Large Language Models (LLMs) excel at general-purpose reasoning by leveraging broad commonsense knowledge, but they remain limited in tasks requiring personalized reasoning over multifactorial personal data. This limitation constrains their applicability in domains such as healthcare, where decisions must adapt to individual contexts. We introduce Personalized Causal Graph Reasoning, a framework that enables LLMs to reason over individual-specific causal graphs constructed from longitudinal data. Each graph encodes how user-specific factors influence targeted outcomes. In response to a query, the LLM traverses the graph to identify relevant causal pathways, rank them by estimated impact, simulate potential outcomes, and generate tailored responses. We implement this framework in the context of nutrient-oriented dietary recommendations, where variability in metabolic responses demands personalized reasoning. Using counterfactual evaluation, we assess the effectiveness of LLM-generated food suggestions for glucose control. Our method reduces postprandial glucose iAUC across three time windows compared to prior approaches. Additional LLM-as-a-judge evaluations further confirm improvements in personalization quality.

Health monitoring systems have revolutionized mod-ern healthcare by enabling the continuous capture of physio-logical and behavioral data, essential for preventive measures and early intervention. Integrating this data with Large Lan-guage Models (LLMs) shows promise in delivering interactive health advice, but traditional methods like Retrieval-Augmented Generation (RAG) and fine-tuning often fail to fully utilize the complex, multi-dimensional data from wearable devices. These approaches typically provide limited actionable and personalized health insights due to their inadequate capacity to dynamically integrate and interpret diverse health data streams. This pa-per introduces a graph-augmented LLM framework designed to enhance the personalization and clarity of health insights. Utilizing a hierarchical graph structure, the framework captures inter and intra-patient relationships, enriching LLM prompts with feature importance scores from a Random Forest Model. The effectiveness of this approach is demonstrated through a sleep analysis case study involving 20 college students during the COVID-19 lockdown, highlighting the potential of our model to generate actionable and personalized health insights efficiently. We leverage another LLM to evaluate the insights for relevance, comprehensiveness, actionability, and personalization. Our findings show that augmenting prompts with our framework yields significant improvements in all four criteria, eliciting well-crafted, thoughtful responses tailored to a specific patient.

Wearable technology enables ubiquitous health monitoring where multivariate physiological and behavioral data can be captured over time. Such multivariate time series (MTS) data in healthcare applications needs technique to interpret the analysis results. However, existing deep learning models for MTS data analysis often lack interpretability, and current explainable AI (xAI) techniques fail to capture the temporal and inter-variable complexities inherent in MTS. This hinders the trust and integration of these AI-based systems in clinical decision-making. In this paper, we propose an attention-based xAI method to classify and interpret MTS data collected from wearable devices. Our approach leverages self-attention mechanisms and graph attention layers (GAT) to capture both temporal and inter-variable dependencies, providing interpretability at both the temporal and modality levels. We evaluate our method using a longitudinal affect status monitoring. The dataset was collected from 21 college students via wearable devices over one year. We train separate models for positive (PA) and negative affect (NA) prediction, and compare their performance with a Transformer-based method. Our method achieves robust classification performance, with 78.62% accuracy for PA and 76.30% for NA, while offering transparent explanations of its decisions. These findings highlight the potential of our xAI method for reliable and interpretable MTS classification in healthcare applications.

The adverse effects of loneliness on both physical and mental well-being are profound. Although previous research has utilized mobile sensing techniques to detect mental health issues, few studies have utilized state-of-the-art wearable devices to forecast loneliness and comprehend the physiological manifestations of loneliness and its predictive nature. The primary objective of this study is to examine the feasibility of forecasting loneliness by employing wearable devices, such as smart rings and watches, to monitor early physiological indicators of loneliness. Furthermore, smartphones are employed to capture initial behavioral signs of loneliness. To accomplish this, we employed personalized machine learning techniques, leveraging a comprehensive dataset comprising physiological and behavioral information obtained during our study involving the monitoring of college students. Through the development of personalized models, we achieved a notable accuracy of 0.82 and an F-1 score of 0.82 in forecasting loneliness levels seven days in advance. Additionally, the application of Shapley values facilitated model explainability. The wealth of data provided by this study, coupled with the forecasting methodology employed, possesses the potential to augment interventions and facilitate the early identification of loneliness within populations at risk.

