In a previous post, a setup using an F# script to perform machine learning with Microsoft.ML is described. A very import aspect for a successful model is picking the right features that will predict the label, i.e. the exposure that is associated with outcome. In this post a simple F# feature selection algorithm is described that automatically figures out which features result in the ‘best’ model.
Machine learning is the new kid on the block and has become accessible with the advent of the Microsoft.ML library. In medicine machine learning hasn’t been used that much. Most scientific epidemiological research still relies on established statistical analysis. The core principle, however, is the same. Known exposure is used to predict outcome. In Pediatric Intensive Care, prediction of mortality is used to benchmark and monitor performance. In the Netherlands, for this purpose, data is gathered at a national basis. This blog will discuss a machine learning setup to analyze data using a data set with 13.793 PICU admissions.
The ability to Extract, Transform and Load data to a format that enables data analysis and machine learning is essential to make use of the vast amount of observational data that is nowadays available. F# can be a very efficient tool to achieve these goals.
In a previous post a described an event driven domain design to run a resuscitation protocol. I also mentioned that running this protocol in practice can have some pitfalls. One of these pitfalls is the use of a defibrillator that is required to apply a cardio conversion by delivering a shock. We have to test that this cannot ever occur.
For some time now I have been thinking of writing an app to help with running a resuscitation protocol. The resuscitation protocol as provided by the Dutch Advanced Pediatric Life Support is rather straight forward. Yet to adhere to this protocol in a stressful situation for a patient with a life threatening condition is a challenge to many physicians or health care providers.
The first thing I will trying when using a new framework or library is to start writing code in a script file and see what happens in the interactive. Normally, I am just to stupid to understand the guidelines, but by experimenting and finding it out myself by writing code I manage to get my head around the new framework or library I am using.
Testing a user interface is a challenging task in programming. However, new tools have emerged to make this better feasible. Using the setup of an web app using the SAFE template, canopy is a testing library that enables UI testing in the browser.
The first hit you get when looking for the answer how ‘medical decision support’ is defined points to clinical decision support systems. The definition for wich is formulated as (attributed to Robert Hayward of the Centre for Health Evidence ):
Clinical decision support systems link health observations with health knowledge to influence health choices by clinicians for improved health care
The next logical step is to look at the topic of artificial intelligence in medicine.
However, in clinical practice there are more immediate and practical needs to have support for simpel calculation and lookup actions.
The things I will be doing this week
- Setting up this site:
- Themes and appearance
- Links to existing sites and blogs
- Setting up and adding all my code projects