Engineering can take you places. Just ask Ben Sprenger (Year 4 MIE), who spent two weeks in the summer of 2019 talking to Mongolian nomads about portable solar generators.
“I wanted to apply the skills that I had learned through classes and extracurriculars to solving important global problems,” he says. “I was also really interested in the opportunity to work in a multidisciplinary team with people in other fields, such as Arts & Science students.”
The Mongolian project was coordinated by the Reach Alliance from U of T’s Munk School of Global Affairs & Public Policy. Working in a team that included fellow U of T students Hannah Rundle (Munk School MGA candidate), Rushay Naik (Health Policy MSc candidate) and Tanvi Shetty (Munk School MGA candidate) Sprenger conducted interviews to understand the impact of the Mongolian government’s 100,000 Solar Ger (Yurt) Electrification Program.
Begun in 2000, the program provided electricity to nomadic families in the form of portable solar home systems. After a decade of operation, it had surpassed its target, reaching more than 70% of nomadic herders across the country. The U of T team wanted to better understand the key innovations that led to the program’s success.
“I will never forget driving through the steppe, not a road to be seen for what felt like a million miles, an endless expanse in front of us completely unmarked by human activity,” says Sprenger. “I loved meeting the local herders, and they welcomed us into their gers. I hope that the information we gathered will be useful for future projects, both in Mongolia and elsewhere.”
The trip wasn’t Sprenger’s only journey abroad during his degree program. As the Team Lead for the University of Toronto Formula Racing Team, he oversaw all aspects of the team’s entry in two international competitions: one in Brooklyn, Mich. and the other in Most, Czech Republic.
“I am an obsessive motorsports fan,” says Sprenger. “The sheer ingenuity and innovation that comes out of the sport is staggering, and its inventions have changed the world more than you could imagine.”
The obsession is more than a hobby. In August 2019 Sprenger moved to Oxfordshire, U.K. to take up a 12-month work placement with Williams Advanced Engineering, one of the world’s 10 Formula One teams. The job was facilitated through U of T Engineering’s Professional Experience Year Co-op Program.
“I wanted to push myself in a very competitive and fast-paced workplace, and learn about the inner workings of what makes a winner,” he says.
Founded by Williams Grand Prix Engineering, Williams Advanced Engineering was contracted by Panasonic Jaguar Racing to design and build its vehicles for the Formula E championship, the world’s only all-electric racing series.
Sprenger contributed to both the 2020 and 2021 vehicles, and says one of his proudest moments was when the team placed first at the ABB FIA Formula E Championship, held in Mexico City in February 2020. Later, the team would also win the 2021 Diriyah E-Prix, taking the lead in the constructor’s championship.
As part of his PEY Co-op internship, Ben Sprenger (Year 4 MIE) worked at Williams Advanced Engineering, designing and creating components for the vehicle that won the 2020 ABB FIA Formula E Championship in Mexico City.
“Seeing all the hard work that everyone on the team did pay off in a race win was inspirational,” says Sprenger. “It was such a thrill seeing the car cross the line on TV, knowing that there were parts that I designed on there.”
Apart from the work itself, Sprenger says that moving to a new continent for a work term was one of the highlights of his degree.
“I was able to travel all around the U.K. and Ireland on weekend trips, and have made great friends with people from across Europe,” he says. “Being away from Canada for a full year has allowed me to grow so much as a person and helped me do things that I never thought I would be able to do.”
Sprenger will be graduating in June 2021, and he’ll soon be jetting off again: he’s secured an internship in California with Tesla, working on battery development.
“Engineering is truly a global discipline,” says Sprenger. “It has become increasingly clear to me that developing an understanding of the unique cultures of the world has become an essential part of the engineering ethos. I’m thankful that UofT has given me the opportunity to explore this at such an early stage in my career.”
Researchers from U of T Engineering and LG AI Research have developed a novel explainable artificial intelligence (XAI) algorithm that can help identify and eliminate defects in display screens.
The algorithm, which outperformed comparable approaches on industry benchmarks, could potentially be applied in other fields, such as the interpretation of data from medical scans. It is the first AI achievement to emerge from a collaboration between LG and University of Toronto in AI research, an investment that LG recently committed to expanding.
The team includes Professor Kostas Plataniotis (ECE), recent graduate Mahesh Sudhakar (MEng 2T0) and ECE Master’s candidate Sam Sattarzadeh, as well as researchers led by Dr. Jongseong Jang at LG AI Research Canada, part of the company’s global research-and-development arm.
