This project aims at evaluating whether recent results in deep learning models, trained to exploit weak labels (Hwang, 2016) can serve to extract meaningful lesion localizations from image-level labels, either from individual scans or given a (longitudinal) sequence thereof. To this end, we will scale up existing models that have been shown to work on 2D images to a 3D context, studying labeling performance as the dataset size grows.
Online communities abound today, arising on social networking sites, on the websites of real-world communities like schools or clubs, on web discussion forums, on the discussion boards of videogames, and even on the comment pages of news sites and blogs. Some of these communities are “healthy” and foster polite discussion between respectful members, but others are “toxic” and devolve into virulent fights, trolling, cyber-bullying, fraud, or worse even, incitation to suicide, radicalization, or the sexual predation and grooming of minors.
Some diseases and brain injuries can seriously impair language. Patterns in an individual's speech can allow computers to describe these impairment with a high degree of accuracy. These techniques can be used to test large groups of people for drug trials and potentially replace pen-and-paper based testing methods. To fully automate this process, speech recognition systems can be used to automatically transcribe speech. Unfortunately, these technologies continue to perform relatively poorly for elderly speakers, or for individuals with speech disorders.
Blockchain is an emerging technology that has the potential to change the way financial participants transact with each other. It enables direct transfer of value and financial assets between participants over networks without the need for a central authority (internet of value). It does this by combining the functionality of different technologies - distributed systems, smart contracts, mutual consensus verification, and cryptography. Given its potential Scotiabank is investing in technical research and business application.
TurnMeUp is an iOS app for always-on voice communications. Users leave the app running in the background and can talk to the recipient (also using the app) at any time. This app is especially useful for coworkers listening to their own music in the background without needing to enter and exit voice call sessions manually. To conserve bandwidth and ensure that users listen to music without being unnecessarily interrupted, TurnMeUp sends voice signals to the recipient only if the user is speaking.
We are conducting research on using techniques from Artificial Intelligence (specifically Machine Learning, Reinforcement learning, and computer vision) to automate the ability of a robotic arm equipped with a hand-like gripper to pick a wide variety of items. The robot uses the visual scene, provided through cameras, in order to choose which item to pick, and needs to then plan and execute a grasp. This is an open research problem at the cutting edge of robotics and AI and we plan to use a combination of state of the art academic research as well as internally developed algorithms.
The company wants to develop a state of art recommendation system for the clients. A recommendation system is a piece of software that provides productsâ suggestions to customers on a website. For example the products suggestions that can be seen on Amazonâs web page are generated by its recommendation engine.
The typical recommendation engines work by utilizing the existing user-product preferences information. They recommend products to a user by comparing his preferences to other similar usersâ preferences. The typical example of this is Users who bought item-A also bought item-B.
Developing a model for a system can come with a lot of uncertainty, especially in the early stages of development. Recent research has be done into removing uncertainty during early stage models. Doctalk plans to use modern research to develop a viable product for market, while contributing to the process of the research being applied to the development of the product.
Using web crawling technology in coordination with state of the art machine learning techniques, the project aims to mine useful, structured information about the worldâs suppliers from the web. Recent advances in artificial intelligence have increased the viability of such autonomous systems for extracting coherent information from arbitrary human-produced content. By leveraging these technologies, our goal is to build improved supplier discovery and recommendation systems.
During surgeries, it is important to keep track of what is happening with the patient, the steps being taken during the surgery by the operating staff, and unforeseen events that occur. All the previous correspond to the surgical workflow. Keeping track of the workflow is essential to achieving a better and safer surgery. In the past, computational tools have been developed to track each step the surgeon takes during the surgery, and dividing the separate surgical phases. However, the adverse events have not been tracked.