韩国裸舞

Electrical and Computer Engineering
韩国裸舞.CA / Faculty of Engineering / Students / Undergraduate Students / Undergraduate Research (SURE) / Student Section / Projects

Computer, Electrical & Software Engineering 2025

ECSE 001: Temporal Generative Modelling for Personalized Medicine in Multiple Sclerosis (Arbel)

Professor Tal听Arbel

tal.arbel [at] mcgill.ca
5143988204
Research Area
generative modeling, temporal modeling, deep learning, machine learning, artificial intelligence, computer vision, medical image analysis, multiple sclerosis, personalized medicine
Description
This project aims to create cutting-edge deep learning tools for personalized medicine in multiple sclerosis. This involves developing generative modeling techniques, such as Diffusion Models, to accurately predict patient-specific disease trajectories in Magnetic Resonance Imaging (MRI) over time across various treatments. When implemented in clinical practice, these tools could rapidly identify the most effective therapies, representing a major breakthrough in patient care and AI-driven personalized medicine.
Tasks per student
Students will work closely with graduate students in Prof. Arbel鈥檚 lab and a Research Scientist at MILA, and assist with (1) the development of generative modeling techniques (e.g. Diffusion Models) to accurately predict patient-specific disease trajectories in MRI over time across various treatments, and (2) with supporting infrastructure.

听
Deliverables per student
A deep learning model, or functional subcomponents, of a larger deep learning system for personalized medicine in multiple sclerosis.
Number of positions

2

Academic Level

Year 3

Location of project

in-person

ECSE 002: Physio-Video (Armanfard)

Professor听Narges听Armanfard

narges.armanfard [at] mcgill.ca
5143982939
Research Area
Artificial Intelligent, Machine Learning, Computer Vision and Signal Processing
Description
This project focuses on recording video data of participants to analyze and measure key physiological metrics using video-based techniques. The goal is to leverage advanced AI and signal processing methods to extract accurate health data from video footage.
Tasks per student
Data Collection:
- Record video and physio signals from participants, ensuring high-quality and consistent data across sessions.

-Collaborate in setting up the experimental environment, ensuring the proper placement of cameras and equipment.

Experimental Setup & Troubleshooting:
- Assist in the configuration and maintenance of experimental setups, including cameras, lighting, and other sensor equipment.
- Identify and troubleshoot any technical issues that arise during the data recording process.

Signal and Video Processing:
- Develop and implement AI-driven algorithms for processing video and signal data to extract physiological metrics.

- Explore advanced image processing techniques to enhance data quality and optimize model inputs.

- Analyze the results of processed data and suggest improvements.

Model Testing & Validation:
- Test and validate the currently developed AI models on the recorded video data.
- Contribute to refining the models based on performance metrics and results.
- Conduct experiments to evaluate model accuracy and propose adjustments to improve outcomes.

听
Deliverables per student
Biweekly Reports:
- Provide detailed reports every two weeks outlining progress, challenges, and milestones achieved.

AI Code and Implementation:
- Deliver well-documented AI code for video and signal processing tasks.
- Implement, test, and refine AI models for physiological measurement.

Documentation of Experimental Setup:
- Document the entire data collection and experimental setup process, including cameras specifications, lighting, and equipment configurations.
- Create troubleshooting guides for setup maintenance and issue resolution.

Data Analysis & Evaluation:
- Provide thorough analysis of the collected data, summarizing findings and identifying potential improvements.
- Contribute to the interpretation of results in the context of the model鈥檚 performance.

Model Testing Results & Improvements:
- Submit results from model tests, including performance evaluations and suggested model adjustments.
- Provide recommendations for future testing or improvements to both the experimental setup and processing pipelines.
Number of positions

2

Academic Level

Year 3

Location of project

in-person

ECSE 003: Identity and failure 鈥� putting it all together听(Chen)

Professor Lawrence Chen

lawrence.chen [at] mcgill.ca

438 496 3495
Research Area
Engineering education
Description听
Research on engineering identity has been conducted for many years. The development of an engineering identity, especially amongst undergraduate students, is important for persistence, motivation, engagement, and success. A commonly used measured of engineering identity in students is based on Godwin鈥檚 measures and involves three constructs: performance/competence, interest, and recognition.

