Power BI Research: Improving Healthcare in Europe

Our health is our wealth. Apart from being the key to one’s happiness and well-being, health contributes to our productivity, economic progress, and longer lives.

We thought: “What if we can analyze the healthcare policy and mortality trend?”

So we chose to conduct a healthcare market research and analyze the mortality trend statistics in Europe; along with this, we used machine learning for healthcare.

There is historical data of past years, so there is a chance to lower this indicator for the present time – for example, by paying special attention to groups with a high death rate. No need wondering how to make a decision tree for such research – keep on reading!

Location: Ukraine
Industry: Healthcare

Client Goal:

The main goal was to analyze the mortality statistics for the past years and determine the causes of death. We aimed to apply business intelligence for the healthcare industry and make research in different gender and age groups. We were looking for this information, as far as it is possible to improve healthcare and lower the mortality by paying special attention to the risk groups.

During our work, we used business intelligence and analytics in healthcare. We used analysis tools such as:

Maps for displaying countries and numerical characteristics.
Time series for analyzing mortality by year.
Decision tree in machine learning (PowerBI) for mortality rate modeling.
Hexbin scatterplot for analyzing data with several numerical characteristics.
Heatmap for visualization of the dependence of various characteristics and classification of diabetes (high, medium, low incidence).

The business intelligence in healthcare study process included the following steps:

Results:

Our visualization showed the mortality per 100,000 people in Europe, so we were able to determine which countries need reforms and measures to reduce this indicator. Among these are CIS countries, Malta, Romania, Latvia, Montenegro, Bulgaria, Poland and Hungary.
The statistics showed that the issue of diabetes is an important and relevant topic of healthcare.
The decision tree in machine learning helps to simulate the mortality rate per 100,000 people by selecting a country.
Cause-of-death statistics help health authorities to determine the focus of their public health actions – for instance, starting a vigorous program to prevent particular illnesses.

Business intelligence software for healthcare (Power BI) technologies are popular and widely used in business and social researches. You can use these instruments for any statistical tasks. Our Health BI Study only proves that the Power BI instruments and machine learning in healthcare are perfect for any kind of such data analytics.

Our process

Timeline:

2019

Team:

1 machine learning developer

Technologies we used

Server-side

Pandas