Ross Williams, MSc
Ross obtained his MSc (2017) in Data Science from King's College London having previously obtained his BSc in Physics and Philosophy from the same institution. His thesis focused on predicting the change in diastolic blood pressure level for patients being treated for hypertension based upon demographics and pharmacological intervention, a particular focus of this was on external validation of the developed models.
His current focus is on methodological research into external validation of prediction models. Including but not limited too assessment metrics, transportability and recalibration. Accordingly he has contributed code to the PatientLevelPrediction R package along these lines. Further lines of interest include the creation of a prediction model library, implementation of Association Rule mining, temporal data analysis and methods for dealing with class imbalance.
Implementation of the COVID-19 Vulnerability Index Across an International Network of Health Care Data Sets: Collaborative External Validation Study
JMIR Medical Informatics
2021-04-05 | journal-article
Sarcopenia in older people with chronic airway diseases: the Rotterdam study
ERJ Open Research
2021-01-15 | journal-article
Renin-angiotensin system blockers and susceptibility to COVID-19: an international, open science, cohort analysis.
The Lancet. Digital health
2020-12 | journal-article
PMID: 33342753PMC: PMC7834915DOI: 10.1016/s2589-7500(20)30289-2
Renin-angiotensin system blockers and susceptibility to COVID-19: a multinational open science cohort study2020-06 | preprint
OTHER-ID: PPR174750DOI: 10.1101/2020.06.11.20125849
Feasibility and evaluation of a large-scale external validation approach for patient-level prediction in an international data network: validation of models predicting stroke in female patients newly diagnosed with atrial fibrillation.
BMC medical research methodology
2020-05 | journal-article
PMID: 32375693PMC: PMC7201646DOI: 10.1186/s12874-020-00991-3
Development and validation of a prognostic model predicting symptomatic hemorrhagic transformation in acute ischemic stroke at scale in the OHDSI network.
2020-01 | journal-article
PMID: 31910437PMC: PMC6946584DOI: 10.1371/journal.pone.0226718
Sarcopenia in COPD: a systematic review and meta-analysis.
European respiratory review : an official journal of the European Respiratory Society
2019-11-13 | journal-article
PMID: 31722892DOI: 10.1183/16000617.0049-2019
Opioid use, postoperative complications, and implant survival after unicompartmental versus total knee replacement: a population-based network study
The Lancet Rheumatology
2019 | journal-article
- Assists Dr. Rijnbeek in the teaching of data science to students of the “Klinische Technology” Master of Science program. This course aims to provide students the fundamentals of machine learning in a medical context. The course includes practical exercises that focus on the development of clinical prediction models. Ideally, this course is further extended to a full curriculum on health data science to teach all medical students at the Erasmus MC the basics of this exciting multidisciplinary field.
- Teaches on the Patient-Level Prediction tutorial day. A day organised around the OHDSI Symposiums and at other times during the year which teaches students the why, what and how of running a clinical prediction model using the OHDSI tools.
Prediction Modelling Autumn School, King's College London, November 2017.
Population Health Management course Advanced Risk Stratification, Leiden UMC, May 2018.
- Young Researcher in the Spotlight, HealthySciencesDay, ErasmusMC, April 2019.
- "The Prediction Model Library, OHDSI Symposium 2021
- "Predicting adverse events following total knee replacement", ISPE 2019, Philadelphia, August 2019
- "Predicting Heart Failure in PAtients newly initialising treatment for type 2 diabetes", OHDSI Symposium 2018, October 2018.
2019 Titan Award for Clinical Application, OHDSI Symposium, Bethesda, September 2019
2021 Titan Award for Community Support, OHDSI Symposium, September 2021