It’s been great attending the Intelligence 2018 seminar, with many great talks and inspiring speakers. Here’s a quick summary of the important points I learned in the day.
Issues with medical data
- Volume: medical imaging data are abundant compared to other specialties, but it’s still small compared to general data such as road data for automated vehicles, everyday images such as ImageNet.
- Privacy: already a huge hurdle for many research projects, but another issue lies in the fact that current anonymisation techniques may not be sufficient
- Source and bias: although we have many promising results from ML with the current data sets, there are always issues with “rubbish in, rubbish out”.
- Labels and ground truth
- Rare cases: we simply don’t have enough of rare medical cases to train the machine
A few issues with results
- clinical significance
- false positives
Future of radiology
- guide treatment
- participate in machine learning
Future of medical innovation
- workflow improvement rather than systems innovation
- infrastructure improvements will offer further higher yield: fax machines, electronic health records