Parkinson's Disease (PD) is one of the most critical neurodegenerative diseases, yet there is no cure for it, and the state-of-the-art treatment is to slow its progression. Thus, the earlier a patient with PD is recognized, the better he can be treated. Our project joins the research effort that aims to support early PD diagnosis by designing a Virtual Reality (VR)-based system to monitor pupil diameter patterns as new biomarkers (e.g., Pupil Light Reflex and Task-evoked Pupil Response) and provide early warning of potential PD onset. A follow-up experiment with 55 participants shows that the accuracy of recognizing early PD from healthy controls could reach 0.8942. Our study shows early results of a promising research direction that leverages VR-based technology to non-intrusively recognize patterns and provide alerts to early PD patients who would otherwise not know their symptoms until much later.

Artificial intelligence (AI) technology has been increasingly used in the implementation of advanced Clinical Decision Support Systems (CDSS). Research demonstrated the potential usefulness of AI-powered CDSS (AI-CDSS) in clinical decision making scenarios. However, post-adoption user perception and experience remain understudied, especially in developing countries. Through observations and interviews with 22 clinicians from 6 rural clinics in China, this paper reports the various tensions between the design of an AI-CDSS system (“Brilliant Doctor”) and the rural clinical context, such as the misalignment with local context and workflow, the technical limitations and usability barriers, as well as issues related to transparency and trustworthiness of AI-CDSS. Despite these tensions, all participants expressed positive attitudes toward the future of AI-CDSS, especially acting as “a doctor’s AI assistant” to realize a Human-AI Collaboration future in clinical settings. Finally we draw on our findings to discuss implications for designing AI-CDSS interventions for rural clinical contexts in developing countries.

Chatbots systems, despite their popularity in today's HCI and CSCW research, fall short for one of the two reasons: 1) many of the systems use a rule-based dialog flow, thus they can only respond to a limited number of pre-defined inputs with pre-scripted responses; or 2) they are designed with a focus on single-user scenarios, thus it is unclear how these systems may affect other users or the community. In this paper, we develop a generalizable chatbot architecture (CASS) to provide social support for community members in an online health community. The CASS architecture is based on advanced neural network algorithms, thus it can handle new inputs from users and generate a variety of responses to them. CASS is also generalizable as it can be easily migrate to other online communities. With a follow-up field experiment, CASS is proven useful in supporting individual members who seek emotional support. Our work also contributes to fill the research gap on how a chatbot may influence the whole community's engagement.

Modeling the endpoint uncertainty of moving target selection with crossing is essential to understand factors such as speed-accuracy trade-off and interaction efficiency in crossing-based user interfaces with dynamic contents. However, there have been few studies looking into this research topic in the HCI field. This paper presents a Quaternary-Gaussian model to quantitatively measure the endpoint uncertainty in crossing-based moving target selection. To validate this model, we conducted an experiment with discrete crossing tasks on five factors, i.e., initial distance, size, speed, orientation, and moving direction. Results showed that our model fit the data of μ and σ accurately with adjusted R2 of 0.883 and 0.920. We also demonstrated the validity of our model in predicting error rates in crossing-based moving target selection. We concluded with a set of implications for future designs.

Providing users with rich sensations is beneficial to enhance their immersion in Virtual Reality (VR) environments. Wetness is one such imperative sensation that affects users' sense of comfort and helps users adjust grip force when interacting with objects. Researchers have recently begun to explore ways to create wetness illusions, primarily on a user's face or body skin. In this work, we extended this line of research by creating wetness illusion on users' fingertips. We first conducted a user study to understand the effect of thermal and tactile feedback on users' perceived wetness sensation. Informed by the findings, we designed and evaluated a prototype---Mouillé---that provides various levels of wetness illusions on fingertips for both hard and soft items when users squeeze, lift, or scratch it. Study results indicated that users were able to feel wetness with different levels of temperature changes and they were able to distinguish three levels of wetness for simulated VR objects. We further presented applications that simulated an ice cube, an iced cola bottle, and a wet sponge, etc, to demonstrate its use in VR.

