Brought to you by NumFOCUS Foundation and O’Reilly Media Inc.
The official Jupyter Conference
August 22-23, 2017: Training
August 23-25, 2017: Tutorials & Conference
New York, NY

Intro to Machine Learning for the Healthcare Professional

Moderated by: Harold Mitchell

Who is this presentation for?

Primarily healthcare professionals, but researchers and academics as well. Or Anyone.

Prerequisite knowledge

1) General understanding of data sets (i.e. rows, columns) with a class or response variable 2) Familiarity with mean, standard deviation, percentiles (will touch on) 3) Familiarity with metrics (e.g. ROC curves, confusion matrix) 4) Basic knowledge of computer programming

What you'll learn

1) The understanding of how to apply a classification algorithm to a data set with the goal of predicting an outcome. 2) The ability to implement tools to visualize data 3) The general knowledge how-to use of common machine learning tools like Jupyter, Python, SciKit‐Learn, pandas, numpy, and matplotlib

Description

There is so much data collected in spreadsheets and CSV files. In many cases, this data is simply used to build some reports that summarize some historical or current phenomena of interest. However, this data might possess useful patterns that would provide insight into the future. So imagine data that can not only provide a historical and current account of events, but insight into to future events as well. Undoubtedly, in the past we relied primarily on statistics to do this. But today, we have a “new sheriff” in town – machine learning. Machine learning brings together statistics and computer science to perform some unbelievable predictions on all kinds of data. Moreover, it is revolutionizing healthcare from predicting hospital readmission to heartbeat detection.

The goal of this lesson is to teach the healthcare and research professionals how-to build their own predictive model given a data set. This will be accomplished by taking the learner on a journey through the world of learning from the eyes of a machine. So, how does a machine learn data? We’ll answer this question. Being presented will be a step-by-step demonstration of how machine learning can be applied to a data set to produce a predicted outcome. The tools that will be used include: a publicly available data set, Jupyter notebook, Python programming, SciKit-Learn algorithms, panda dataframes, numpy arrays, and matplotlib graphs. By the end of the lesson, each participate should have a good understanding of how-to apply machine learning methods to predict an outcome using these tools. Finally, the lesson will conclude by providing links to materials from this lesson and to information on opportunities for on-going study in this emerging and exciting field.