Shwetak Patel

Shwetak Patel

Auto-generated profile, based on publicly available information

Shwetak Patel

University of Washington, Seattle, USA

Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 December 2022 in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
Verified
Journal Article
Open access
XXXuhai Xu
Xuhai Xu
XLXin Liu
Xin Liu
HZHan Zhang
Han Zhang
WWWeichen Wang
Weichen Wang
SNSubigya Nepal
Subigya Nepal
JMJennifer Mankoff
Jennifer Mankoff
There is a growing body of research revealing that longitudinal passive sensing data from smartphones and wearable devices can capture daily behavior signals for human behavior modeling, such as depression detection. Most prior studies build and evaluate machine learning models using data collected from a single population. However, to ensure that a behavior model can work for a larger group of users, its generalizability needs to be verified on multiple datasets from different populations. We present the first work evaluating cross-dataset generalizability of longitudinal behavior models, using depression detection as an application. We collect multiple longitudinal passive mobile sensing datasets with over 500 users from two institutes over a two-year span, leading to four institute-year datasets. Using the datasets, we closely re-implement and evaluated nine prior depression detection algorithms. Our experiment reveals the lack of model generalizability of these methods. We also implement eight recently popular domain generalization algorithms from the machine learning community. Our results indicate that these methods also do not generalize well on our datasets, with barely any advantage over the naive baseline of guessing the majority. We then present two new algorithms with better generalizability. Our new algorithm, Reorder, significantly and consistently outperforms existing methods on most cross-dataset generalization setups. However, the overall advantage is incremental and still has great room for improvement. Our analysis reveals that the individual differences (both within and between populations) may play the most important role in the cross-dataset generalization challenge. Finally, we provide an open-source benchmark platform GLOBEM- short for Generalization of Longitudinal BEhavior Modeling - to consolidate all 19 algorithms. GLOBEM can support researchers in using, developing, and evaluating different longitudinal behavior modeling methods. We call for researchers' attention to model generalizability evaluation for future longitudinal human behavior modeling studies.

28 March 2025 in Science Advances
Verified
Journal Article
Open access
YYYuzhe Yang
Yuzhe Yang
YLYujia Liu
Yujia Liu
XLXin Liu
Xin Liu
AGAvanti Gulhane
Avanti Gulhane
DMDomenico Mastrodicasa
Domenico Mastrodicasa
SPShwetak Patel
Shwetak Patel
Advances in artificial intelligence (AI) have achieved expert-level performance in medical imaging applications. Notably, self-supervised vision-language foundation models can detect a broad spectrum of pathologies without relying on explicit training annotations. However, it is crucial to ensure that these AI models do not mirror or amplify human biases, disadvantaging historically marginalized groups such as females or Black patients. In this study, we investigate the algorithmic fairness of state-of-the-art vision-language foundation models in chest x-ray diagnosis across five globally sourced datasets. Our findings reveal that compared to board-certified radiologists, these foundation models consistently underdiagnose marginalized groups, with even higher rates seen in intersectional subgroups such as Black female patients. Such biases present over a wide range of pathologies and demographic attributes. Further analysis of the model embedding uncovers its substantial encoding of demographic information. Deploying medical AI systems with biases can intensify preexisting care disparities, posing potential challenges to equitable healthcare access and raising ethical questions about their clinical applications.

13 May 2024 in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
Verified
Journal Article
Open access
ZEZachary Englhardt
Zachary Englhardt
CMChengqian Ma
Chengqian Ma
MMMargaret E. Morris
Margaret E. Morris
CCChun-Cheng Chang
Chun-Cheng Chang
XXXuhai Xu
Xuhai Xu
VIVikram Iyer
Vikram Iyer
Passively collected behavioral health data from ubiquitous sensors could provide mental health professionals valuable insights into patient's daily lives, but such efforts are impeded by disparate metrics, lack of interoperability, and unclear correlations between the measured signals and an individual's mental health. To address these challenges, we pioneer the exploration of large language models (LLMs) to synthesize clinically relevant insights from multi-sensor data. We develop chain-of-thought prompting methods to generate LLM reasoning on how data pertaining to activity, sleep and social interaction relate to conditions such as depression and anxiety. We then prompt the LLM to perform binary classification, achieving accuracies of 61.1%, exceeding the state of the art. We find models like GPT-4 correctly reference numerical data 75% of the time. While we began our investigation by developing methods to use LLMs to output binary classifications for conditions like depression, we find instead that their greatest potential value to clinicians lies not in diagnostic classification, but rather in rigorous analysis of diverse self-tracking data to generate natural language summaries that synthesize multiple data streams and identify potential concerns. Clinicians envisioned using these insights in a variety of ways, principally for fostering collaborative investigation with patients to strengthen the therapeutic alliance and guide treatment. We describe this collaborative engagement, additional envisioned uses, and associated concerns that must be addressed before adoption in real-world contexts.

