Lead contamination in urban water systems remains a prevalent public health threat, affecting millions of American households and disproportionately endangering vulnerable population groups. Current municipal risk assessment and inspection strategies are overwhelmingly based on random sampling and complaint-driven protocols that overlook spatial complexity, reinforce inequities, and squander limited resources, leaving critical exposure areas unidentified. This paper presents a lead contamination risk prediction framework from socio-demographic housing features analytics, first of its kind, by drawing on partially anonymized residential testing data as ground truth and applying graph neural networks alongside gradient-boosted ensembles. Specifically, our method integrates spatial Deep Graph Attention Networks classifiers to capture inter-neighborhood contamination dependencies, fuse demographic and spatial evidence, and produce interpretable risk scores. Those scores are actionable by municipal water authorities at the intra-neighborhood level. Through extensive experiments on newly constructed Chicago block-group level datasets, our framework achieves a balanced accuracy of 84.8% and reduces false positive lead contamination by up to 44% versus spatial-only baselines and 21% over current practice, without sacrificing recall on contaminated blocks. Our approach not only extends technical boundaries in spatial-ensemble learning and privacy-preserving urban health modeling, but also provides policymakers and public health officials with a means to assess and address contamination risks, supporting efforts to protect community health and safety.

We present a new semi-supervised domain adaptation framework that combines a novel auto-encoder-based domain adaptation model with a simultaneous learning scheme providing stable improvements over state-of-the-art domain adaptation models. Our framework holds strong distribution matching property by training both source and target auto-encoders using a novel simultaneous learning scheme on a single graph with an optimally modified MMD loss objective function. Additionally, we design a semi-supervised classification approach by transferring the aligned domain invariant feature spaces from source domain to the target domain. We evaluate on three datasets and show proof that our framework can effectively solve both fragile convergence (adversarial) and weak distribution matching problems between source and target feature space (discrepancy) with a high ‘speed’ of adaptation requiring a very low number of iterations.

We develop a Multimodal Spatiotemporal Neural Fusion network for MTL (MSNF-MTCL) to predict 5 important students' retention risks: future dropout, next semester dropout, type of dropout, duration of dropout and cause of dropout. First, we develop a general purpose multi-modal neural fusion network model MSNF for learning students' academic information representation by fusing spatial and temporal unstructured advising notes with spatiotemporal structured data. MSNF combines a Bidirectional Encoder Representations from Transformers (BERT)-based document embedding framework to represent each advising note, Long-Short Term Memory (LSTM) network to model temporal advising note embeddings, LSTM network to model students' temporal performance variables and students' static demographics altogether. The final fused representation from MSNF has been utilized on a Multi-Task Cascade Learning (MTCL) model towards building MSNF-MTCL for predicting 5 student retention risks. We evaluate MSNF-MTCL on a large educational database consists of 36,445 college students over 18 years period of time that provides promising performances comparing with the nearest state-of-art models. Additionally, we test the fairness of such model given the existence of biases.

Person re-identification is a critical privacy breach in publicly shared healthcare data. We investigate the possibility of a new type of privacy threat on publicly shared privacy insensitive large scale wearable sensing data. In this paper, we investigate user specific biometric signatures in terms of two contextual biometric traits, physiological (photoplethysmography and electrodermal activity) and physical (accelerometer) contexts. In this regard, we propose PhysioGait, a context-aware phyiological signal model that consists of a Multi-Modal Siamese Convolutional Neural Network (mmSNN) which learns the spatial and temporal information individually and performs sensor fusion in a Siamese cost with the objective of predicting a person's identity. We evaluated PhysioGait attack model using 4 real-time collected datasets (3 collected datasets and one publicly available data) and two combined datasets achieving 89% - 93% accuracy of re-identifying persons.