“Explainability and interpretability are about meeting the quality standards we set for ourselves as engineers and demanded by the end user,” says Plataniotis. “With XAI, there’s no ‘one size fits all.’ You have to ask whom you’re developing it for. Is it for another machine learning developer? Or is it for a doctor or lawyer?”
XAI is an emerging field, one that addresses issues with the ‘black box’ approach of machine learning (ML) strategies.
In a black box ML model, a computer might be given a set of training data in the form of millions of labelled images. By analyzing this data, the ML algorithm learns to associate certain features of the input (images) with certain outputs (labels). Eventually, it can correctly attach labels to images it has never seen before.
The machine decides for itself which aspects of the image to pay attention to and which to ignore. But it keeps that information secret — not even its designers know exactly how it arrives at a result.
This approach doesn’t translate well when ML is deployed in non-traditional areas such as health care, law and insurance. These human-centred fields demand a high level of trust in the ML model’s results. Such trust is hard to build when nobody knows what happens inside the black box.
“For example, an ML model might determine a patient has a 90% chance of having a tumour,” says Sudhakar. “The consequences of acting on inaccurate or biased information are literally life or death. To fully understand and interpret the model’s prediction, the doctor needs to know how the algorithm arrived at it.”
In contrast to traditional ML, XAI is a ‘glass box’ approach that makes the decision-making transparent. XAI algorithms are run simultaneously with traditional algorithms to audit the validity and the level of their learning performance. This approach also provides opportunities to carry out debugging and find training efficiencies.
Sudhakar says that, broadly speaking, there are two methodologies to develop an XAI algorithm, each with advantages and drawbacks.
The first, known as backpropagation, relies on the underlying AI architecture to quickly calculate how the network’s prediction corresponds to its input. The second, known as perturbation, sacrifices some speed for accuracy, and involves changing data inputs and tracking the corresponding outputs to determine the necessary compensation.
“Our partners at LG desired a new technology that combined the advantages of both,” says Sudhakar. “They had an existing ML model that identified defective parts in LG products with displays, and our task was to improve the accuracy of the high resolution heat maps of possible defects while maintaining an acceptable run time.”
The team’s resulting XAI algorithm, Semantic Input Sampling for Explanation (SISE), is described in a recent paper for the Thirty-Fifth AAAI Conference on Artificial Intelligence.
“We see potential in SISE for widespread application,” says Plataniotis. “The problem and intent of the particular scenario will always require adjustments to the algorithm — but these heat maps or ‘explanation maps’ could be more easily interpreted by, for example, a medical professional.”
“LG’s goal in partnering with University of Toronto is to become a world leader in AI innovation,” says Jang. “This first achievement in XAI speaks to our company’s ongoing efforts to make contributions in multiple areas, such as functionality of LG products, innovation of manufacturing, management of supply chain, efficiency of material discovery and others, using AI to enhance customer satisfaction.”
ECE Chair Professor Deepa Kundur says successes like this are a good example of the value of collaborating with industry partners: “When both sets of researchers come to the table with their respective points of view, it can often accelerate the problem-solving. It is invaluable for graduate students to be exposed to this process.”
The team says that while it was a challenge to meet the aggressive accuracy and run-time targets within the fixed year-long project — all while juggling Toronto/Seoul time zones and working under COVID-19 constraints — the opportunity to generate a practical solution for a world-renowned manufacturer will not soon be forgotten.
“It was good for us to understand how exactly industry works,” says Sudhakar. “LG’s goals were ambitious, but we had very encouraging support from them, with feedback on ideas or analogies to explore. It was very exciting.”
It may make you squirm to think about it, but your insides are home to trillions of organisms from hundreds of different species. Scientists are studying this “gut microbiome” as a factor in a range of human health conditions, but current techniques could use some improvement.
“Gut microbes in humans are typically studied using stool samples, which inform about microbes in the colon,” says Professor Eric Diller (MIE). “While colonoscopy could allow for access deeper within the intestines, it is of course highly invasive. This means there is no simple, non-invasive method to sample within the small intestine, which is the most important area of the microbiome to study and observe.”
Diller and his team have a solution: a tiny magnetic capsule about the size of a vitamin pill. When swallowed by a patient, the capsule slowly works its way through the digestive system. When it gets to the desired point, it can be activated by holding a large magnet over the abdomen.
