Online education has become mainstream in 2020 and 2021. Students of all ages have become accustomed to logging onto online meeting software such as Zoom, Microsoft Teams, Google Meet (and many others) to attend their classes virtually. Educators too have become accustomed to this new method of course delivery, using a combination of live webcam video and the screen sharing capabilities inherent in all modern online meeting platforms.
While sharing your screen to present a slide deck is perfectly fine from an image quality perspective, it simply doesn’t meet the needs of microscopy education. During the process of teaching microscopy, educators often load whole slide images (WSI) relevant to the course material, and walk through an analysis of those images – zooming in on features of interest, panning to others, measuring the sizes of, and distances between features, and so on. This style of education requires that ultra-high-quality WSI imagery be delivered to the student’s workstation with as little latency as possible, wherever that student may be – a level of quality which is far from guaranteed with platforms like Zoom, Teams, Meet, and others, especially with highly-detailed slides which can be many gigabytes in size.
Consider that students are logging into your online classroom using a variety of internet connection types and speeds, and are likely sharing their connection with multiple people simultaneously. Educators face many of the same challenges – they may be located in an urban area with exceptional internet bandwidth, but may be sharing their bandwidth with their children and spouse who are in online meetings at the same time the educator is teaching the course. Students may be sharing their internet connection with roommates, or may be connecting via coffee shop Wi-Fi, or rural internet connection.
In cases where bandwidth (either from the educator to the internet, or from the internet to the student) is restricted, online meeting software will prioritize the voice connection, and degrade the webcam video and screen sharing connections. For most meetings, this is fine – the brain has a difficult time with laggy or garbled speech, so the audio channel is prioritized - but if a few frames of the educator’s webcam drop, or the letters in a PowerPoint presentation are fuzzy for a few seconds, very little is lost.
This tradeoff simply isn’t acceptable when teaching microscopy – the quality of the images is paramount.
Here are the two main ways online meeting software compensate for a low-bandwidth connection (or one which temporarily has reduced bandwidth):
Strategy | Effect on Web Meeting | Effect on Microscopy Education |
Reducing frame rate. | Minimal.While ideally, we want 30 frames per second (fps) for screen sharing, momentarily reducing it to 2-5 fps, or even pausing video output for 30 seconds poses little problem when the slides in a slide deck are only changing once every 1-2 minutes. | Unacceptable.When a whole slide image (WSI) is being presented, the educator may be zooming and panning through a slide, moving constantly (but slowly) at some points.If the web meeting platform can’t keep up, the student will be quite confused. |
Reducing image quality by blurring image details. | Minimal.Compression algorithms are designed and tuned so that the blur isn’t noticed by the human brain.These algorithms do this quite well for faces and text (e.g. presentation decks). | Unacceptable. Correct interpretation of a WSI requires the student to see the minute details of the image – blurring the details will cause students to miss key insights, and to come to the wrong conclusions. |
Aurora mScope Education is an online learning toolset designed specifically for microscopy-based educators. Its interface is optimized to deliver the highest-quality imagery over a wide variety of bandwidths, with as little latency as possible. With Aurora mScope, users are only sent the pieces of the image which they are currently viewing – not the entire multi-gigabyte image. This allows them to start reviewing the image right away, and when they zoom in or pan, the system sends them additional pieces.
This is very similar to the very efficient way Google Maps works. Google has street-level maps for every major city which are hundreds of gigabytes in size. Is all of that downloaded to your workstation right away? Of course not – Google loads a high level map in your browser to start. Then when you zoom in, pieces of maps with higher detail are loaded. In this way, Google ensures that the Maps GUI is extremely responsive to user input, even over low bandwidth.
A short, 3 ½ minute video of some of the ways Aurora mScope can enhance your microscopy courses.
An article by Dr. Erik Yeo (Professor of Medicine at the University of Toronto, and CEO of Aurora mScope) on cloud-based learning in microscopy.
Comments