Machine Learning in Medicine
Weekly Roundup #1
Machine learning is revolutionizing healthcare. Here are some highlights of interesting applications of how research in artificial intelligence is helping people.
1) Preventing selfie related deaths
Yes, believe it or not, taking a selfie can be dangerous. Since 2014, A shocking 124 people have died from snapping a pic in dangerous places. In this paper computer scientists determine if someone is at risk while taking a selfie based on their location, what is in their background, and the text related to their post. Read more here:
Me, Myself and My Killfie: Characterizing and Preventing Selfie Deaths
2) Learning from your retina
Ever wondered what is so futuristic about retina scans?
The retina is a complicated network of blood vessels. That network holds rich information about different diseases such as diabetes, hypertension, glaucoma, to name a few. In this paper researchers describe a way to
train algorithms to outline (segment) and identify images of the retina.
3) Automatically detecting when people fall
The over 80 year old population is the fastest growing demographic in the world. Unfortunately among this age group, 60% of accidents occur at home and 80% are due to falls. Often when elderly people fall they may be alone and not get help as quickly as they need. These researchers have come up with a clever way to detect falls using tiles implanted with accelerometers. When someone falls on the floor it exerts a pattern of force that is unique and distinct from walking, jumping, etc. A fascinating way to detect and get help quickly to elderly people who have fallen.
Read more here: Automatic Fall Detection System using Sensing Floors
4) A collective intelligence messaging system to help people quit smoking