“The magnet opens a trap-door on the capsule, which then closes to tightly seal the sample inside,” says Diller. “The capsule can then be recovered from the stool and sent to a lab for analysis.”
The project is one of seven from across U of T Engineering to receive funding from the Connaught Innovation Awards, The Connaught Innovation Awards recognize and support innovations that have strong socio-economic or commercial potential. A total of 10 research teams from across U of T will share up to $500,000 in funding in this year’s cohort.
Diller and his team will use the funding to create an even smaller capsule that is easier to swallow, as well as to develop a way to track the capsule and enable samples to be taken at a more precise location.
“Tracking will be done using a vest of magnetic field sensors worn by the patient, as well as with ultrasound,” he says. “Along with project co-investigator John Parkinson at SickKids, we have already developed a prototype which we tested in animals, and we found that we could indeed collect samples with our non-invasive method.”
If successful, the tool could be invaluable to researchers as they continue to study the role of gut microbiota in all kinds of health conditions, from irritable bowel syndrome — which affects 226,000 Canadians — to diabetes.
The other six projects funded by this year’s Connaught Innovation Awards are:
- Energy-efficient coded-exposure-pixel cameras for accurate imaging without motion artifacts — Professors Roman Genov (ECE), Andreas Moshovos (ECE) and Kiriakos Kutulakos (Computer Science)
- Computational framework for fast uncertainty quantification and decision analytics — Professor Prasanth Nair (UTIAS)
- 2-ZYME: Two-step biocatalytic conversion of underused biorefinery side-streams to glucaric acid — Professor Emma Master (ChemE)
- Development of an automated system for blastocyst biopsy with minimal invasiveness in IVF clinics — Professor Yu Sun (MIE)
- Process Scale up for a novel method of nickel extraction — Professor Mansoor Barati (MSE)
- Iron fortification of tea — Professor Levente Diosady (ChemE)
You can’t grow food without water, and in many parts of the world, there isn’t a lot to spare. Two U of T Engineering graduates have a solution they think could help.
Austin Mclean (MechEng 1T5+PEY, MEng 1T9) and Rashmi Satharakulasinghe (ChemE 1T7) are the co-founders of Corridor Water Technologies, a social enterprise that aims to commercialize a passive water irrigation controller for use in areas without electricity.
“Irrigation is water intensive, but it can be made a lot more efficient if you only turn on the water when the plants require it,” says Satharakulasinghe. “There are a number of smart sensing systems available to help farmers make that call, but right now they all require electricity, which isn’t always available to farmers outside of the developed world.”
The team’s solution grew out of research conducted over the past several years in the lab of Professor Amy Bilton (MIE). Bilton and her students made several trips to the community of Pedro Arauz, Nicaragua to meet with local farmers, understand their needs and test out various designs. Bilton continues to serve as a senior advisor to Corridor Water Technologies.
Known as a passive irrigation controller system (PICS), the team’s device includes a probe that is capable of sensing how wet or dry the soil is via physical and chemical characteristics. Using a property known as “soil suction pressure,” the device mechanically opens or closes a valve to regulate the flow of water, all without the use of electricity.
“We have been through numerous iterations of the design, experimenting with different materials and mechanical principles,” says Mclean. “We are still looking to improve some of the smaller mechanics of the design and integrate unique options where possible.”
Watch this video to learn more about the passive irrigation controller developed by Corridor Water Technologies.
By turning irrigation on and off automatically to maintain a desired soil saturation level, the PICS can reduce water use by up to 20% compared to current practices of flood irrigation. Farmers can “set it and forget it,” freeing up time for other tasks while secure in the knowledge that their plants won’t get too dry or too wet, and will grow at their optimum rate.
“We designed our device to be simple in terms of features, operation and maintenance,” says Satharakulasinghe. “It can easily be tuned to the different crops common in the community, or even adapted for new crops that would help farmers diversify their range.”
In June of 2020, Corridor Water Technologies placed in the top three at ISHOW USA, a competition organized by the American Society of Mechanical Engineers. The win netted them a $10,000 seed grant, which they have been using to further improve their design, as well as one-on-one coaching sessions with industry professionals.
“We have been fortunate that during the pandemic, we had access to a U of T greenhouse to do testing in,” says Mclean. “This summer we are planning to expand our field testing, hopefully including warmer areas closer to the conditions of our intended farming community outside of Canada.”
