A Brief Introduction into Machine Learning
Machine Learning is, in essence, a way to accurately predict and classify something based on previous training data.
For example, if you wanted to train a model to detect if an image contained an apple, you’d supply training images of apples and instruct the model to remember what they are.
Eventually, the model should be able to accurately detect images with apples without your assistance.
Case Study: Using ML to diagnose facial disorders
The applications of ML in genomics is limitless, but one strong example is automatically detecting phenotypes and classifying them. For example, the shape of one’s nose, the color of one’s eyes, etc., are all phenotypes that are commonly varying between people.
This information can be extracted by training the ML model to recognize such features (through nodes as seen by the green dots). However, the real benefit is in disease tracking, as any abnormalities are surely to be detected and flagged, saving time and money from patients and doctors.
Works Cited:
“Artificial Intelligence, Machine Learning and Genomics.” Genome.Gov, www.genome.gov/about-genomics/educational-resources/fact-sheets/artificial-intelligence-machine-learning-and-genomics.
Gurovich, Yaron, et al. “Identifying Facial Phenotypes of Genetic Disorders Using Deep Learning.” Nature News, Nature Publishing Group, 7 Jan. 2019, www.nature.com/articles/s41591-018-0279-0.
Google, Google, teachablemachine.withgoogle.com/.
コメント