causal Inference Network to infer causation under continuous treatment using recurrent neural networks and a novel probabilistic balancing technique. More specifically, we propose a neural network based potential outcome model using the long-short-term-memory (LSTM) layers for time-delayed factual and counterfactual predictions with a custom weighted loss. To tackle the confounding bias, we experiment with multiple balancing strategies, namely TCINet with the inverse probability weighting (IPTW), TCINet with stabilized weights using Gaussian Mixture Model (GMMs) and TCINet without any balancing technique. Through experiments on synthetic and observational data, we show how our research can substantially improve the ability to quantify leading causes of Arctic sea ice melt, further paving paths for causal inference in observational Earth science.

Despite the active legislation in many countries supporting the accessibility of public spaces by mobility-impaired users, the reality is far from ideal. Wheelchair users often struggle to navigate the built environment let alone the natural areas. While barriers to wheeled mobility can be caused by broken/uneven surfaces, steep slopes, and unfavorable weather conditions, the effects of many such factors and others are not properly investigated. In this paper, we aim to classify various built and natural surfaces through their characteristic vibration patterns using different deep learning algorithms. The surface vibration data is collected from various cities in Europe (including Paris (FR), Mannheim (DE), Dresden (DE), Munich, Nuremberg (DE), and Salzburg (AT)) while a user drives a manual wheelchair attached with three differently oriented smartphones placed at different heights. Extensive experiments show that a Deep Neural Network model classifies surfaces using a denoised dataset with a 98.9% accuracy which is significantly higher than our previous state-of-the-art.

The nature of the surface has a significant effect on how wheelchair users experience locomotion. The preferred surfaces for wheeled mobility must be even, firm and smooth while generating adequate friction. The development of accessible road maps that include ground conditions is therefore of utmost importance. Our prior work has shown how such maps can be created using surface-induced vibration data collected by motion sensors embedded in smartphones and then classifying them with machine learning algorithms. To make data collection scalable, participatory crowd-sensing can be used, where users collect and transmit sensor data while traveling on wheelchairs. The complexity here is that wheelchairs widely vary in type (manual, power-assist, power), weight, number and nature of wheels, therefore the sensor data generated by different wheelchairs varies greatly. Collecting training data on each individual wheelchair type to develop classification models is not feasible. To address this problem, in this paper we explore the possibility of transferring knowledge from known wheelchairs to unknown types. We develop a transfer learning algorithm to classify 15 surfaces with minimal training data from different wheelchairs. Our experiments with 47 subjects show that surface classification knowledge, learned from sensor data generated by manual wheelchairs, can be transferred to a power wheelchair with up to 90.02% accuracy. This allows crowd-sensing to be used effectively for data collection for generating accessible route maps. We integrate our transfer learning approach into our system for accessible routing, which we developed in previous work.

With the rapid growth of on-demand taxi services, like Uber, Lyft, etc., urban public transportation scenario is shifting towards a personalized transportation choice for most commuters. While taxi rides are comfortable and time efficient, they often lead to higher cost and road congestion due to lower overall occupancy than bigger vehicles. One efficient way to improve taxi occupancy is to adopt ride sharing. Existing ride sharing solutions are mostly centralized and proprietary. Moreover, given the wide spatio-temporal variation of incoming ride requests designing a dynamic and distributed shared-ride scheduling system is NP-hard. In this paper, we have proposed a publisher (passengers) and subscriber (taxis) based ride sharing system that provides effective real-time ride scheduling for multiple passengers. A particle swarm based route optimization strategy has been applied to determine the most preferable route for passengers. Empirical analysis using large scale single-user taxi ride records from Chicago Transit Authority, show that, our proposed system, ensures a maximum of 91.74% and 63.29% overall success rates during non-peak and peak hours, respectively.

Use of mobile health (mHealth) systems has increased with the advancement and proliferation of mobile computing and related technologies. It has improved efficiency and effectiveness of healthcare services. Non-invasive pain level detection from facial images is one of the promising mHealth applications. Effective pain treatment requires regular and continuous pain assessment. Most of the pain research tools are study or disease specific while some are pain (lumbar pain, cancer pain, etc.) and patient group specific (neonatal, adult, woman, etc.). This results in recurrent but potentially avoidable costs such as time, money, and workforce to develop similar services or software research tools for each research study. In this study, we have proposed, designed, and implemented a customizable personalized pain study platform that offers real-time data collection, research participant management, role-based access control, research data anonymization etc. It is also used to investigate pain level detection accuracy using evidence-based continuous learning from the facial expression data, collected from Bangladesh, Nepal and USA, which yielded about 71% classification accuracy.

The growing popularity of shared transportation enables researchers to explore a wide range of competing solutions. These propositions, in turn, uncover new challenges. Continuous research is done to design systems that are easy to scale and sustain. This survey categorizes ride sharing systems broadly into static and dynamic models. It also presents the system architectures for dynamic ride sharing systems such as central, distributed and hybrid design. This paper outlines the various algorithm designs adopted by the researchers over the years as well as provides an insight into the recent research conducted in this field. It also serves as a guide for potential future research directions towards identified open challenges.

Hemoglobin level detection is necessary for evaluating health condition in the human. In the laboratory setting, it is detected by shining light through a small volume of blood and using a colorimetric electronic particle counting algorithm. This invasive process requires time, blood specimens, laboratory equipment, and facilities. There are also many studies on non-invasive hemoglobin level detection. Existing solutions are expensive and require buying additional devices. In this paper, we present a smartphone-based non-invasive hemoglobin detection method. It uses the video images collected from the fingertip of a person. We hypothesized that there is a significant relation between the fingertip mini-video images and the hemoglobin level by laboratory "gold standard." We also discussed other non-invasive methods and compared with our model. Finally, we described our findings and discussed future works.

Every year millions of people gather at Makkah, Saudi Arabia during the Hajj, an annual Islamic pilgrimage. The area at Makkah is small, and the number of attendees increases each year, which has created an ongoing and ever increasing problem of crowd management. In this paper, we present our integrated solution to the localization challenge of tracking specific users in a highly crowded area where GPS signal may be weak or even unavailable. Smartphone based Human Activity Recognition (HAR) uses various sensors that are built into the smartphone to sense a person's activity in real time. Applications that incorporate HAR can be used to track a person's movements and are very useful in areas such as health care. We also propose a group-tracking mechanism that can be applied when a group member appears to get lost. Other members of the group will be immediately notified and receive an estimation of the lost member's location. Using wireless signals (RSSI) and inertial sensor data, we have developed a mathematical model and a system for both outdoor and indoor localization. The experimental results show that the proposed system is able to detect locations of users with high accuracy, with an error of less than 2.5 meters. The system will be used by millions of users in Makkah, where there have been thousands of reported cases of pilgrims getting lost during the Hajj, however, it is scalable to accommodate any other crowded population.

Nowadays with the dispersion of wireless networks, smartphones and diverse related services, different localization techniques have been developed. Global Positioning System (GPS) has a high rate of accuracy for outdoor localization but the signal is not available inside of buildings. Also other existing methods for indoor localization have low accuracy. In addition, they use fixed infrastructure support. In this paper, we present a novel system for indoor localization, which also works well outside. We have developed a mathematical model for estimating location (distance and direction) of a mobile device using wireless technology. Our experimental results on Smartphones (Android and iOS) show good accuracy (an error less than 2.5 meters). We have also used our developed system in asset tracking and complex activity recognition.