With the advancement of deep neural networks and computer vision-based Human Activity Recognition, employment of Point-Cloud Data technologies (LiDAR, mmWave) has seen a lot interests due to its privacy preserving nature. Given the high promise of accurate PCD technologies, we develop, PALMAR, a multiple-inhabitant activity recognition system by employing efficient signal processing and novel machine learning techniques to track individual person towards developing an adaptive multi-inhabitant tracking and HAR system. More specifically, we propose (i) a voxelized feature representation-based real-time PCD fine-tuning method, (ii) efficient clustering (DBSCAN and BIRCH), Adaptive Order Hidden Markov Model based multi-person tracking and crossover ambiguity reduction techniques and (iii) novel adaptive deep learning-based domain adaptation technique to improve the accuracy of HAR in presence of data scarcity and diversity (device, location and population diversity). We experimentally evaluate our framework and systems using (i) a real-time PCD collected by three devices (3D LiDAR and 79 GHz mmWave) from 6 participants, (ii) one publicly available 3D LiDAR activity data (28 participants) and (iii) an embedded hardware prototype system which provided promising HAR performances in multi-inhabitants (96%) scenario with a 63% improvement of multi-person tracking than state-of-art framework without losing significant system performances in the edge computing device.

Enabling mobile robots for solving challenging and diverse shape, texture, and motion related tasks with high fidelity vision requires the integration of novel multimodal imaging sensors and advanced fusion techniques. However, it is associated with high cost, power, hardware modification, and computing requirements which limit its scalability. In this paper, we propose a novel Simultaneously Learned Auto Encoder Domain Adaptation (SAEDA)-based transfer learning technique to empower noisy sensing with advanced sensor suite capabilities. In this regard, SAEDA trains both source and target auto-encoders together on a single graph to obtain the domain invariant feature space between the source and target domains on simultaneously collected data. Then, it uses the domain invariant feature space to transfer knowledge between different signal modalities. The evaluation has been done on two collected datasets (LiDAR and Radar) and one existing dataset (LiDAR, Radar and Video) which provides a significant improvement in quadruped robot-based classification (home floor and human activity recognition) and regression (surface roughness estimation) problems. We also integrate our sensor suite and SAEDA framework on two real-time systems (vacuum cleaning and Mini-Cheetah quadruped robots) for studying the feasibility and usability.

We develop CACE (Constraints And Correlations mining Engine), a framework that significantly improves the recognition accuracy of complex daily activities in multi-inhabitant smarthomes. CACE views the implicit relationships between the activities of multiple people as an asset, and exploits such constraints and correlations in a hierarchical fashion, taking advantage of both personspecific sensor data (generated by wearable devices) and person-independent ambient sensor data (generated by ambient sensors). To effectively utilize such couplings, CACE first uses a multi-target particle filtering approach over ambient sensors captured movement data, to identify the number of distinct users and infer individual-specific movement trajectories. We then utilize a Hierarchical Dynamic Bayesian Network (HDBN)-based model for activity recognition. This model utilizes the inter-and-intra individual correlations and constraints, at both micro-activity and macro-activity levels, to recognize individual activities accurately. These constraints are learnt automatically using data-mining techniques, and help to dramatically reduce the computational complexity of HDBN-based inferencing. Empirical studies using a real-world testbed of five multi-inhabitant smarthomes shows that CACE is able to achieve an activity recognition accuracy of 95%, with a 16-fold reduction in computational overhead compared to traditional hybrid classification approaches.

Wireless wearable sensor networks (WSN)-based activity recognition has implicit impacts on context-aware application and connected health research for older adults. Although, many existing researches focus on different WSN integration, signal processing and intelligent detection to recognize different activities, none of the existing works address the Artificial Intelligence (AI) fairness in activity recognition of older adults domain. We argue that AI-fairness towards detecting activities of older adults is different than to fairness for other protected attributes such as age, gender or race. The primary reason behind this difference is the diversity of same activity among different older adults based on their functional abilities (walking using crutches, walker). However, the diversity also exists among the same older adults based on time and space as well. The above constraints limit the AI capabilities and causes unfair detection of daily activities that has potential impacts on healthcare interventions for older adults. In this paper, we investigate, first of its kind, AI-fairness of activity recognition for older adults using a single wearable WSN sensor in presence of diverse disabilities. In this regard, we (i) employ signal processing and Bi-directional LSTM model to recognize diverse multi-label activities of older adults using single WSN; (ii) identify the existence of biases in activity recognition considering the age and functional ability as protected attributes; (iii) mitigate the biases by applying different bias mitigation techniques in different stages (pre-, in- and post-processing) of machine learning model development; and finally (iv) we experimentally evaluate the proposed AI-fairness framework using older adults data collected from a retirement center.

