Bart Steverink & Laurens Schinkelshoek

Ai

Discover how a medical AI startup predicts surgical infections and provides risk scores to doctors using electronic health records, machine learning, and medical decision support systems, with real-world results and future plans for growth and adoption.

Key takeaways
  • The company’s product aims to predict infections in surgical patients and provide a risk score for doctors to make informed decisions.
  • The data used is extracted from electronic health records (EHR) systems, which are diverse and lack standardization.
  • The company uses a medical decision support system to analyze data and provide predictions.
  • The product has been tested in several hospitals in the Netherlands and is currently being expanded to other countries.
  • The company is working on getting CE certification, which will allow them to supply hospitals in all of Europe.
  • The product uses supervised machine learning and has been trained on a dataset of over 10,000 patients.
  • The company is also working on integrating their product with EHR systems and developing a Python-only version of the product.
  • The company has faced resistance to adopting their product, but sees potential for growth and adoption in the future.
  • The company is working on addressing the issue of data standardization and is developing tooling to speed up the process.
  • The company is also working on addressing the issue of free unstructured text data and is developing a structure mapping to a uniform set of columns.
  • The company is planning to do a study to compare the use of their product to not using the product and to determine if it provides an advantage for hospitals.
  • The company is also planning to do a study to determine if their product can be used across different hospitals and populations.