From Higgs to Healthcare Big Data for better business and society! Egge van der Poel Parttime Clinical Data Scientist @ Erasmus MC, Parttime Freelance Data Scientist @ EggeWel, fulltime curious
ANP
De Telegraaf
That awful feeling It could have been so much better
Storyboard Personal Introduction Big Data @ CERN Big Data @ KPMG Big Data @ Healthcare
Personal Introduction
Who am I?
What is Big Data? For me, Big Data represents the emergence of the digital enterprise the ability for an organization to take full advantage of its digital assets which collectively can be described as large amounts of data and more. Bourne PE. J Am Med Inform Assoc 2014;21:194
Big Data @ CERN
Why do p have article s mas s? 10
60 TB/s 01 010 1101 11001 110110 0111000 01001110 110111001 1111010001 010110000101 t H χ 0 W Z 300 MB/s
Big Data @ KPMG
Behavioral patterns: Location Aware Services (macroscopic) Service Background KPMG Cases Conference venue Our work for a large Dutch telecom provider allowed us to gain experience with various cases regarding crowd control and location based services. Especially for the coronation of the King in 2013, a website and widget were developed in close collaboration with several parties. Requirements / Challenges Intensity and movement of crowd on street level based on antenna data Sentiment analysis of social media Real-time feedback Resources Data sources: aggregated transaction data of mobile phones, Tweets including GPS coordinates Duration (develop phase): 5 10 weeks Solution Using open standards for Big Data Analytics, we have developed a base solution that is scalable to millions of mobile phone users, tweets and other data points. An intuitive user interface is developed, which is important to represent analysis results to the public. Benefits By combining crowd and sentiment monitoring, valuable real-time information is available for crowd control and safety purposes. This information can be used to prevent hazardous situations and instantly respond to incidents. The strength of this tool is its flexibility to adapt to different types of use; e.g. emergency services or taxi companies.! 2014 KPMG Advisory N.V., registered with the trade register in the Netherlands under number 33263682, is a member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative ( KPMG International ), a Swiss entity. All rights reserved. Printed in the Netherlands. The KPMG name, logo and cutting through complexity are registered trademarks of KPMG International. #
Service Behavioral patterns: Location Aware Services (microscopic) KPMG Cases Background Conference venue Tracking the customer behaviour is important information which can be gathered using several types of sources. Wi-Fi signals can for instance be used to detect shopping patterns from customers, but one can think of many different data sources and techniques, like security cameras, ibeacon or Bluetooth. Requirements / Challenges Real-time collection and analysis Wi-Fi device signals. Analysing the customer behaviour and plotting patterns, graphs and trend lines. Storing sales data, data from RFID chips or motion sensors for extra information about certain products. Storing other relevant (e.g. Social Media) information. Resources Data sources: Wi-Fi device signals. Additionally: sales data, RFID data, social media data, camera footage, ibeacon. Duration (develop phase): 8 16 weeks Solution The analysis of any of these separate data streams will provide vital information about the shopping behaviour of customers. Combining all the data creates even richer information. Benefits The goal is to define which variables affect a purchase and make changes accordingly. For example, a store might deploy more salespeople, alter displays or put out red blouses instead of blue, according to a story on Bloomberg.! 2014 KPMG Advisory N.V., registered with the trade register in the Netherlands under number 33263682, is a member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative ( KPMG International ), a Swiss entity. All rights reserved. Printed in the Netherlands. The KPMG name, logo and cutting through complexity are registered trademarks of KPMG International. #
Service Customer store card use: pattern finding, modelling, and global reporting KPMG Cases Background Conference venue Over 80 million gift and loyalty cards are distributed to customers globally within a large multinational chain of furniture stores. The client s current live database system could not provide the insights required to declare liability correctly for profit and loss calculation, especially since a large group of customers will never actually spend some portion of their open balance. Requirements / Challenges Fifty different legal entities across several countries used the customer cards in different ways. Additionally different groups of customers used cards in different ways. These differences needed to be correctly located, understood and accounted. Bringing the power of a truly statistical-based big-data analysis and modelling into a world of accountancy and audit presents many challenges. Proving that we understand customer behavior is one of them. Solution Our team worked closely together with client experts to understand the data, the business strategy, and the customer trends We leveraged our powerful secure big-data platform KAVE to develop insights into the data without interfering with the live system We discovered several factors with a significant influence on customer spending behavior and so categorized customers to produce a stable and accurate predictive model with known and quantified uncertainties. Benefits Complete and accurate reporting of more than 100 M open balance. Evaluation and accounting of cross-legal-entity spending. Accurate predictions of the amounts that are will not be spent in the future KAVE: Our big data platform Fraction spent Customers spent around 95% of all money loaded Card age Customers spent the majority of their money in the first few weeks Flows of money move back and forth between legal entities as customers use their store cards in different places Hundreds of customer categories! Europe/US Asia 2014 KPMG Advisory N.V., registered with the trade register in the Netherlands under number 33263682, is a member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative ( KPMG International ), a Swiss entity. All rights reserved. Printed in the Netherlands. The KPMG name, logo and cutting through complexity are registered trademarks of KPMG International. #
Big Data @ Healthcare
My drive
Algorithms in Healthcare [1/2]
Algorithms in Healthcare [2/2] http://youtu.be/wncclbzr_i4
How we learn in Healthcare 100% : 0,1% Peter M. Rothwell External validity of randomised controlled trials: "to whom do the results of this trial apply? Lancet. 2005 Jan 1-7;365(9453):82-93.
The Big Data way Use all available information to provide the best evidence at the right time to support decision making! Pacmed example: real-time analysis of GP data from around NL providing information for similar patient profiles shared decision making
Case Operational Excellence: Asset management and planning KPMG Cases Background Monte Carlo simulation to analyze operational efficiency of Operation Room planning at a large academic hospital. A new blue print for OR planning was created and needed to be evaluated on performance, expected production numbers and where issues might arise. Requirements / Challenges Handling poor data quality Translate highly flexible human decision-making to rules in model Client involvement (expert knowledge) was needed throughout the process in order to create an expert system for planning Resources Data sources: blueprint, historical OR usage data Duration (develop phase): 10 16 weeks Solution Built a Monte Carlo simulation model to evaluate the expected performance of the blueprint in various scenario s and to quantify these results. Benefits With our results, the client gained insight into the expected performance of the new blueprint, including the expected performance for several alternative scenario s. Reality Simulation 2012 2014 2014 KPMG Advisory N.V., registered with the trade register in the Netherlands under number 33263682, is a member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative ( KPMG International ), a Swiss entity. All rights reserved. Printed in the Netherlands. The KPMG name, logo and cutting through complexity are registered trademarks of KPMG International. #
Case Difference between reality and simulation KPMG Cases Human planning Machine planning Colors indicate different medical specialisms. Human planning less strict in separation between specialisms. 2014 KPMG Advisory N.V., registered with the trade register in the Netherlands under number 33263682, is a member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative ( KPMG International ), a Swiss entity. All rights reserved. Printed in the Netherlands. The KPMG name, logo and cutting through complexity are registered trademarks of KPMG International. #
Caution is needed http://www.tylervigen.com 24
Closing remarks Signal vs Noise and Correlation vs Causation My PhD thesis http://www.xkcd.com
Dare to do it yourself: Where do you see opportunities in your organisation? football24.ua
Make it cool and specific specific vague this is the best place to start?this is where most people would place Big Data analyses cool scary
Go for it and you might just avoid the awful It could have been so much better KNVB
Thank you! dr. ir. Egge van der Poel info@eggewel.nl +31624817682