Tutorial: Biosignal-driven Models for Improved Control


Opportunities for solving medical problems with wearable devices are growing. However, improving the intelligence of these devices to provide new solutions relies on collecting, processing, and using biological signals (biosignals). Through analyzing and using biosignals, device performance can be enhanced, and more complex types of interactions can be provided. We propose a tutorial to educate researchers and engineers on methods for sensing and processing biosignals, and the development of models that capture biological properties of the human.

This tutorial will discuss a set of processing methods and models that we have used to improve the intelligence of our wearable devices. The tutorial will be divided into six sections (Figure 1), with each section presented by one of the team members. Upon completion of our tutorial, researchers and engineers will have a better understanding of what it takes to sense and process biological signals for medical applications. They will also have an understanding of different modelling approaches that can be taken to enhance the control of wearable devices.

Figure 1: Outline and ordering of topics covered in the tutorial. The tutorial will begin with the presentation on Sensing Biosignals.

Team Members

Dr. Michael Naish, University of Western Ontario
Dr. Yue Zhou, University of Western Ontario
Parisa Daemi, University of Western Ontario
Anas Ibrahim, University of Western Ontario
Dr. Tyler Desplenter, University of Western Ontario
Dr. Ana Luisa Trejos, University of Western Ontario


1) Sensing Biosignals – Dr. Michael Naish
Dr. Naish will discuss the types of biosignals used for controlling wearable mechatronic devices and the design of sensing systems to acquire these signals.

2) Motion Estimation – Dr. Yue Zhou
Dr. Zhou will examine the characteristics of motion signals and the development of models for estimating components of these signals.

3) Motion Prediction – Anas Ibrahim
Anas will discuss the use of neural networks for processing biosignals and predicting motion.

4) Kinematic Modelling of the Hand – Parisa Daemi
Parisa will discuss the use of motion signals to develop a kinematic model of the hand.

5) Musculotendon Motion Modelling – Dr. Tyler Desplenter
Dr. Desplenter will discuss the use of electromyography and motion signals to estimate musculotendon contributions to joint torque.

6) Data Analysis and Management – Dr. Ana Luisa Trejos
Dr. Trejos will discuss the analysis of biosignal-driven models and the management of the biosignal data.

Note: The tutorial titled “An Introduction to Reinforcement Learning: From Theory to Application” has been cancelled

Tutorial: Hybrid Microfluidic CMOS Sensing Platforms for Life Science Applications


Ebrahim Ghafar-Zadeh, P.Eng., Ph.D., Associate Professor, York University
Sebastian Magierowski, P.Eng., Ph.D., Associate Professor, York University

Motivation and Focus

Recent advances in micro- and nanotechnologies have enabled the development of High-Throughput Screening (HTS) techniques for various applications including drug discovery [1]. Owing to the seminal advances in micro-fabrication technologies, HTS is moving towards massively parallel, miniaturized, and label-free platforms. A state-of-the-art DNA sequencing platform featuring millions of Ion-Selective Field Effect Transistors (ISFETs) has convincingly demonstrated the advantage of using standard microelectronic technologies such as Complementary Metal Oxide Semiconductor (CMOS) process in HTS applications [2].

Similarly, many researchers have addressed the challenge of developing HTS systems for monitoring cellular activities on a single chip. Due to the significant advantages of CMOS-based biosensors, such as non-invasive long term recordings, fast response times and label-free processes, they have been widely applied in many biological and medical fields concerning the study of living-cell samples such as neural cell recording and stimulation [3–6], monitoring metabolic activity [7], cell manipulation [8, 9], and extracellular pH monitoring [10,11], nanopore DNA sequencing [16-17] and NMR spectroscopy [18-19].

Among various CMOS sensing techniques, we have reported the advantages of capacitive sensors as low- complexity, high precision, label-free sensing methods for monitoring cellular activities such as cell viability, proliferation and morphology [1, 12, 13]. Also, we investigated the advantages of nano-pore and pH ISFET sensors for DNA sequencing [16 which have revolutionized DNA sequencing technologies. Furthermore, we elaborate on the nuclear magnetic resonance (NMR) sensors a non-invasive method for drug discovery.

For the references cited above and the presenter CVs please see the attached PDF.

Basic Structure

1. Introduction An overview of CMOS biosensors and their applications and Post-CMOS Microfluidic Processes will be provided.
2. CMOS Capacitive Sensor
The design and implementation of capacitive sensing techniques including Charge based
measurement method (CBCM) will be discussed.
3. Nano-pore Techniques
The principle and application on Nanopore Sensors for various applications including DNA
sequencing will be discussed.