A Novel AI-Based LED Grow Light

Indoor farming offers the opportunity to grow produce close to the consumer and has recently been gaining attention due to its efficiency, controllability, and sustainability. In indoor farming, food is grown indoors in a controllable environment, with stable environmental conditions. This approach reduces the land footprint required, the demand for energy and water, the nutrient input requirements, and can eliminate the need for pesticides.

Machine learning-based optimization of a small-molecule suppressor of the cellular prion protein

A reduction in the levels of the cellular prion protein (PrPC) is expected to ameliorate cellular toxicity in both Alzheimer’s disease (AD) and prion diseases. The latter are invariably fatal diseases that include Creutzfeldt-Jakob disease in humans and bovine spongiform encephalopathy (BSE), also known as ‘Mad Cow Disease’, in cattle. To identify a rational method for reducing PrPC levels, the Schmitt-Ulms group has been studying the evolution, function and molecular environment of PrPC for more than ten years.

Micro- and Nanofluidics to Measure Fluid Performance in Unconventional Hydrocarbon Extraction

Oil recovery from underground reservoirs with small pore-scales can have environmental impacts that can be minimized with prior knowledge of the physics behind fluids/fluids and fluids/rock interaction and the type of fluid to employ for oil extraction. Currently, these analyses can be performed in laboratories at reservoir conditions with rock samples in large pressurized vessels capable high pressure (~15 MPa) and temperature (~150 C). However, these measurements take weeks to complete and there is a test-to-test variation due to the lack of repeatability in the rock sample.

Development of a microfluidic point-of-care diagnostic device for the discrimination between viral and bacterial infections as a means to reduce antibiotic resistance

The discrimination between viral and bacterial infections has long been a goal in the field of point-of-care (PoC) diagnostics. Such a diagnostic tool would prevent the over-prescription of antibiotics, a leading cause of antimicrobial resistance. Current standard methods involve sending patient samples (throat swabs, blood, urine) to specialized clinical labs. This usually involves expensive and time-consuming protein biomarker assays to identify the cause of infection.

Plant community responses to climate change in the northern boreal mountains

Globally, mountain regions (especially those at high latitudes) are undergoing rapid environmental change and plant communities are expected to respond by changing their locations or timing of flowering. Exactly how future plant communities in these regions will look and function is unknown but has important implications for local animals and human communities. Through this project, we will focus on understanding the response of high-latitude plant communities to ongoing warming and predict how these communities will continue to change.

Deep-Learning for Distributed Intelligent Systems with Applications in Robotics and Computer Vision

Agile manufacturing via adaptive robots is the provision of Industry 4.0 for advanced manufacturing that enables more efficient, lean and cost-effective production. It is considered to be the ultimate solution for mass customization of many manufacturing industries such as aerospace industry hindered by their heavy reliance on manual labor. The current practice of programming a robot for every specific task is limited, if not futile, in the many manufacturing industries.

Development of MCT4-targeting small molecule inhibitors for management of castration-resistant prostate cancer

Late-stage, therapy-resistant prostate cancer (PCa) remains a difficult-to-treat disease that urgently needs better therapeutics. Advanced PCa cells use glucose (sugar) differently than normal cells, substantially increasing lactic acid secretion into the surrounding environment. This supports cancer growth in numerous important ways, including helping PCa avoiding destruction by the patient’s immune system. One critical protein involved in this process is MCT4, which transports lactic acid out of cells.

Stratifying colorectal cancer liver metastases using unsupervised clustering of quantitative imaging phenotypes

Personalized and precise treatments are the keys to improve prognosis of cancer patients and are also the main strategy of Sunnybrook Hospital. This project aims to stratify patients with colorectal cancer liver metastases (CRM) based on their disease subtype and risk using magnetic resonance imaging, which is routinely used in the diagnosis, staging and operative planning.

Classifying Innovation Management Forms Using Ontology Reasoning

In this project, the goal is first to design a domain ontology that models the innovation management forms semantically. At this step, the ontology contains domain-specific background knowledge, which is expressed using terminological statements. Then every completed form and the value of its fields are asserted as instances of different concepts of the ontology. Afterwards, an ontology reasoning algorithm is deployed to classify every completed form into various categories defined in the ontology.

Assessing and Identifying Clinical Dead-ends in Intensive Care Settings

type of treatment they will provide to patients. With technological improvements and the availability of a significant volume of data, it is increasingly difficult for care providers to properly evaluate and analyze the options available to them. The current health condition of the patient--reflected in the monitored observations which are recorded in EMR--may depend on all the relevant information from all prior observations and selected treatments, not just those most immediate (e.g., the trend of various health measures).