Influence maximization (IM) is a problem of selecting k nodes from social networks to make the expected number of the active node maximum. Recently, with the popularity of Internet technology, more and more researchers have paid attention to this problem. However, the existing influence maximization algorithms with high accuracy are usually difficult to be applied to the large-scale social network. To solve this problem the paper proposes a new algorithm, called community-based influence maximization (ComIM). Its core idea is “divide and conquer”. In detail, this algorithm first utilizes the Louvain algorithm to divide the large-scale networks into some small-scale networks. Afterwards, the algorithm utilizes the one-hop diffusion value (ODV) and two-hop diffusion value (TDV) functions to calculate the influence of a node and select nodes on these small-scale networks, which can improve the accuracy of our proposed algorithm. By using the above methods, the paper proposes a community influence-estimating method called CDV, which can improve the efficiency of the algorithm. Experimental results on six real-world datasets demonstrate that our proposed algorithm outperforms all comparison algorithms when comprehensively considering the accuracy and efficiency.

Recovery of a 3D head model including the complete face and hair regions is still a challenging problem in computer vision and graphics. In this paper, we consider this problem using only a few multi-view portrait images as input. Previous multi-view stereo methods that have been based, either on optimization strategies or deep learning techniques, suffer from low-frequency geometric structures such as unclear head structures and inaccurate reconstruction in hair regions. To tackle this problem, we propose a prior-guided implicit neural rendering network. Specifically, we model the head geometry with a learnable signed distance field (SDF) and optimize it via an implicit differentiable renderer with the guidance of some human head priors, including the facial prior knowledge, head semantic segmentation information and 2D hair orientation maps. The utilization of these priors can improve the reconstruction accuracy and robustness, leading to a high-quality integrated 3D head model. Extensive ablation studies and comparisons with state-of-the-art methods demonstrate that our method can generate high-fidelity 3D head geometries with the guidance of these priors.

Refractive errors, such as myopia and astigmatism, can lead to severe visual impairment if not detected and corrected in time. Traditional methods of refractive error diagnosis rely on well-trained optometrists operating expensive and importable devices, constraining the vision screening process. Advance in smartphone camera has enabled novel low-cost ubiquitous vision screening to detect refractive error or ametropia through eye image processing, based on the principle of photorefraction. However, contemporary smartphone-based methods rely heavily on hand-crafted features and sufficiency of well-labeled data. To address these challenges, this paper exploits active learning methods with a set of Convolutional Neural Network features encoding information of human eyes from pre-trained gaze estimation model. This enables more effective training on refractive error detection models with less labeled data. Our experimental results demonstrate the encouraging effectiveness of our active learning approach. The new set of features is able to attain screening accuracy of more than 80% with mean absolute error less than 0.66, meeting the expectation of optometrists for 0.5 to 1. The proposed active learning also requires significantly fewer training samples of 18% in achieving satisfactory performance.

Uncorrected refractive errors can lead to permanent debilitating eye conditions if not corrected in a timely manner. Contemporary diagnostic methods rely on the professional acumen of optometrists and the use of expensive devices, which may not be easily accessible to all. According to the optical principle of photorefraction, refractive error can be estimated based on a relative pupil and crescent size of an eye image taken by a camera from a specified working distance. A low-cost approach would be to leverage smartphones with cameras for this purpose. However, the poor image quality generated from basic smartphones poses a challenge for the current approach as they often fail to accurately distinguish the crescent from the iris. We propose a novel method to detect and accurately measure the iris and crescent from smartphone photos. Based on this method, we further propose a set of features for machine learning to build our refractive error estimation model. The performance of our models are evaluated in an in-depth experiment.