Tensor based machine learning with human computer interaction applications

In this research, first we will using tensor network techniques speedup the processing in neural network since computational cost is a major bottleneck in neural network based deep learning. Note that the weight matrix in each layer of neural network is with huge size, because of millions of parameters. This may cause large time complexity to calculate all the parameters. To overcome the large time complexity, we could use tensor decomposition to calculate the low-rank weight matrix, to reduce a large amount of parameters, therefore to reduce the time complexity.

Development of nano-colloid with gold nanoparticles to detect Legionella pneumophila using the principle of localized surface plasmon resonance

Legionnaires is a disease caused by the bacteria Legionella pneumophilia present mostly in aquatic environments. The first outbreak of this disease was recognized in 1976 in Philadelphia and the most recent one in July 2019 in Atlanta. Diagnosis of the disease isn’t early and thus need to be prevented by regular treatment. Treatment of water needs information about the water quality which needs on-site based sensors to detect the different pathogens present. At present, there is no viable solution to detect Legionella on site with confidence.

Realizing Large-Scale Production of High-Quality Human Induced Pluripotent Stem Cells through a Systematic Investigation on the Characterization and Protocol Optimization of Single-Use, Scalable, Vertical-Wheel Bioreactors

For stem cell discoveries to translate into improved health solutions for Canadians, we must use engineering manufacturing practices to grow enough cells safely and efficiently. Pluripotent stem cells (PSCs) have the unique ability to transform into any cell in the body when subjected to specified environmental conditions. They are invaluable in studying disease and gene functions and can be differentiated for potential use in transplantation.

Automated Multi-Factor Medium Optimization of Human Hematopoietic Stem Cell Expansion Using Design of Experiments

This project aims to support ExCellThera's current clinical trial (ECT-001) to improve blood stem cell transplantation outcomes for patients suffering from a host of blood malignancies. A critical aspect of developing clinical therapies using stem cells involves accurate and robust control of stem cell behavior in controlled laboratory conditions. Cell growth conditions (media) should mimic the native environment of the cell of interest.

Development of a personalized in-vitro model for dystrophic muscle endogenous repair for drug discovery

Duchenne muscular dystrophy (DMD) is a genetic disorder resulting in progressive muscle degeneration. Satellos Bioscience Inc is developing small molecule drugs that target and modulate DMD muscle stem cells to repair the dystrophic muscle. Despite the advantages of DMD animal models, they are not always predictive of human DMD phenotypes, or their response to drug treatments.

RALI-DX: Rapid Diagnostic Tools for Acute Lung Injury

Sepsis, severe pneumonia and respiratory viruses are key causes of Acute Respiratory Distress Syndrome (ARDS)—a detrimental lung condition that results in >3 million ICU admissions/year and can lead to organ failure and death. In Canada, ~18,000 patients with ARDS are admitted to hospital annually and those requiring critical care place an immense burden on hospital resources. Early detection of ARDS is vital for treatment to be effective.

Blood flow within microfluidic conduits

A current area of focus at SMT Americas is to develop injection-moulding methods for medical lab-on-chip (LOC) devices, particularly to transport and process blood and plasma while eliminating conduits clogging and reducing sample loss. This project will evaluate the effect of flow conditions on viscosity of samples, and the adequacy of conduit structures achieved with SMT Americas’ technology for cell separation.

AccessiBelt - Car Seatbelt for Users with Limited Hand Dexterity

Our hand function can be affected by different diseases and disorders, such as arthritis and neurological injuries. Having limited hand function will immensely affect our life and will make everyday simple tasks challenging. There are about 1 million Canadians who reported struggling with daily activities due to this limitation. One task that is challenging for a person with limited hand function is buckling and unbuckling a car seatbelt. Car seatbelts are life-saving devices and are crucial for the protection of passengers in a car.

A cloud?based, multi?modal, cognitive ophthalmic imaging platform for enhanced clinical trial design and personalized medicine in blinding eye disease

Age Related Macular Degeneration is the leading cause of irreversible blindness in Canada and the industrialized world, yet there are no treatments for most patients. Tracery Ophthalmics has developed an imaging method, “AMD Imaging” (AMDI) that has generated unprecedented pictures of the retina (film of the eye) that include both known and unknown “flavours” of disease (phenotype). These complex images will be compared against an individual’s genetic makeup (their genotype) and their concurrent illnesses, medications, and lifestyle history (their epigenetics).

Three dimensional (3D) ultrasound image analysis of the female pelvic floor

Pelvic organ prolapse (POP) is the descent of the pelvic organs into the vaginal canal, affecting up to 50% of the female population. Pessaries are intravaginal medical devices used to alleviate the symptoms of prolapse. However, pessaries are fit through trial and error and finding the right fit is not guaranteed for all patients. Cosm Medical aims to develop custom pessaries to overcome these challenges. The process of developing custom pessaries involves a novel ultrasound imaging technique to provide quantitative measurements of the vaginal canal.