Human activity recognition (HAR) is a an important field of study concerned with identifying the specific movement or action of a person based on sensor data. Movements are often typical activities, such as walking, lying, standing, and sitting which has high significance in assessing health related vital signs. While accuracy of HAR been improved significantly using advanced deep learning techniques, employment of LiDAR (Light detection and Ranging) has been rarely explored due to the high cost of sensor device as well as the complexity of LiDAR point-cloud data. In this paper, we propose LAMAR (LiDAR-based Adaptive Multi-inhabitant Activity Recognition), a low-resolution low-cost (<300$) solid-state LiDAR-based multiple-inhabitant activity recognition system. LAMAR incorporates a series of existing signal processing techniques and a novel adaptive deep learning model to improve multi-inhabitant HAR performance. In this regard, at first, LAMAR posits a voxelized feature representation and clustering based multiple person tracking method to separate individual person related point-clouds. Then, it employes a novel variational autoencoder based domain adaptation technique to improve activity recognition performance by transferring domain knowledge from high resolution to low resolution LiDAR data. Our experimental evaluation on two datasets (one real-time collected data and another publicly available available data) provided significant improvement of HAR performance (94%) in multiple-inhabitant scenario with 63% improvement of multiple person tracking than existing work.

Cognitive impairment has become epidemic in older adult population. The recent advent of tiny wearable and ambient devices, a.k.a Internet of Things (IoT) provides ample platforms for continuous functional and cognitive health assessment of older adults. In this paper, we design, implement and evaluate AutoCogniSys, a context-aware automated cognitive health assessment system, combining the sensing powers of wearable physiological (Electrodermal Activity, Photoplethysmography) and physical (Accelerometer, Object) sensors in conjunction with ambient sensors. We design appropriate signal processing and machine learning techniques, and develop an automatic cognitive health assessment system in a natural older adults living environment. We validate our approaches using two datasets: (i) a naturalistic sensor data streams related to Activities of Daily Living and mental arousal of 22 older adults recruited in a retirement community center, individually living in their own apartments using a customized inexpensive IoT system (IRB #HP-00064387) and (ii) a publicly available dataset for emotion detection. The performance of AutoCogniSys attests max. 93% of accuracy in assessing cognitive health of older adults.

Veteran mental health is a significant national problem as large number of veterans are returning from the recent war in Iraq and continued military presence in Afghanistan. While significant existing works have investigated twitter posts-based Post Traumatic Stress Disorder (PTSD) assessment using blackbox machine learning techniques, these frameworks cannot be trusted by the clinicians due to the lack of clinical explainabil-ity. To obtain the trust of clinicians, we explore the big question, can twitter posts provide enough information to fill up clinical PTSD assessment surveys that have been traditionally trusted by clinicians? To answer the above question, we propose, LAXARY (Linguistic Analysis-based Exaplainable Inquiry) model, a novel Explainable Artificial Intelligent (XAI) model to detect and represent PTSD assessment of twitter users using a modified Linguistic Inquiry and Word Count (LIWC) analysis. First, we employ clinically validated survey tools for collecting clinical PTSD assessment data from real twitter users and develop a PTSD Linguistic Dictionary using the PTSD assessment survey results. Then, we use the PTSD Linguistic Dictionary along with machine learning model to fill up the survey tools towards detecting PTSD status and its intensity of corresponding twitter users. Our experimental evaluation on 210 clinically validated veteran twitter users provides promising accuracies of both PTSD classification and its intensity estimation. We also evaluate our developed PTSD Linguistic Dictionary's reliability and validity.

