Deep and Shallow Learning Projects

We’ve talked before about the importance and increased prominence of AI across industries, but for the most part, it’s still been something that has been looked at as the technology of the future rather than the present.  Partners for Kids, associated with the Nationwide Children’s Hospital in Ohio has been looking to implement AI on a much more accessible basis.

Traditionally, when people think of AI, they think of massive data sets that allow machine learning to make better decisions as a result of insights that can be found within the data.  At Extract, we certainly feel that a large data set can help us better identify information in unstructured documents that needs to be identified, incorporated into a medical record, or even redacted.

What Partners for Kids, the group featured in the article, is positing though, is that AI doesn’t have to be used exclusively for deep learning projects.  There is still plenty to be gained from limited data sets that can influence management and care decisions.

What this ‘shallow’ learning project is concerned with is predicting utilization and costs.  Budgeting can be extremely difficult for healthcare organizations to do, and better and more predictive data can give a group a more accurate picture of what their expenditures will look like for a given year.  Whether it’s getting an idea of repeat hospitalizations or which patients have the potential to be very high cost, it’s data that hospital systems haven’t always been able to predict.

Extract certainly believes that we can use data to help identify where critical or sensitive pieces of information can be found, and that shallow learning is certainly a great start to being able to more accurately read your incoming documents.

What we also believe though, is that there is a wealth of data that comes from scans, faxes, and paper documents.  This is why we automate the process of getting this information into your system so it can be used for projects within your organization to not only predict expected costs, but also patient outcomes.

Reach out today if you'd like to learn more.


About the Author: Chris Mack

Chris is a Marketing Manager at Extract with experience in product development, data analysis, and both traditional and digital marketing.  Chris received his bachelor’s degree in English from Bucknell University and has an MBA from the University of Notre Dame.  A passionate marketer, Chris strives to make complex ideas more accessible to those around him in a compelling way.