[MA 2022 04] Internship Dataverification – Clinical Outcomes (MRDM)

Amsterdam

Proposed by: Mark Wevers [mark.wevers@logex.com]

At LOGEX we turn data into better healthcare

LOGEX delivers analytics solutions that enable data-driven decision making in every area of healthcare – from operations and finance, to clinical and patient outcomes - in order to achieve the best possible value. The right people make this all possible. That’s why we’re constantly on the lookout for brilliant minds from a wide range of expertise. Our team of over 300 international experts consists of data scientists, analysts, consultants and experts in technology, healthcare economics, medicine, mathematics and statistics. We work together to support our customers and build long-term, mutually fulfilling partnerships with them. If you enjoy solving complex problems for the greater good and working in a dynamic, entrepreneurial culture, then LOGEX is the company for you.

Background internship data verification

The Netherlands has clinical registries for various diseases, which improve the quality of care by providing insights through research on the collected data. Clinicians enter data, for a significant number of variables per patient, manually in the EHR which is then (re)used to fill the clinical registry. Consequently, errors occur in the data due to various reasons, e.g. typing mistakes, incorrect interpretation of variable definitions, the existence of information silo’s, etc. However, in order to obtain valuable insights, data quality must be high. After all, the quality of insights obtained from clinical registry data cannot be better than the quality of the corresponding clinical registry data itself. To address this issue, MRDM has developed a methodology called automatic data verification which seeks out potential discrepancies in clinical registries using variable-specific models based on either business rules or machine learning.

What you will be doing as an intern data verification?

At MRDM we are constantly pushing to further improve the data verification methodology and we are looking for a master student to strengthen our team for a period of approximately 6 months! We have created a master thesis assignment which focusses on increasing the prediction performance of the data verification methodology. The topic of the assignment is a mix of Data Science and Machine Learning and will require you to program R or Python. As mentioned before, the data verification methodology seeks out potential discrepancies using variable-specific models based on either business rules or machine learning. For both methods we have the following concrete expectations:

Business rules

1. Automate the generation of business rules. Currently we create variable-specific business rules ourselves. We would like you to investigate automating this process via for example Bayesian rule lists.

Machine Learning

1. Feature engineering and creation of predictive meta-features to improve prediction accuracy. You experiment with predictive features for both chirurgical and non-chirurgical registries and create or derive these predictive features from other features and/or raw data.

2. Implementing an efficient feature selection method. We have a significant number of features at our disposal. This makes it hard for machine learning models to understand which features are important, due to the so-called “curse of dimensionality”. To address this issue, we would like you to look into efficient feature selection methods.

Internship Dataverification – Clinical Outcomes (MRDM)

What you bring to the table?

• You are studying a MSc degree with a quantitative / beta / computer science / technical background.

• You have experience- and affinity with programming (preferably with R or Python).

• You have experience with analysing large datasets and are interested in learning to work with data pipelines.

• You are flexible and have a learning mindset.

• You have good communication skills.

• You speak and write both Dutch and English fluently.

• Knowledge/experience on the topic of Data Science, Machine Learning and/or AI is a plus, especially if this is related to healthcare.

What we offer

• A competitive internship allowance.

• A challenging assignment in a fast growing, international company.

• An informal work environment with ambitious colleagues.

• Laptop and home office supplies to create a comfortable and healthy work environment.

• Monthly digital team events. As soon as the RIVM regulations allow, weekly Friday drinks and many more events will be hosted.

Interested?

We are happy to meet you and would like to tell you more about this exciting opportunity. You can apply via the button below and upload your motivation letter and CV. For more information, or in case you have any questions, you can contact Mark Wevers via mark.wevers@logex.com or +31648531467 (Phone or WhatsApp).