Satharakulasinghe says that in addition to agricultural benefits, the PICS could have several social benefits in the communities where it could one day be deployed.
“Often in these communities, it is the women of the household who run the farming activities while the men travel to the city for labour work,” she says. “By bringing additional income to the family, PICS could also promote gender equality.”
“These types of projects provide students with opportunities to work across cultures and disciplines, at the same time as addressing a hard and constrained engineering challenge,” says Bilton.
“Austin and Rashmi are doing a fantastic job, and it’s very rewarding to see a group of former students take some tech which was developed in our lab and turn it into a venture like Corridor Water. The reason we do the work we do is to have an impact.”
One year into the COVID-19 pandemic, vaccines are being deployed and administered around the world — with vaccine development, manufacturing and distribution taking place at record-breaking speed.
As Canada races to vaccinate its citizens amid an increase in variant infections, writer Liz Do spoke to Professor Omar F. Khan (BME), an immunoengineering expert. Khan, whose lab designs nanotechnology devices that can deliver RNA technology to cells for better disease treatment, explains common concerns and questions around COVID-19 vaccines.
There are four approved vaccines in Canada. How are they different and how do they work?
Two are mRNA vaccines — Pfizer-BioNTech and Moderna — and two are viral vector vaccines — Johnson & Johnson and AstraZeneca. They are different methods of delivering an antigen.
An antigen is a unique protein that is used to sensitize the immune system to a pathogen. This unique antigen is only found on the pathogen and nowhere else in your body. Once the immune system is sensitized to it, the immune system then goes on to make antibodies. Antibodies are specialized protein complexes that float around in your blood and if they ever recognize the antigen, they stick to it. That ‘sticking’ sets off a cascade of signals and the rest of the immune system converges and destroys the pathogen connected to the antigen.
The Pfizer-BioNTech and Moderna vaccines contain a piece of genetic information in the form of messenger RNA, or mRNA. This molecule is separate from DNA, so they don’t intermix. The mRNA molecule is an instruction set that your body can use to make the antigen.
The viral vector vaccines use a modified virus — not the virus that causes COVID-19, a different one — that has been changed so that instead of carrying instructions to make more of itself, it just carries instructions to make the antigen for the COVID-19 virus.
These vaccines were developed when there was one version of the virus. Now there are variants — how did this come about?
Because we’ve had so many cases of COVID-19, these viruses kept on replicating in people. Every time a virus replicates, there is an opportunity for a mistake to happen: this is a mutation. Most mutations are not favourable, but some enable the virus to survive and spread more effectively, and so become a favourable evolutionary event. In this case, a variant emerges that outruns the original version because it can spread more easily and make more copies.
Do we know how effective the vaccines are against new COVID-19 variants?
The antigen that we’re using as part of the vaccine is called a spike protein, which we hope is relatively invariant in that it shouldn’t change too much. So, with these vaccines, we’ve trained the immune system to recognize an earlier version of the spike protein. If the variant has a slightly different version, then the fidelity isn’t quite there.
One way we can overcome this difference is to have a very high immune response by overwhelming the variant virus with antibodies. Booster shots can help maintain a high level of antibodies in your blood and is the reasoning behind the recommendation for a second shot.
What this all ends up meaning: these vaccines can no longer completely prevent disease. However, they can still reduce the severity of disease. We are in the middle of a very dynamic pandemic, where the virus keeps changing.
What this means for the person receiving the vaccine is that you have to be more cognizant of what ‘efficacy’ means. Here, efficacy means a reduction in the severity of disease.
Why is there a public health push to get everyone vaccinated if we’re not completely ‘stopping’ COVID-19?
There are several reasons. Mass vaccination can slow down the viral replication in the population. That leads to fewer variants, and our vaccines stay more effective. We are preventing the severity of the disease, which means less hospitalization and fewer patients in intensive care. Hence, we do not overwhelm the health-care system. I think the major confusion is that the vaccines are primarily preventing severe disease, and not necessarily completely preventing disease.
Does it matter which vaccine you get?
It doesn’t matter. The goal is to stop viral replication and the creation of new variants. All of these vaccines have shown, through clinical trials, that they will reduce the severity of disease. Get whatever you can get, because we don’t want to be responsible for creating a variant that is even more pathogenic. Let’s try to all fight this.
