Dr. Michael Barrowman, PhD

Data Scientist

Evolution Software

Biography

Dr. Michael Barrowman is a Data Scientist, currently employed by Evolution Sofware, a company which manages online casinos and builds their internal software structure. With a background in both the public and private sector and a wealth of experience working with medical, education and financial datasets; he has produced stunning visualisations for previous teams as well as publication-ready reports and production-ready scripts.

He has recently completed his doctoral Thesis on Multi-State Clinical Prediction Models in Renal Replacement Therapy, earning him a PhD from the University of Manchester. His PhD project encompasses the development and validation of a multi-state clinical predication model, as well as the methodological advancements to produce such a model. This has led to multiple publications and the creation of software as a by-product.

He is interested in Data Science, particularly using R and python to their fullest potential, encouraging others to do the same and is an advocate for neat and reproducible coding practices. He enjoys web development and learning about web paradigms as wel as learning new programming languages such as C++ and Typescript.

He lives in Merseyside, UK with his partner, two children and two step-children. He enjoys walks down by the local canal, through nearby forested areas and trips to the park as often as possible as his daughter’s favourite outdoor activity is “going on adventures”.

Interests

Data Science
Financial Risk Modelling
Clinical Prediction Modelling
Multi-State Models

Education

PhD in Medicine, 2022

University of Manchester
MSci in Mathematics, 2013

University of Lancaster

Skills

R

SQL

python

C++

Statistics

Communication

Data Visualisation

LaTeX

CI

Power BI

HTML/CSS/JS

AWS

Posts

The lapply() family

This question on Reddit, got me thinking about the lapply() family of functions, and how a beginner might want to learn about them. Here is my take Introduction The easiest one to understand is lapply(), I’ll work through that and then extend to the others.

Last updated on Dec 6, 2020 11 min read

The New Base Pipe

Here, we’re going to take a quick look at the new pipe introduced in the development version of R 4.1.0, and compare it to the well-known %>% pipe from the {magrittr} package that is used throughout the {tidyverse}.

Last updated on Dec 6, 2020 8 min read

Convert all Character variables to Factors

Introduction First, let’s load up what we need! set.seed(15102020) library(tidyverse) #We'll use tidyverse functions library(magrittr) #A few extra pipes from magrittr library(lexicon) #For a word dictionary When dealing wth complex datasets, it is common that a variable may be stored as a character variable, when in reality what you want is a factor variable.

Last updated on Oct 15, 2020 7 min read

Getting a variable name in a pipeline

There is no greater staple of the {tidyverse} than the pipe, %>%, however not a lot of people understand what’s going on “under-the-bonnet” of the pipe. To be fair, not many people have to worry about it.

Last updated on Jul 24, 2020 7 min read

Deploy to GitHub Pages

Within git, the default branch is usually named master, however in recent times, the negative connotations of that word are coming to the forefront of a lot of people’s minds, and so they are wishing to diverge away from that kind of terminology.

Last updated on Jul 22, 2020 12 min read

Packages

rando

The goal of rando is to provide easier generating of random numbers in a manner that is context aware, and reproducible.

Jan 18, 2021

mutils

The goal of mutils is to provide useful functions to make data processing smoother. Most functions contained here are ‘nifty’, rather than ‘innovative’.

Jul 21, 2020

mpipe

The mpipe package is designed to add extra functionality to the pipeline process in tidyverse style R usage

Jun 22, 2020

typos

The goal of typos is to provide a flexible warning when commonly mis-typed functions are called. Functions with typing errors will still be evaluated and a warning will be output. It also provides the user with a convenient function to define their own typos.

Jun 14, 2020

Experience

Data Analyst & Scientist

EvoSoft

Nov 2022 – Present Manchester, Greater Manchester, UK

Analysing customer and transactional data to build dashboards, interact with stake holder APIs and develop business driven models to improve performance of our softwares and ease the customer journey

PhD Candidate

University of Manchester

Oct 2016 – Apr 2022 Manchester, UK

The goal of this PhD was to improve the academic knowledge surrounding Multi-State Clinical Prediction Models (MSCPMs). To accomplish this, I wrote articles to solve methodological issues that were yet to be addressed and applied these novel techniques (along with the literature) to develop and validate an MSCPM to predict outcomes for Chronic Kidney Disease patients.

Publications

Glen Philip Martin, David A Jenkins, Lucy Bull, Rose Sisk, Lijing Lin, William Hulme, Anthony Wilson, Wenjuan Wang, Dr. Michael Barrowman, PhD, Camilla Sammut-Powell, Alexander Pate, Matthew Sperrin, Niels Peek, Predictive Healthcare Analtics Group (2020). Toward a Framework for the Design, Implementation, and Reporting of Methodology Scoping Reviews. J CLIN EPIDEMIOL.

PDF DOI

Dr. Michael Barrowman, PhD, Niels Peek, Mark Lambie, Glen Philip Martin, Matthew Sperrin (2019). How Unmeasured Confounding in a Competing Risks Setting Can Affect Treatment Effect Estimates in Observational Studies. BMC Med Res Methodol.

PDF DOI

Alexander Pate, Dr. Michael Barrowman, PhD, David Webb, Jeanne M Pimenta, Kourtney J Davies, Rachael Williams, Tjeerd Van Staa, Matthew Sperrin (2018). Study Investigating the Generalisability of a COPD Trial Based in Primary Care (Salford Lung Study) and the Presence of a Hawthorne Effect. BMJ Open Respir Res.

PDF DOI