Entering commands on touchscreens can be noisy, but existing interfaces commonly adopt deterministic principles for deciding targets and often result in errors. Building on prior research of using Bayes' theorem to handle uncertainty in input, this paper formalized Bayes' theorem as a generic guiding principle for deciding targets in command input (referred to as "BayesianCommand"), developed three models for estimating prior and likelihood probabilities, and carried out experiments to demonstrate the effectiveness of this formalization. More specifically, we applied BayesianCommand to improve the input accuracy of (1) point-and-click and (2) word-gesture command input. Our evaluation showed that applying BayesianCommand reduced errors compared to using deterministic principles (by over 26.9% for point-and-click and by 39.9% for word-gesture command input) or applying the principle partially (by over 28.0% and 24.5%).

Parkinson's disease (PD) is a chronic neurological disorder causing progressive disability that severely affects patients' quality of life. Although early interventions can provide significant benefits, PD diagnosis is often delayed due to both the mildness of early signs and the high requirements imposed by traditional screening and diagnosis methods. In this paper, we explore the feasibility and accuracy of detecting motor impairment in early PD via sensing and analyzing users' common touch gestural interactions on smartphones. We investigate four types of common gestures, including flick, drag, pinch, and handwriting gestures, and propose a set of features to capture PD motor signs. Through a 102-subject (35 early PD subjects and 67 age-matched controls) study, our approach achieved an AUC of 0.95 and 0.89/0.88 sensitivity/specificity in discriminating early PD subjects from healthy controls. Our work constitutes an important step towards unobtrusive, implicit, and convenient early PD detection from routine smartphone interactions.

Intensive exploration and navigation of hierarchical lists on smartphones can be tedious and time-consuming as it often requires users to frequently switch between multiple views. To overcome this limitation, we present PinchList, a novel interaction design that leverages pinch gestures to support seamless exploration of multi-level list items in hierarchical views. With PinchList, sub-lists are accessed with a pinch-out gesture whereas a pinch-in gesture navigates back to the previous level. Additionally, pinch and flick gestures are used to navigate lists consisting of more than two levels. We conduct a user study to refine the design parameters of PinchList such as a suitable item size, and quantitatively evaluate the target acquisition performance using pinch-in/out gestures in both scrolling and non-scrolling conditions. In a second study, we compare the performance of PinchList in a hierarchal navigation task with two commonly used touch interfaces for list browsing: pagination and expand-and-collapse interfaces. The results reveal that PinchList is significantly faster than other two interfaces in accessing items located in hierarchical list views. Finally, we demonstrate that PinchList enables a host of novel applications in list-based interaction?

Instant photo taking and sharing has become one of the most popular forms of social networking. However, taking high-quality photos is difficult as it requires knowledge and skill in photography that most non-expert users lack. In this paper we present SmartEye, a novel mobile system to help users take photos with good compositions in-situ. The back-end of SmartEye integrates the View Proposal Network (VPN), a deep learning based model that outputs composition suggestions in real time, and a novel, interactively updated module (P-Module) that adjusts the VPN outputs to account for personalized composition preferences. We also design a novel interface with functions at the front-end to enable real-time and informative interactions for photo taking. We conduct two user studies to investigate SmartEye qualitatively and quantitatively. Results show that SmartEye effectively models and predicts personalized composition preferences, provides instant high-quality compositions in-situ, and outperforms the non-personalized systems significantly.

Understanding the selection uncertainty of moving targets is a fundamental research problem in HCI. However, the only few works in this domain mainly focus on selecting 1D moving targets with certain input devices, where the model generalizability has not been extensively investigated. In this paper, we propose a 2D Ternary-Gaussian model to describe the selection uncertainty manifested in endpoint distribution for moving target selection. We explore and compare two candidate methods to generalize the problem space from 1D to 2D tasks, and evaluate their performances with three input modalities including mouse, stylus, and finger touch. By applying the proposed model in assisting target selection, we achieved up to 4% improvement in pointing speed and 41% in pointing accuracy compared with two state-of-the-art selection technologies. In addition, when we tested our model to predict pointing errors in a realistic user interface, we observed high fit of 0.94 R2.

