The IT Nerd

A new virtual reality (VR) training lab at the University of Waterloo’s School of  Optometry and Vision Science, will help Canada’s next generation of optometrists learn how to diagnose vision problems and eye diseases more quickly and accurately.

The new lab, funded through an $800,000 investment by national eye care provider FYidoctors, will accelerate the development of essential clinical skills for optometrists.  The authentic experience simulates a patient’s eye and provides students with an opportunity to use virtual reality to practice on many cases that range from wellness through to diseases such as macular degeneration and diabetes.

At a total cost of $1.5 million, the FYidoctors Simulation Lab is the first of its kind in Canada and will ensure the School remains at the forefront in optometric education in North America.

The simulation equipment will supplement real-life training, which is limited.  It will also provide hundreds of real patient cases to practice binocular indirect ophthalmoscopy (BIO), which will help students to recognize, identify and diagnose eye and systemic disease more quickly and confidently.  FYidoctors leadership support represents the most significant sponsorship every received by the Schools optometry program.

The University of Waterloo is one of two optometry schools in Canada, making the University of Waterloo a national resource for training vision care professionals and vision science researchers.

The Equipment:
The lab will initially include 5 Eyesi Binocular Indirect Ophthalmoscopes (BIO) are state of the art augmented reality simulator for training of retinal examinations and provides a highly realistic and dynamic 3D simulation of the anatomical structures of the eye and ophthalmoscope optics.

Phase two of the lab, to be unveiled at a later date, will include the addition of the Eyesi® Slit Lamp simulators.  This technology will allow students to practice basic handling of the device and skills required to conduct a corneal exam, retinal exam and Gonioscopy & Tonometry. Through multi-tiered training provided by the Slit Lamp simulation technology, it will promote pattern recognition for retinal pathologies and provide competency-based assessments.

With Eyesi BIO and Slit Lamp technology, students can examine a wide range of clinically relevant cases before they meet their first real patient. The simulated cases are based on real patient histories and were developed in cooperation with universities and eye specialists.

Simulation technology provides an authentic experience that offers the look and feels for the diagnostic skills of BIO and slit lamp examination.  Students can practice as much as they need to reach certain milestones in a modular system that progresses in difficulty and complexity.  The process begins with mastering the technical skills to image the fundus, and later incorporates real cases and images for examination and diagnosis.  The goal is to use precious laboratory time with faculty to transition students from good foundational skills to great diagnostic skills.

The conditions and diseases it will train students to diagnose:
Conditions and diseases that can be assessed and diagnosed through the simulation technology range from wellness through diseases.  The simulation technology focuses on looking in the back of the eye to identify a wide variety of eye and health-related problems.

Examples include: 
Vision:
• Reviewing and assessing healthy eyes
• Trauma – concussion
• Foreign bodies
• Amblyopia – lazy eye, a vision development disorder
• Uveitis – inflammation of the middle layer of the eye
• Keratoconus (curving of the lens)
• Detached retina

Diseases include:
• Macular degeneration
• Diabetes
• Hypertension
• Cancer
• Glaucoma

Popular methods of protecting smartphone personal identification number (PINs) may only be successful in safeguarding your personal information 20 per cent of the time, according to a new study out of the University of Waterloo.

The study found that methods such as tilting the smartphone, a widely adopted defence strategy, does not guard against people close to you such as romantic partners and co-workers who might be angling for access to your device.

The study also found that even when the attacker is observing from across the room they still have a good success rate of stealing your PIN from a distance.

In conducting the study, videos were recorded of 30 people entering a PIN from different positions with different conditions, such as the screen of the device tilted away from the camera. Thirty attackers were then recruited to mount over 1,000 shoulder surfing attacks, which involved watching videos of users entering PINs on a phone.

The researchers found that attackers who paid attention to the pattern of relative finger movement, movement in direction and distance relative to the previous tap, were more successful than the attackers who guessed only based on the current position of the finger and the layout of the keypad.

With attackers having to observe the victim entering their PIN only four times or less to figure out PINs 80 per cent of the time, even when the device is tilted, Khan said a better mechanism than tilting the device screen away needs to be considered.

The study, Evaluating Attack and Defense Strategies for Smartphone PIN Shoulder Surfing which was co-authored by Khan, Urs Hengartner and Daniel Vogel, all of Waterloo’s Cheriton School of Computer Science, was presented at the 36th Annual ACM Conference on Human Factors in Computing Systems (CHI 2018).

The University of Waterloo launched the Waterloo Artificial Intelligence Institute today, bringing together a large group of researchers and businesses to advance technology and prepare Canada for future economic disruption.

Waterloo AI uniquely focuses on end-to-end innovation ranging from foundational to operational AI. Foundational AI advances the field as a whole through research in a number of areas, including statistical learning, deep learning, game theory and data science. Operational AI develops scalable, secure and transparent solutions for a wide range of applications.

As part of its mandate, Waterloo AI will pursue new areas of research with societal and business impact including healthcare, environmental protection, urban planning, manufacturing, autonomous systems and human-machine interaction and will emphasize timely access to expertise to individuals and industry.

Waterloo AI is a joint venture of the University’s Faculties of Engineering and Mathematics and will also include researchers from Arts, Applied Health Sciences, Environment and Science. It is affiliated with a number of existing labs, centres and institutes at Waterloo.

Waterloo AI is led by two co-directors: Fakhri Karray, professor of electrical and computer engineering, and Peter van Beek, professor of computer science.

Sponsors of Waterloo AI include General Motors Canada, Huawei Technologies Co, Loblaw Companies Limited, Nutrien, Shopify, Borealis AI, Magna International, Miovision Technologies, NexJ Systems and Scribendi.