For most part of history, doctors have held the monopoly on interpreting medical data. However, with the advance of powerful computers and sophisticated algorithms, in combination with the availability of structured electronic medical records, this is slowly changing. Predictive Medical Technologies is developing a system that uses real-time intensive care monitoring data to predict clinical events like cardiac arrest up to 24 hours ahead of time. Ciara Byrne from O’Reilly Radar had an interview with the company’s CEO Bryan Hughes. In summary, the system integrates with computer systems and extracts raw data coming from the monitors, lab reports and nursing notes. That data is then fed into models based on non-hypothesis machine learning techniques, which eliminate any human expert bias from the models. So far the system predicts cardiac arrest and respiratory failure, and in the future this will be expanded to include sepsis, renal failure and re-intubation risk as well. The system has already proven its value on retrospective data, and a prospective validation trial is about to start. It will be at least 12 months before any formal FDA trial will be started. For the full interview, follow the link below.
Interview at O’Reilly: Algorithms are the new medical tests…
Company homepage: Predictive Medical Technologies…