LURA/USRA Posting Information
Lassonde Undergraduate Research Award- summer research
Research Assistant/summer researcher
Location: Farq 108
Professor: Peter Lian
Department: Electrical Engineering and Computer Science
Contact for Professor (Email, phone): firstname.lastname@example.org
# of positions available: 4
Project Description (200-500 words maximum)
This project aims to develop a machine learning algorithms to extract heart rate information from biosignals recorded by wireless wearable electrocardiogram (ECG) sensors. The biosignals recorded by a wearable biomedical sensor normally contain motion artefacts and other noises, which greatly reduce the diagnostic value of these recordings. Traditionally, the recorded biosignals are processed by digital signal processing algorithms to remove noises before extracting useful information. With the advancement of machine learning techniques, it is possible to extract useful information directly from noise corrupted biosignals. These algorithms are very useful in detecting heart condition from ECG signals. Knowing the heart condition will be a great help for patients with cardiovascular diseases, especially for heart failure (HF) patients. Heart failure is a growing epidemic in Canada. It is a significant health issue for hundreds of thousands of Canadians and their families, and its reach is expanding according the 2016 Report on the Health of Canadians from HEART & STROKE Foundation. 600,000 Canadians are living with HF. 50,000 Canadians are diagnosed each year with the HF. One in two Canadians has been touched by the HF. HF costs more than $2.8 billion per year. HF, in simple terms, arises due to damaged valves or heart muscles. HF normally leads to reduced pumping capacity of the heart (cardiac output) that can reduce exercise capacity and cause fatigue. HF can also result in an imbalance of blood pressure leading to fluid retention in the body, particularly congestion in the lungs, which can cause breathlessness. An acute episode of breathlessness typically sends the HF patient to the emergency department. The patient then spends on average five days in the hospital to stabilize his/her condition but remains susceptible to future decompensation. Prior to discharge, patient gets some time to learn a huge list of HF self-care instructions, which is usually overwhelming to the sick patient. This in turn leads to poor self-care at home and poor compliance to medication and diet, which are common causes for the patient’s HF rehospitalization. Thus, the rehospitalization rate is high, in the range of 14% to 24%, in Canada. With the machine learning based heart rate detection algorithm and wearable wireless sensors, it is possible to allow doctor to remotely access patient’s heart condition, which may provide just-in-time help for patient. It will also empower patient, their home-caregivers and practitioners to achieve better home care, and reduce rehosptialization rate.
Duties and Responsibilities of the student:
The successful candidate will be responsible for the development of machine learning algorithm, neural network training, and APP development.
The student should have the following qualifications:
1. Basic knowledge of digital signal processing.
2. Good knowledge in programming and APP development
3. Knowledge in machine learning or AI is a plus.
4. Passionate about programming.
Degrees, courses and Disciplines prerequisite*:
Minimum requirement: EECS2030, EECS2031 or equivalent
Preferred: EECS2311, EECS3401 or EECS3451
Are you willing to host external students? (There is an additional cost.): Yes
Duration: 16 weeks minimum
Start Date: 05/01/2018 (estimated)
End Date: 08/31/2018 (estimated)
Materials required for application: TBC
Please contact email@example.com directly for more information.