19 April 2023
Verified
Conference Paper
Open access
AWAnandghan Waghmare
Anandghan Waghmare
YTYoussef Ben Taleb
Youssef Ben Taleb
ICIshan Chatterjee
Ishan Chatterjee
ANArjun Narendra
Arjun Narendra
SPShwetak Patel
Shwetak Patel
We present Z-Ring, a wearable ring that enables gesture input, object detection, user identification, and interaction with passive user interface (UI) elements using a single sensing modality and a single point of instrumentation on the finger. Z-Ring uses active electrical field sensing to detect changes in the hand’s electrical impedance caused by finger motions or contact with external surfaces. We develop a diverse set of interactions and evaluate them with 21 users. We demonstrate: (1) Single- and two-handed gesture recognition with up to 93% accuracy (2) Tangible input with a set of passive touch UI elements, including buttons, a continuous 1D slider, and a continuous 2D trackpad with 91.8% accuracy, <4.4 cm MAE, and <4.1cm MAE, respectively (3) Object recognition across six household objects with 94.5% accuracy (4) User identification among 14 users with 99% accuracy. Z-Ring’s sensing methodology uses only a single co-located electrode pair for both receiving and sensing, lending itself well to future miniaturization for use in on-the-go scenarios.

21 April 2018
Verified
Conference Paper
Open access
PDF file
AMAlex Mariakakis
Alex Mariakakis
SPSayna Parsi
Sayna Parsi
SPShwetak N. Patel
Shwetak N. Patel
JWJacob O. Wobbrock
Jacob O. Wobbrock
Breathalyzers, the standard quantitative method for assessing inebriation, are primarily owned by law enforcement and used only after a potentially inebriated individual is caught driving. However, not everyone has access to such specialized hardware. We present drunk user interfaces: smartphone user interfaces that measure how alcohol affects a person's motor coordination and cognition using performance metrics and sensor data. We examine five drunk user interfaces and combine them to form the "DUI app". DUI uses machine learning models trained on human performance metrics and sensor data to estimate a person's blood alcohol level (BAL). We evaluated DUI on 14 individuals in a week-long longitudinal study wherein each participant used DUI at various BALs. We found that with a global model that accounts for user-specific learning, DUI can estimate a person's BAL with an absolute mean error of 0.005% ± 0.007% and a Pearson's correlation coefficient of 0.96 with breathalyzer measurements.

5 January 2024 in Science Advances
Verified
Journal Article
Open access
MSManuja Sharma
Manuja Sharma
VNVidelis Nduba
Videlis Nduba
LNLilian N. Njagi
Lilian N. Njagi
WMWilfred Murithi
Wilfred Murithi
ZMZipporah Mwongera
Zipporah Mwongera
DHDavid J. Horne
David J. Horne
Recent respiratory disease screening studies suggest promising performance of cough classifiers, but potential biases in model training and dataset quality preclude robust conclusions. To examine tuberculosis (TB) cough diagnostic features, we enrolled subjects with pulmonary TB (N= 149) and controls with other respiratory illnesses (N= 46) in Nairobi. We collected a dataset with 33,000 passive coughs and 1600 forced coughs in a controlled setting with similar demographics. We trained a ResNet18-based cough classifier using images of passive cough scalogram as input and obtained a fivefold cross-validation sensitivity of 0.70 (±0.11 SD). The smartphone-based model had better performance in subjects with higher bacterial load {receiver operating characteristic–area under the curve (ROC-AUC): 0.87 [95% confidence interval (CI): 0.87 to 0.88],P< 0.001} or lung cavities [ROC-AUC: 0.89 (95% CI: 0.88 to 0.89),P< 0.001]. Overall, our data suggest that passive cough features distinguish TB from non-TB subjects and are associated with bacterial burden and disease severity.