UbiComp 2024
User preferences for mobile sensing and mental health interventions vary significantly across data types, with some sensors more acceptable than others. A university-wide survey reveals that individuals willing to share one type of data are often open to others, highlighting distinct engagement patterns. These insights support the design of inclusive, scalable mental health apps tailored to student needs.
IJROBP 2024
Contouring assessments from novice radiation oncology residents reveal consistent patterns of error across disease sites, including undercontouring lymph nodes and missing key at-risk volumes. Using an interactive online platform during early clinical rotations, the study identifies specific structures frequently overlooked, such as the inguinal nodes in anal cancer and the base of skull in nasopharyngeal cases. These insights can inform targeted educational interventions and support the development of more standardized contouring curricula.
UIST 2023PaperToPlace is a novel workflow comprising an authoring pipeline, which allows the authors to rapidly transform and spatialize existing paper instructions into an MR experience, and a consumption pipeline, which computationally places each instruction step at an optimal location that is easy to read and does not occlude key interaction areas.
ASSETS 2023Assessing accessibility for wheelchair users can be challenging, due to lack of accessibility details needed for individual users. Embodied Exploration is a VR technique to deliver the experience of a physical visit while keeping the convenience of remote assessment. Embodied Exploration allows wheelchair users to explore high-fidelity digital replicas of physical environments with themselves embodied by avatars, leveraging the increasingly affordable VR headsets.
CHI 2023We present a prototype platform for tumor-contouring training that gives radiation-oncology residents overlap scores, visual cues for over- or under-contoured regions, and toxicity forecasts, enriching apprenticeship learning with timely, detailed feedback.
CHI 2023UnMapped merges a live 3-D “window” into the novice’s scene with a static 3-D replica of the expert’s own desk, letting mentors anchor instructions to familiar spatial landmarks. Compared with fully immersive views, this hybrid layout sped up remote guidance, reduced communication overhead, and eased expert workload.
IMWUT 2023Mites is a scalable hardware–software platform that deploys hundreds of privacy-aware, high-fidelity sensors across entire buildings—demonstrated with 314 nodes over five years in a five-story campus hall—to power diverse, machine-learning smart-building applications.
AMIA 2023
Implicit biases in healthcare harm communication, decision-making, and care quality for marginalized patients. Through co-design workshops with 32 BIPOC, LGBTQ+, and QTBIPOC individuals, we identified four patient-centered solutions: accountability measures, real-time correction, enablement tools, and provider resources. These insights advance patient-focused approaches to addressing bias in care.