OBJECTIVE: Prior studies often examine single telehealth encounter types or aggregate all digital care, overlooking how patients combine multiple digital and in-person modalities in hybrid care. To address this gap, we derived hybrid care engagement phenotypes and assessed sociodemographic differences and associations with glycemic control among adults with type 2 diabetes (T2DM). METHODS: We conducted a retrospective cohort study of 10 671 adults with T2DM receiving primary care at an academic (UCSF) or safety-net system (SFHN) from April 2021 to March 2023. K-medoids clustering was applied to five encounter modalities (in-person, video, telephone visits; portal messages; unscheduled telephone calls) to derive four engagement phenotypes. We assessed sociodemographic differences using chi-square and Kruskal-Wallis tests and evaluated associations between phenotype and follow-up HbA1c control using logistic regression. We tested interactions with baseline HbA1c and estimated predicted probabilities using Tukey-adjusted contrasts. RESULTS: Four phenotypes emerged per system: Digitally Engaged Multimodal, Traditional High Utilizers, Digitally Leaning (UCSF), Telephone Reliant (SFHN), and Low Digital. UCSF patients belonged to digitally forward phenotypes, whereas SFHN patients concentrated in traditional, lower-tech phenotypes. Among patients with uncontrolled diabetes, digitally forward phenotypes had 13-20 percentage points higher predicted probability of achieving control (UCSF: 56% Digitally Leaning vs 36% Traditional; SFHN: 53% Multimodal vs 40% Telephone). DISCUSSION: Phenotypes varied by health system and sociodemographic factors, with modest, system-specific associations between digitally forward phenotypes and glycemic control among patients with uncontrolled diabetes. Findings underscore structural and sociodemographic inequities in hybrid care engagement and the need for proactive, tailored strategies to promote equitable hybrid care.

In this brief, a high-speed [multiway parallel (MWP)] hardware-implemented deflate data compressor (DDC) is proposed for reducing the storage of solid-state drives (SSDs). To minimize the area of the DDC, registers instead of static random access memories (SRAMs) are utilized for building hash tables because multiway data within the DDC are able to access a register-based hash table simultaneously. To further reduce the area of the DDC, the output data of indefinite length are concatenated with a tree-type hardware architecture for reducing the overall concatenation complexity. Moreover, a solid mathematical foundation is established for optimizing the latency values of Lempel–Ziv (LZ)77 circuit, the Huffman encoding circuit, and the output data concatenation circuit within the MWP DDC. The results show that the proposed MWP DDC is capable of achieving a 12.1-Gb/s throughput and a 1.76 compression ratio (CR) with a 1.17-mm2area and 0.103-$\mu $s latency, under the synthesis of SMIC 55-nm process design kits (PDKs). Hence, the proposed DDC satisfies the SSD compression requirement for a universal serial bus (USB) 3.2 connector.

Smart devices often use natural language understanding (NLU) to infer user goals. NLU works by extracting user intents from text, then mapping these intents to rigid sets of rules that specify system responses (i.e., slot filling). NLU behaves best when intents and their targets are well-specified (e.g., "set brightness to 50%"), and when the rules implement all desirable responses. NLU falters, however, when responses demand a deeper level of reasoning than predefined rules can provide, e.g., when intents are under-specified (e.g., "make a warm glow") or context-sensitive information is requested (e.g., "why is it so hot in here?"). Toward more flexible user-device interactions, we re-frame the challenge as one of building thoughtful things: devices that leverage a local capacity for reasoning about their internal states to generate responses rather than retrieving them from constrained rule sets. We propose an end-to-end framework for "teaching things to think" that trains a language model to generate valid, device-specific actions and natural language explanations in response to unconstrained language that is out-of-scope for slot filling NLU. Our key insight is that we can bootstrap this generative model from purely synthetic data constructed using a formal definition of a device’s state. The resulting model is small enough to run on-device, with no runtime cloud dependency. We implement two thoughtful things (a lamp and a thermostat) on real hardware and explore their potential to generate and explain device states in response to commands and inquiries.

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.

This paper introduces a novel hybrid simulation system (HybriSim) tailored for simulating distributed learning in mobile settings, such as those involving vehicles and pedestrians navigating through cities. Designed to be learning-method independent, the system is compatible with decentralized learning, federated learning, or a combination of the two. It has special relevance for decentralized learning systems that are sensitive to mobility patterns and rely on direct, device-to-device communication. Existing tools for evaluating resource-intensive tasks in opportunistic networks are either purely simulated, which may not accurately reflect system performance, or take the form of testbeds of real devices, which are difficult to scale to use cases involving huge numbers of devices, such as distributed learning. By integrating real devices with virtual simulated devices, HybriSim more accurately mirrors real-world performance and dynamics. This integration not only mitigates the biases associated with pure simulations but also resolves the deployment complexities of conducting simulations entirely on real devices. Our system sets a new benchmark for academic and industry researchers, facilitating more reliable and actionable insights into distributed learning systems in mobility contexts.

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 development and usage of Internet-of-Things (IoT) and smart-home devices, researchers continue efforts to improve the "smartness" of those devices to address daily needs in people's lives. Such efforts usually begin with understanding evolving user behaviors on how humans utilize the devices and what they expect in terms of their behavior. However, while research efforts abound, there is a very limited number of datasets that researchers can use to both understand how people use IoT devices and to evaluate algorithms or systems for smart spaces. In this paper, we collect and characterize more than 50,000 recipes from the online If-This-Then-That (IFTTT) service to understand a seemingly straightforward but complicated question: "What kinds of behaviors do humans expect from their IoT devices?" The dataset we collected contains the basic information of the IFTTT rules, trigger and action events, and how many people are using each rule.