At the same time, the motivational strategies for learning questionnaire (MSLQ), the achievement emotions questionnaire (AEQ), and Dweck鈥檚 implicit theories of intelligence scale (growth mindset scale) are frequently used to examine the interconnection between student motivation, emotions, and learning. These frameworks are especially useful for understanding students鈥� perspectives and attitudes towards academic failure.

However, there are few studies that examine the relationship between engineering identity and factors such as self-regulation, growth mindset, and correspondingly academic failure.

Our hypotheses are the following:

H1 鈥� a strong engineering identity predicts a strong growth mindset
H2 鈥� a strong engineering identity predicts strong self-regulation
H3 鈥� a strong engineering identity solicits more positive emotional responses towards academic failure

The objective of this project is to conduct a pilot study of a new survey tool that combines items from Godwin鈥檚 engineering identity scale, Dweck鈥檚 implicit theories of intelligence scale, the MSLQ, and the AEQ. Based on data collected, analysis will include exploratory and confirmatory factor analyses, followed by structural equation modeling or multiple regression to test the hypotheses. The results may be useful for engineering educators and educational developers to design interventions that enhance engineering identity, develop growth mindsets, and promote greater resilience to failure.
Tasks per student
The research involves a literature review and quantitative (statistical) data analysis:

1. Complete the TCPS 2 : Core-2022 (Course on Research Ethics)

2. Conduct a literature review on the use of the MSLQ, AEQ, and Dweck鈥檚 scale for understanding student perspectives, attitudes, and emotions towards academic failure

3. Perform exploratory and confirmatory factor analyses with survey data

4. Use the results from the exploratory and confirmatory factor analyses to develop different structural models to test the hypotheses and evaluate the model fit using different measures. Alternatively, use multiple regression analysis to test the hypotheses and examine for the presence of mediating factors. Examine whether demographic variables such as gender or year of study impact the relationships.

5. Interpret the results to highlight the implications on pedagogy and teaching if the hypotheses are supported or identify reasons if the hypotheses are not supported. Identify unanticipated relationships in the data.

听
Deliverables per student
The SURE student will complete a detailed report that includes the literature review and summarizes the results of the statistical analysis.
Number of positions

1

Academic Level

No preference

Location of project

in-person

ECSE 004: Using LLMs to make cockpit messaging smarter (Cooperstock)

Professor Jeremy听Cooperstock

jer [at] cim.mcgill.ca
5143985992
Research Area
Intelligent Systems
Description
Modern aircraft presently use voice communication and text messaging systems to communicate information between air traffic controllers and pilots. Text messaging is done over controller-pilot data link communications (CPDLC), which has been almost universally adopted in commercial aircraft cockpits. To cope with the increase in global air traffic, which is expected to double between 2017 and 2037, there is a push for CPDLC to become the primary mode of communication for pilots, in order to reduce the demand on already crowded radio channels.

CPDLC communication currently takes the form of pre-formatted messages whose parameters can be specified by the sender. For example, controllers can choose a message like 鈥淐LIMB TO FL_____鈥� and then fill in the flight level (FL) parameter to get 鈥淐LIMB TO FL340鈥�. These messages can then be loaded by pilots into their flight management system (FMS), allowing them to quickly change altitude. Despite this advantage, CPDLC is currently under-utilized by pilots. This may be due in part to the inflexible nature of the current messaging system, which requires pilots to send additional messages to provide context. As a result, pilots often switch back to voice communication over radio. We are investigating how controller-pilot messaging would function if pilots could send messages that are typed freely, more similarly to the way they speak over the radio. The system, however, still needs to be able to extract relevant instructions from messages, so that they can be loaded into the FMS.
Tasks per student
You will be refining a language model that extracts codified CPDLC instructions from natural language text messages and integrate said model into simulation software for use in testing.