There is also vaccine hesitancy, with concerns of health side effects from getting the vaccine. We’ve seen in Europe the suspension of the AstraZeneca vaccine due to reports of blood clots, for example.
It’s good to be concerned and good to ask questions. But let’s help people understand why these things are happening.
Vaccines are a very special case of health-care technology. There’s a special burden on the regulations for vaccines that says, ‘if you give this to a healthy person and they become sick, we need to figure out if they became sick because of your vaccine, or because people become sick from an unrelated reason’ and to decouple the two.
The AstraZeneca vaccine has been widely distributed around the world and administered to millions of people, particularly in the U.K. Of the people vaccinated there, they looked at these potential side effects and the rate at which they were occurring, in order to determine if there’s a link.
So when the U.K. did this investigation, they found that the rate of blood clots happening in people was not different than the rate at which it happened in the overall public. Other countries, such as in France and Spain, want to make sure they are making the right call out of an abundance of caution — and that’s understandable and important to be resolved.
These safety procedures are in fact normal; it’s just that the awareness has been amplified because these protocols are under a big magnifying glass.
When do you think Canadians will see some semblance of normalcy, given the rise of variant infections in parallel with the rate of vaccines being administered?
If we look at the rate of which vaccines are being shipped, how quickly people can be vaccinated, but then also factoring in how long it takes your body to develop immunity, I think we are looking at 2022. By the time we have enough vaccines to vaccinate all Canadians, we’re already pushing into the Fall.
How will mRNA technology change the treatment of diseases and vaccine development going forward?
This is a mosaic of technologies working together. We saw the different rates at which vaccines were developed, manufactured, distributed — not to mention the challenges with temperature-control distribution logistics — these are all important elements that are part of the ecosystem of pathogen preparedness.
I would say that of the more traditional vaccine technologies versus the newer technologies, I don’t think there’s necessarily a silver bullet that will supplant the other. But what this has shown is that we are fortunate to have many more new options and tools that improve vaccine accessibility globally and hopefully prevent this from happening again.
Only two years after its creation, U of T Engineering startup Reeddi, Inc. (pronounced “ready”) is well on its way to fulfilling its mission to bring sustainable, affordable electricity to places where reliable power is hard to come by.
“Right now, our technology serves a combined 600 households and businesses monthly in Nigeria,” says Olugbenga Olubanjo (CivE MASc 1T9), founder and CEO of Reeddi. “We have plans to increase that number.”
In 2017, Olubanjo’s first year at U of T, he would make phone calls to family and friends in Nigeria, where he grew up. Often, these calls would be disrupted by power outages that are all too common in that part of the world.
Those who can afford them buy diesel generators, but these units and the fuel needed to run them are costly, and they produce harmful emissions, including greenhouse gases.
Olubanjo knew that the cost of solar power had plummeted in recent years, and he wanted to make this technology more accessible to rural Nigerians. The solution he hit upon was a solar-powered “electricity bank” where portable power packs could be rented on a short term basis.
In its current iteration, a Reeddi bank contains 20 capsules, each of which holds about 250 Watt-hours of energy, enough to charge three mobile phones or power a laptop for four hours. Capsules are rented for 24 hours, after which they can be returned to the bank to recharge in the sun.
Olubanjo developed the company with support from U of T’s entrepreneurship ecosystem, including The Entrepreneurship Hatchery at U of T Engineering. He quickly started attracting attention: in 2019 alone, Reeddi won awards at the Cisco Global Problem Challenge, the MIT Clean Energy Prize competition, and the IEEE Empower a Billion Lives competition.
The past year has brought even more accolades. Last summer, Olubanjo and his team placed in the top three at ISHOW USA, a competition organized by the American Society of Mechanical Engineers. The win netted them a $10,000 seed grant, as well as the $1,000 “fan favourite” prize.
“We benefitted from fantastic insights from the ASME judges,” says Olubanjo. “We are equally excited to explore the potential networking and advisory opportunities that come with the prize to scale our venture for global impact.”
Olubanjo and his team have enrolled in Third Derivative, a technology accelerator focused on clean energy. The team also received a $25,000 award from the Nigeria Off-grid energy challenge and are currently among the finalists at the Royal Academy of Engineering African Prize. They have presented their innovation to a World Energy Council Panel group and at the Africa Indaba Energy Conference.
“Our current efforts are directed at upscaling local operations and manufacturing more Reeddi capsules for our customers in Nigeria,” says Olubanjo. “The future is bright.”