Autonomous longitudinal functional health assessment is critically important to support the rehabilitation of older adults in Skilled Nursing Facilities (SNFs). Although the wide availability of commodity smarthome sensors and internet-of-things (IoT) is facilitating continual monitoring of individuals' health-related vital signs and behaviors, missing values, presence of multi-inhabitants and diversity of smarthomes interfere with successful longitudinal assessment and impact the scalability of autonomous health assessments. In this paper, we propose a novel scalable framework to provide health assessments of older adults living in varied smarthome environments. As a critical first step, we propose a novel algorithm to track individuals in a multi-inhabitant smarthome environment. We then propose a novel data curation technique to address missing sensor signals in a multi-modal ambient sensor-assisted environment. Finally, we propose a novel trajectory featurization method inspired Deep Convolutional Neural Network TDCNN, leveraging appropriate samples from a well-labeled source smarthome, to transfer functional health assessment knowledge to unlabeled diverse smarthomes, boosting the scalability of autonomous health assessment. Our evaluation on real SNFs data, collected over 5 months from 95 individuals residing in 9 diverse sensored SNFs environments shows promising results (93% accuracy) with respect to the scalability of our framework.

Human activity recognition (AR) is an essential element for user-centric and context-aware applications. While previous studies showed promising results using various machine learning algorithms, most of them can only recognize the activities that were previously seen in the training data. We investigate the challenges of improving the recognition of unseen daily activities in smart home environment, by better exploiting the hierarchical taxonomy of complex daily activities. We first (a) design a hierarchical representation of complex activity taxonomy in terms of human-readable semantic attributes, and (b) develop a hierarchy of classifiers which incorporates a cluster tree built on the domain knowledge from training samples. Though this model is rich in recognizing complex activities that are previously seen in training data, it is not well versed to recognize unseen complex activities without new training samples. To tackle this challenge, we extend Hierarchical Active Transfer Learning (HATL) approach that exploits semantic attribute cluster structure of complex activities shared between seen (source) and unseen (target) activity domains. Our approach employs transfer and active learning to help label target domain unlabeled data by spawning the most effective queries. We evaluated our approach with two real-time smart home systems (IRB #HP-00064387) which corroborates radical improvements in recognizing unseen complex activities.

We propose CACE (Constraints And Correlations mining Engine) which investigates the challenges of improving the recognition of complex daily activities in multi-inhabitant smart homes, by better exploiting the spatiotemporal relationships across the activities of different individuals. We first propose and develop a loosely-coupled Hierarchical Dynamic Bayesian Network (HDBN), which both (a) captures the hierarchical inference of complex (macro-activity) contexts from lower-layer microactivity context (postural and improved oral gestural context), and (b) embeds the various types of behavioral correlations and constraints (at both micro-and macro-activity contexts) across the individuals. While this model is rich in terms of accuracy, it is computationally prohibitive, due to the explosive increase in the number of jointly-defined states. To tackle this challenge, we employ data mining to learn behaviorally-driven context correlations in the form of association rules, we then use such rules to prune the state space dramatically. To evaluate our framework, we build a customized smart home system and collected naturalistic multi-inhabitant smart home activities data. The system performance is illustrated with results from real-time system deployment experiences in a smart home environment reveals a radical (max 16 fold) reduction in the computational overhead compared to traditional hybrid classification approaches, as well as an improved activity recognition accuracy of max 95%.

To promote independent living for elderly population activity recognition based approaches have been investigated deeply to infer the activities of daily living (ADLs) and instrumental activities of daily living (I-ADLs). Deriving and integrating the gestural activities (such as talking, coughing, and deglutition etc.) along with activity recognition approaches can not only help identify the daily activities or social interaction of the older adults but also provide unique insights into their long-term health care, wellness management and ambulatory conditions. Gestural activities (GAs), in general, help identify fine-grained physiological symptoms and chronic psychological conditions which are not directly observable from traditional activities of daily living. In this paper, we propose GeSmart, an energy efficient wearable smart earring based GA recognition model for detecting a combination of speech and non-speech events. To capture the GAs we propose to use only the accelerometer sensor inside our smart earring due to its energy efficient operations and ubiquitous presence in everyday wearable devices. We present initial results and insights based on a C4.5 classification algorithm to infer the infrequent GAs. Subsequently, we propose a novel change-point detection based hybrid classification method exploiting the emerging patterns in a variety of GAs to detect and infer infrequent GAs. Experimental results based on real data traces collected from 10 users demonstrate that this approach improves the accuracy of GAs classification by over 23%, compared to previously proposed pure classification-based solutions. We also note that the accelerometer sensor based earrings are surprisingly informative and energy efficient (by 2.3 times) for identifying different types of GAs.