听
Deliverables per student
1) connecting the language model to simulation software (e.g. XPlane) so that information can be passed back and forth between the software and the model, and 2) implementing a graphical user interface that allows users to interact with the language model while in the simulation environment.
Number of positions

1

Academic Level

No preference

Location of project

in-person

ECSE 005: High Performance Computational Electromagnetics (Giannacopoulos)

Professor Dennis听Giannacopoulos

dennis.giannacopoulos [at] mcgill.ca
514-398-7128
Research Area
Computational Electromagnetics
Description
To model the electromagnetic fields accurately and efficiently within sophisticated microstructures of modern engineering systems, devices, and biological structures, high performance computing (HPC) methods, such as parallel and distributed simulations on emerging multicore/manycore platforms, are deemed promising for overcoming current computational bottlenecks. While robust and reliable 3-D automatic mesh generation procedures and solution strategies for electromagnetics are emerging, major computational challenges remain for effective parallel and distributed 3-D adaptive finite element methods (AFEMs). Uniting AFEMs and HPC methods to achieve high gains in efficiency makes it possible to solve previously intractable problems; however, effective implementation of such techniques is still not well understood. AFEMs for parallel/distributed computing introduce complications that do not arise with simpler solution strategies. For example, adaptive algorithms utilize unstructured meshes that make the task of balancing processor computational load more difficult than with uniform structures. To help address these and other challenges, machine learning (ML) based approaches will be leveraged for developing state-of-the-art mesh generation and multiphysics solvers for complex systems including biomedical applications.
Tasks per student
The students in this project will research and develop efficient ML-based parallel and distributed adaptive algorithms for unstructured meshes that use sophisticated data structures for implementing dynamic load balancing strategies for heterogeneous HPC environments such as multicore/manycore architectures. The students鈥� role will include involvement in all aspects of the engineering research process for this project including actual implementation of algorithms as executable code.

听
Deliverables per student
The students are expected to help deliver a functioning, well-documented ML-based 3-D parallel automatic mesh generator and solver suitable for use with AFEM refinement criteria, along with documented case study validation & verification examples relevant to complex engineering systems including biomedical applications.
Number of positions

1

Academic Level

No preference

Location of project

hybrid remote/in-person - a) students must have a Canadian bank account and b) all students must participate in in-person poster session.

ECSE 006: Deep Learning to Predict Histologic Transformation in Low-Grade Lymphoma (Arbel)

Professor Tal Arbel

tal.arbel [at] mcgill.ca

5143988204
Research Area
deep learning, machine learning, artificial intelligence, computer vision, medical image analysis, cancer, personalized medicine
Description
This project focuses on developing cutting-edge deep learning tools to determine whether a cancer (lymphoma) patient is transitioning to a more severe level of disease based entirely on demographic, laboratory, and medical imaging data. To date, a biopsy has generally been required to identify this transition, with patients often undergoing several biopsies over the years, resulting in complications and delays in treatment. As such, developing deep learning techniques to identify this transition using data acquired non-invasively (e.g. imaging, demographic, laboratory), would represent a significant step forward in patient care and AI-based personalized medicine.
Tasks per student
The primary objective of this project is to develop a deep learning model to predict histologic transformation using demographic, laboratory and imaging data (alleviating the need for invasive biopsies). The student will work closely with graduate students in Prof. Arbel鈥檚 lab, a Research Scientists at MILA, and collaborating physicians at the Jewish General Hospital and the 韩国裸舞 Health Centre.

