About me

I am a Postdoctoral Research Associate at Advanced Human Health Analytics Laboratory (AHHA) , College of Information and Computer Sciences, UMass Amherst, where I worked with Prof. Ivan Lee. I completed my Ph.D. and M.S. in the Department of Biomedical Engineering at National Yang-Ming University (newly formed National Yang Ming Chiao Tung University on 2021) in 2019 and 2015. From 2020 to 2023 I was a Postdoctoral Scholar at Dr. Yu Tsao's Bio-ASP Lab, Research Center of Research Center for Information Technology Innovation(CITI), Academia Sinica. In 2023, I was an Assistant Professor in the Department of Electronic Engineering, at National Taipei University of Technology (NTUT) , Taipei, Taiwan. In 2018, I was a visiting scholar in Prof. Bjoern Eskofier's Machine Learning and Data Analytics (MaD) lab at the Friedrich-Alexander University (FAU) Erlangen-Nuernberg (Germany).

I invent novel medical monitoring and assessment systems. I am an expert in medical system integration and building proof-of-concept prototypes that include interdisciplinary skills, such as bio-signal processing, machine learning, human-computer interaction, and clinical trial design and management. My research aims to develop healthcare and computer-assisted systems to support diagnosis, evaluation, monitoring and assessment in clinical and home-based environments through sensors and machine learning.

The main research topic includes pervasive healthcare, wearable computing, machine learning and bio-signal processing.

More About Me

Current Active Projects

-Fall Detection and Monitoring

Deep Learning Based Signal Enhancement of Low-Resolution Accelerometer for Fall Detection Systems

to appear in IEEE Transactions on Cognitive and Developmental Systems (IEEE TCDS)

Kai-Chun Liu, Kuo-Hsuan Hung, Chia-Yeh Hsieh, Hsiang-Yun Huang, Chia-Tai Chan, and Yu Tsao*

To improve the detection accuracy with low-resolution accelerometer signals, this work proposed ASE model based on a deep denoising convolutional autoencoder architecture reconstructs high-resolution (HR) signals from the low-resoution (LR) signals by learning the relationship between the LR and HR signals.

Domain-adaptive Fall Detection Using Deep Adversarial Training

IEEE Transactions on Neural Systems and Rehabilitation Engineering (IEEE TNSRE), vol. 29, pp. 1243-1251, 2021

Kai-Chun Liu, Michael Chan, Heng-Cheng Kuo, Chia-Yeh Hsieh, Hsiang-Yun Huang, Chia-Tai Chan, Yu Tsao*

We proposed domain-adaptive fall detection (DAFD) using deep adversarial training (DAT) to tackle cross-domain problems, such as cross-position and cross-configuration.

Multiphase Identification Algorithm for Fall Recording Systems Using a Single Wearable Inertial Sensor

MDPI Sensors 2021, 21(9), 3302

Chia-Yeh Hsieh, Hsiang-Yun Huang, Kai-Chun Liu, Chien-Pin Liu, Chia-Tai Chan, Steen Jun-Ping Hsu*

This work proposed an automatic multiphase identification algorithm for phase-aware fall recording systems. The proposed multiphase identification algorithm combines machine learning techniques and fragment modification algorithm to identify pre-fall, free-fall, impact, resting and recovery phases in a fall process.

-Frozen Shoulder Assessment

Instrumented shoulder functional assessment using inertial measurement units for frozen shoulder

2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI)

Ting-Yang Lu, Kai-Chun Liu, Chia-Yeh Hsieh, Chih-Ya Chang, Yu Tsao, Chia-Tai Chan

In this work, we propose an IMU-based shoulder functional task assessment with kinematic parameters (e.g., smoothness, power, speed, and duration) in FS patients and analyze the functional performance on complete shoulder tasks and subtasks. The results demonstrate that the used smoothness features can reflect the differences of movement fluency between FS patients and healthy controls Morover, features of subtasks provided subtle information related to clinical conditions that have not been revealed in features of a complete task, especially the defined subtask 1 and 2 of each task

Automatic Functional Shoulder Task Identification and Sub-Task Segmentation Using Wearable Inertial Measurement Units for Frozen Shoulder Assessment

MDPI Sensors 2021, 21(1), 106

Chih-Ya Chang, Chia-Yeh Hsieh, Hsiang-Yun Huang, Yung-Tsan Wu, Liang-Cheng Chen, Chia-Tai Chan, Kai-Chun Liu*

This pilot study aims to propose an automatic functional shoulder task identification and sub-task segmentation system using inertial measurement units to provide reliable shoulder task labeling and sub-task information for clinical professionals. The proposed method combines machine learning models and rule-based modification to identify shoulder tasks and segment sub-tasks accurately. The experimental results show that the proposed method can achieve 87.11% F-score for shoulder task identification, and 83.23% F-score and 427 mean absolute time errors (milliseconds) for sub-task segmentation.

-Assessment and Diagnosis of Vestibular Hypofunction

Instrumented Romberg Test of Postural Stability in Patients with Vestibular Disorders using Inertial Measurement Units

To appear in 2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA)

Yu-Chieh Lin, Kuan-Chung Ting, Kai-Chun Liu, Chia-Yeh Hsieh, Chia-Tai Chan

The study aimed to propose an instrumented Romberg test using inertial measurement units (IMUs) to extract kinematic variables for objective assessment. Kinds of parameters based on IMU signals are estimated to quantify kinematic differences between patients with vestibular disorder and healthy people.

-Automatic Otitis Media Detection and Classification

Automatic Detection of Otitis Media with Effusion Using In-ear Microphones

Paper Preparation

Kuan-Chung Ting, Syu-Siang Wang, Jin Li-You, Chii-Yuan Huang, Tzong-Yang Tu, Chun-Che Shih, Kai-Chun Liu and Yu Tsao

Traditional examination tools require expensive equipment that decreases the penetration rate and availability in clinical usage. Additionally, these tools are designed for clinical professionals in examination rooms. To tackle these issues, this work presents a low-cost and easy-to-use OME detection system using off-the-shelf microphones and machine learning (ML) models. The microphones are placed in the ear canal to record in-ear vowel sounds directly.

Contact Me

Kai-Chun Liu

Email: kaichunliu@umass.edu, t22302856@gmail.com
Office Address: Complex Building 406-1, No.1, Sec. 3, Chunghsiao E. Rd. Taipei City, 10608
Office Phoene: 886-2-2771-2171 Ext.2247

-->