Wearable Intelligence for Smart Health (WISH) Research Lab

Research grants

EXERCISE - Explore Emerging Computing in Science and Engineering
- National Science Foundation (NSF) Research Experiences for Undergraduates (REU) Site
- I am a Faculty Associate, 2022-2025

Prototyping decision support and monitoring tools for equitable management of salt contamination of water supplies in tidal rivers (subaward)
- National Science Foundation (NSF) Convergence Accelerator Track K
- I am an investigator, Jan. 2024 - Dec. 2024

Gait Disturbance Measurement for Preventing Older Adult Falls
- Salisbury University Faculty Mini-Grant
- I am the PI, 2020 - 2021

Note: Please contact me for potential funding opportunities.

----------------------------------------------------------------------------------------

Advised students

Olivia Brague
- Undergraduate student, Salisbury University
- Project: Promote honors student's critical thinking and research abilities via conducting AI research project
- Title: Inferring handwriting inputs on smartphone with side-channel information
- Sep. 2023 - Dec. 2023
- Honors Session Project

Stephora Alberi
- Undergraduate student, Salisbury University
- Project: High-performance deep learning model for emotion detection
- Aug. 2023 - Dec. 2023
- COSC 390 Undergraduate Research Project

Joshua Comfort
- Undergraduate student, Salisbury University
- Project: PCA-Based Efficient CNN Model Construction for ASL Recognition
- Presented as a Lightning Talk at the 2023 IEEE MIT Undergraduate Research Technology Conference (CS News & PowerPoint slides)
- Jan. 2023 - May 2023
- Supported by NSF RUI project

Ejiro Adams Ubini
- Undergraduate student, The Community college of Baltimore County
- Project: Real-time Model Adaptation for WiFi Gesture Recognition
- Jun. 2022 - Aug. 2022
- Supported by NSF REU Site project

----------------------------------------------------------------------------------------

Research devices

Shimmer3 EMG Development Kit (Base6 + 2 Shimmer3)
This Kit contains two Shimmer3 devices. Each device has 9DoF motion sensors (an accelerometer, a gyroscope, and a compass) and two channels of EMGs.

Smartphone
SAMSUNG Galaxy A14 smartphone is available. It contains accelerometer, gyroscope, compass, camera, microphone, etc.

(picture from internet)

Smart wristband
Garmin vívosmart® 5 is available. It contains an optical heart rate monitor, an accelerometer, an ambient light sensor, and a blood oxygen saturation (SpO2) monitor.

(picture from internet)

Note: please feel free to contact me if you need any of the above devices for your research and experiments.

----------------------------------------------------------------------------------------

Undergraduate research projects & assistant positions

Our goal is to develop lightweight, high-performance, and robust machine learning algorithms and applications to recognize diverse activities through mobile and wearable devices and provide smart health-related services to the public.

The following are some interesting research topics in the fields of wearable intelligence and smart health. In each topic, there are some existing or ongoing projects. We will further investigate these projects to either improve the existing solutions or address the unresolved problems.

 

Topic 1: Wearable dietary monitoring

This topic aims to use wearable motion sensors to continuously monitor user's dietary activities, including eating detection, food type recognition, and chewing side detection.

Project 1-1: Eating activity detection & chews counting
- This project aims to detect eating activity and count chews through attaching a triaxial accelerometer on the temporalis.
- Research skills to learn: classification algorithms, digital signal processing, programming language (Matlab or Python).
- Dataset is available.
- Reference: S Wang et al., Eating detection and chews counting through sensing mastication muscle contraction, Elsevier Smart Health 2018
- Two research assistant positions available. One is for high-performance machine learning model; the other is for noisy signal processing. Please contact me if you are interested.

 

Project 1-2: Food type inferring through characterizing mastication dynamics
- This project aims to use wearable motion sensors to sense mastication dynamics and infer food types accordingly.
- Research skills to learn: feature engineering, classification algorithms, programming language (Matlab or Python).
- Dataset is available.
- Reference: S Wang et al., Inferring food types through sensing and characterizing mastication dynamics, Elsevier Smart Health 2021
- One research assistant position available. Please contact me if you are interested.

 

Project 1-3: Chewing side detection
- This project deploys motion sensors on the mastication muscles to sense muscle bulges and skull vibrations and differentiate chewing sides accordingly.
- Research skills to learn: deep learning algorithms, signal processing, programming language (Matlab or Python).
- Dataset is available.
- Reference: S Wang et al., Wearable Motion Sensor-based Chewing Side Detection, Elsevier Smart Health 2021
- One research assistant position available. Please contact me if you are interested.

 

Topic 2: Walking gait analysis

This topic aims to use wearable sensors to analyze user's walking gait and provide personalized services, such as fall prevention and gait balance analysis.

Project 2-1: Walking surface detection for fall prevention
- This project uses motion sensors near the ankle to sense foot-floor friction and recognize walking surfaces.
- Research skills to learn: feature engineering, classification algorithms, digital signal processing, programming language (Matlab or Python).
- Dataset is available.
- Reference: S Wang and G Zhou, Poster: Foot-Floor Friction Based Walking Surface Detection for Fall Prevention Using Wearable Motion Sensors, ACM/IEEE CHASE 2023
- One research assistant position available. Please contact me if you are interested.

 

Topic 3: Hand gesture/interaction analysis

This topic aims to analyze user's hand gestures and/or detect hand-device interaction activities using smartphone/wristband.

Project 3-1: CSI-based sign language recognition
- This project uses CSI information (wireless signals received on WiFi router) to identify user's sign gestures.
- Research skills to learn: deep learning algorithms (e.g. CNN), digital signal processing, programming language (Matlab).
- Dataset is available.
- Reference: Y Ma et al., Signfi: Sign language recognition using wifi, ACM UbiComp 2018
- One research assistant position available. Please contact me if you are interested.

 

Project 3-2: Smartphone touchscreen handwriting recognition
- This project uses smartphone-embedded sensors to recognize the numbers and characters that user inputs through handwriting. This is a security-related research topic.
- Research skills to learn: data preprocessing, feature engineering, machine learning algorithms (e.g. MLP or SVM), digital signal processing, programming language (Matlab or Python).
- Dataset is available.
- One research assistant position available. Please contact me if you are interested.

 

Topic 4: Cognitive impairment assessment

This topic aims to analyze user's hand motions to assess their cognitive capabilities and the trends of improvement or deterioration.

Project 4-1: Cognitive Impairment Assessment Using Motion Sensor-Embedded Wristbands
- This project uses motion sensor-embedded wristbands to sense user's hand motions and assess user's cognitive impairment accordingly.
- Research skills to learn: data preprocessing, feature engineering, machine learning algorithms, programming language (Python or Matlab).
- One research assistant position available. Please contact me if you are interested.

 

Other topics

If you are interested in other wearable computing and smart health topics (e.g. sports training support, elderly care, etc.), please contact me to have a short discussion.

Please check my Google Scholar page for more references.

 

 

Contact Me

I am looking for undergraduate students to work with me on multiple research projects. Interested students are encouraged to contact me.