The challenge of training your orthopedic surgery AI.

Contrary to popular belief, training an AI for surgery is not easy.

So how long does it take for a surgery AI to be trained?

It could take months or years. The answer is dependent on specific hardware, optimization, number of layers in the neural network, size of your dataset, and more.

How much data is needed to make an AI a useful tool for a specific orthopedic surgery in real time?

  • 100 surgeries?
  • 1,000 surgeries?
  • 10,000 surgeries?

Read the real life example below to better answer the AI training questions.

Real life example in agriculture

In northeast England, halfway between Norfolk and Yorkshire, an AI-powered robot spends its days looking at strawberries. It’s not as easy as it sounds.

A human farmer can gauge a strawberry’s ripeness level by sight and weight, but the process involves putting each strawberry on a scale, which can be destructive and time-consuming. The robot can do the same job for up to 4 million strawberries a day by performing a simple scan of the fruit, undisturbed.

Behind the bot…FruitCast, the agricultural AI startup behind the robots, taught its bots how to do their jobs with data from V7 Labs, a London-based startup that helps AI companies automate the training-data process for models.

Training can be one of the most labor-intensive parts of getting an AI system off the ground, since it often calls for not only time and resources, but also vetted and relevant data.

Since its 2018 debut, V7 has used its computer vision platform to train AI models to identify everything from lame cows to grapevine bunches, depending on the client’s needs. In 2020, V7 raised a $10 million total seed round, and so far, its clients include more than 300 AI companies, as well as academic institutions like Stanford, MIT, and Harvard.

How it works.

To create that training data, V7’s model starts off with a “continual learning” approach. That could begin with subject-matter experts in, say, horticulture, drawing boxes around images of fruit and classifying it by ripeness level (e.g., a “level-3 strawberry”). They then either accept or correct each of the model’s attempts to do the same.

After about 100 human-guided examples, a model is able to make relatively confident classifications, so it transitions into what its CEO Alberto Rizzoli calls a “co-pilot approach”—for any given choice, the AI provides its confidence score and the human makes corrections.

The company finds human experts via a network of business process outsourcing companies, agencies, and consultants, which Rizzoli said can find a group of labelers on most topics within 48 hours.

Big picture: “The robots are kind of stupid until you put the intelligence on them,” Raymond Tunstill, CTO of FruitCast.