AI for Healthcare with Keras and Tensorflow 2.0: Design, Develop, and Deploy Machine Learning Models Using Healthcare Data

© Anshik 2021

AnshikAI for Healthcare with Keras and Tensorflow 2.0https://doi.org/10.1007/978-1-4842-7086-8_6

6. Handling Availability of Low-Training Data in Healthcare

Anshik¹  

(1)

New Delhi, India

 

The availability of training data is a critical bottleneck in machine learning applications. This is further augmented by working in a specialized domain like healthcare where one needs to be highly skilled to understand the data and then tag or label it for machine learning to use. In addition to finding a skill steward, there is a heavy investment in terms of time and cost for the organization.

You have already learned one way of handling availability of limited information, which is by transfer learning. Unlike transfer learning, which is an algorithmic ...