Data Driven Decision Making for Hospitals Part 6 – Machine Learning

In this article, we’ll cover another powerful computing tool used in data driven decision making – Machine Learning.

In its most basic context, machine learning is the ability of an algorithm to determine if a piece of data should be classified as either “True” or “False”.  How it determines that classification is driven by the meaning of “True” and “False” and one of many mathematical models that are selected based on the characteristics of the data. For example, if a patient has congestive  heart failure, a recent myocardial infarction, and has chest pain, he should be given immediate attention while a patient who presents with a rash can be designated as non-critical (Emergency Room triage is a good analogy for how ML works).   

Alternatively, the model can be based on continuous data where predictions are based on past performance (regression models).  In this case, historic data shows that the cost of medications rises three percent every six months. The model can determine what to expect for the next price increase.  

Rather than focus on the details of the models, focus on their application in business.

  • Classification systems can be used to identify clusters of clients or customers who have similar traits by using consumer loyalty and credit card data.  
  • Classification systems can be used to determine who receives particular promotional materials. 
  • A hospital can predict the number of emergency room visits based on the demographics of the area served. 
  • Classification process can be further subdivided to investigate questions such as “If there is a preference in treatment, what are the factors that drive this preference?”.
  • Consider if patient A has a pattern consistent with patient B but hasn’t demonstrated the same retreatment pattern as  B, can we determine what factors affected the outcome of A not seen in B ?

The above analyses make use of  both “structured” and “unstructured” data.  Structured data is available in systems that most businesses use to monitor revenue activities.  Using the data from their Electronic Medical Record (EMR) system, Enterprise Resource Planning (ERP) systems along with data from various data services, the hospital can “fuse” or add features about customers that extend beyond what is purchased.  The company begins to leverage features that are characteristic of the subpopulation that they want to target (eg, readmissions).

These examples focus on structured data.  Another popular use of machine learning is to evaluate and classify unstructured data or text.  This is frequently seen in business as sentiment analysis.  The process uses Natural Language Processing and Machine Learning to classify a comment as “positive” or “negative” (or whatever label the user wants to assign to the data).

In either case, machine Learning takes advantage of what machines do well – sorting data.  The systems “learn” based on how the user labels their data. It’s up to the user of the Machine Learning system to define the question that needs to be answered.  The Machine Learning system then sorts the data and provides the classified information to the user.

In the next article, we’ll explore how to assess the results of analyses to then make a decision.

NoviLens is an appliance for data driven decision making. NoviLens handles the tedium of the computing components of the process and frees up the decision maker to focus on their problem solving tasks.

Visit to learn more about NoviLens, and join our email list to get updates on new features and case studies. If you’d like to learn how NoviLens can help your organization, contact us for a 30 minute demo at