听
Deliverables per student
A deep learning model capable of predicting histologic transformation using demographic, laboratory and imaging data.
Number of positions

1

Academic Level

Year 3

Location of project

TBD

ECSE 007:听Soft glass fiber coupling and fusing (Rochette)

Professor Martin Rochette

martin.rochette [at] mcgill.ca
5145767045
Research Area
Electrical and computer engineering
Description
We need to develop recipes for the fusing of optical fibers made of silica and fluoride glasses. In the first thrust, we will aim to create recipes for fabricating fluoride-to-fluoride fiber splices, whereas in the second thrust, we will target silica-to-fluoride splices. A specialty optical fiber splicer is used for this purpose.
Tasks per student
Tasks related to this project include the following:
Learn from the material provided: Instructions from the optical fiber splicer, journal articles.
Report on the functionality of the current splicing system and highlight potential limitations and inaccuracies. Establish the space of input variables that will be used to converge towards successful recipes.
Splices will be characterized with an optical fiber transmission setup
This is a research project, and thus, new orientations may be taken as the project moves forward.
Knowledge of Labview, Matlab, and French language are assets, but not strict requirements.

听
Deliverables per student
At the end of this project, we expect a complete report about the splice fabrication, including specific recipes and optical characterization results.
Number of positions

1

Academic Level

No preference

Location of project

in-person

ECSE 008: Solid contact ion sensor array fabrication and test for controlled environment agriculture (Szkopek)

Professor Thomas听Szkopek

thomas.szkopek [at] mcgill.ca

514 892 1064
Research Area
Electronics, electrochemisty, nanoelectronics, microfabrication
Description
The core objective of this project is to fabricate and test solid contact ion sensor arrays using state-of-the-art electrode design incorporating nanomaterials for a turn-key automated real-time system for nutrient profile measurement of hydroponic water. The end goal is the development of technology that will ultimately enable increased yield and reduced waste in controlled environment agriculture.
Tasks per student
The student will fabricate and test solid contact ion sensor arrays, using processes and procedures already documented within the laboratory.
This will included working with planar manufacturing and electrochemical methods in a wet chemical laboratory enivornment.
Electrochemical measurement techniques will be used for sensor array testing.

听
Deliverables per student
The student will be responsible for delivering:
1) functional solid contact ion sensors for potassium, calcium, magnesium, nitrate
2) a report summarizing fabrication procedures and protocols used for each sensor array
3) a report summarizing electrochemical testing results for each sensor array

听
Number of positions

1

Academic Level

No preference

Location of project

in-person

ECSE 009: 2D Materials Research Instrumentation (Szkopek)

Professor Thomas听Szkopek

thomas.szkopek [at] mcgill.ca

514 892 1064
Research Area
Nanoelectronics
Description
This project entails the coordination of research instrumentation, including lock-in amplifiers, source-measurement units, temperature controllers, gas flow controllers and a magnet power supply for research in the electronic properties of 2D materials. If time permits, there will be an opportunity to contribute to the measurement of various graphene derivatives versus temperature and magnetic field.
Tasks per student
The primary task of the student will be to develop a python code for the coordination of various research instrumentation, including functionality for instrument control, data collection, data storage, and data plotting in real-time. Instrumentation includes a variety of lock-in amplifiers, source-measurement units, temperature controllers (with thermometry and heating), gas flow controllers, a magnet power supply for a superconducting solenoid.
The secondary task of the student, if time permits, will be to contribute to the measurement of the resistivity tensor of chemically modified graphene versus temperature and versus magnetic field.

听

听
Deliverables per student
The student will deliver:
- a modular, well-organized, python code with the aforementioned functionality
- documentation of the functionality achieved
- thorough documentation of the python code itself
If experiments are conducted, the student will deliver:
- complete laboratory book documentation of all experimental details of resistivity tensor measurement
- a brief report explaining the primary findings
Number of positions

1

Academic Level

No preference

Location of project

in-person
Tags: 
Summer Undergraduate Research Opportunities
Undergraduate Research
Back to top