Seth Clifford:

Apple: ResearchKit and CareKit. Centered around individuals, reporting personal data. Assembling tons of it, and allowing for better personal follow through on long-term treatment, and more individualized reporting for research purposes. Gathering of this data is done through traditional channels, but by allowing users to have agency in these processes, Apple affords people the ability to contribute to a large data set, but safely remain an identifiable component variable.

And:

Google: machine learning to aggregate data against the treatment of extremely difficult ailments (diabetic retinopathy was the example presented in the keynote). Very few doctors can detect it accurately, and it’s very hard to do right/well. And this small number of doctors can’t be everywhere at once. But put enough data into a machine and it can pattern match the very intricate details–perhaps better than people, and everywhere at once (since people can only be in one place at a time). Throw incomprehensible amounts of information at an enormous amount of computing power and basically brute-force a treatment protocol that functions better than humans ever could.

Apple, focused on the individual, protecting their privacy. Google, focused on a problem and its associated data. Both approaches valuable and complementary.

Interesting post. [Via MacStories]