The degree and quality of instructor-student interactions are crucial for students' engagement, retention, and learning outcomes. However, such interactions are limited in live online lectures, where instructors no longer have access to important cues such as raised hands or facial expressions at the time of teaching. As a result, instructors cannot fully understand students' learning progresses. This paper presents an explorative study investigating how presenters perceive and react to audience flow prediction when giving live-stream lectures, which has not been examined yet. The study was conducted with an experimental system that can predict audience's psychological states (e.g., anxiety, flow, boredom) through real-time facial expression analysis, and can provide aggregated views illustrating the flow experience of the whole group. Through evaluation with 8 online lectures (N_instructors=8, N_learners=21), we found such real-time flow prediction and visualization can provide value to presenters. This paper contributes a set of useful findings regarding their perception and reaction of such flow prediction, as well as lessons learned in the study, which can be inspirational for building future AI-powered system to assist people in delivering live online presentations.

In contrast to the extensive studies on static target pointing, much less formal understanding of moving target acquisition can be found in the HCI literature. We designed a set of experiments to identify regularities in 1D unidirectional moving target selection, and found a Ternary-Gaussian model to be descriptive of the endpoint distribution in such tasks. The shape of the distribution as characterized by μ and σ in the Gaussian model were primarily determined by the speed and size of the moving target. The model fits the empirical data well with 0.95 and 0.94 R2 values for μ and σ , respectively. We also demonstrated two extensions of the model, including 1) predicting error rates in moving target selection; and 2) a novel interaction technique to implicitly aid moving target selection. By applying them in a game interface design, we observed good performances in both predicting error rates (e.g., 2.7% mean absolute error) and assisting moving target selection (e.g., 33% or a greater increase in pointing accuracy).

Acquiring Chinese tones is often considered as the most difficult task in learning Chinese as a Second Language (CSL). Recently, ToneWars, a collaborative mobile learning game, demonstrated the feasibility and efficacy of connecting CSL learners with native speakers for tone learning. However, the synchronous gameplay nature in ToneWars can be hard to scale due to the time constraint and limited availability of native speakers. We present principled research to make ToneWars scalable and sustainable. First, we address the scalability issue via asynchronous modeling of native speakers. Second, we quantify whether a CSL learner achieves native level mastery for a specific phrase, and explore the use of fine-grained feedback on language mastery as a sustainable motivator for language learning. The insights in this research are generalizable to designing second language learning technologies beyond Chinese. In a longitudinal study with 18 CSL learners, we found that asynchronous gameplay significantly improved learning with an average gain of 29.7 tones and 16.4 syllables, and helped participants achieve native level mastery on 58.2 out of 69 phrases.

We present the iterative design, prototype, and evaluation of CourseMIRROR (Mobile In-situ Reflections and Review with Optimized Rubrics), an intelligent mobile learning system that uses natural language processing (NLP) techniques to enhance instructor-student interactions in large classrooms. CourseMIRROR enables streamlined and scaffolded reflection prompts by: 1) reminding and collecting students' in-situ written reflections after each lecture; 2) continuously monitoring the quality of a student's reflection at composition time and generating helpful feedback to scaffold reflection writing; and 3) summarizing the reflections and presenting the most significant ones to both instructors and students. Through a combination of a 60-participant lab study and eight semester-long deployments involving 317 students, we found that the reflection and feedback cycle enabled by CourseMIRROR is beneficial to both instructors and students. Furthermore, the reflection quality feedback feature can encourage students to compose more specific and higher-quality reflections, and the algorithms in CourseMIRROR are both robust to cold start and scalable to STEM courses in diverse topics.

Recent technological advances have demonstrated the feasibility of measuring people's heart rates through commodity cameras by capturing users' skin transparency changes, color changes, or involuntary motion. However, such raw image data collected during everyday interactions (e.g. gaming, learning, and fitness training) is often noisy and intermittent, especially in mobile contexts. Such interference causes increased error rates, latency, and even detection failures for most existing algorithms. In this paper, we present BayesHeart, a probabilistic algorithm that extracts both heart rates and distinct phases of the cardiac cycle directly from raw fingertip transparency signals captured by camera phones. BayesHeart is based on an adaptive hidden Markov model, requires minimal training data and is user-independent. Through a comparative study of twelve state-of-the-art algorithms covering the design space of noise reduction and pulse counting, we found that BayesHeart outperforms existing algorithms in both accuracy and speed for noisy, intermittent signals.

Wencan Luo, Xiangmin Fan, Muhsin Menekse, Jingtao Wang, Diane Litman. Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations. 2015.