I am a *Designer, Researcher & Art Director.
*Focusing on UX/UI, for products and software systems, interaction design, visual story-telling and problem solving.
InnerEye – Medical Imaging AI to Empower Clinicians
UX/UI designer, In collaboration with colleagues at Microsoft Research Cambridge and Radiation Oncology Consultants at Addenbrooke's hospital, Cambridge UK.
InnerEye develops machine learning techniques for the automatic delineation of tumours as well as healthy anatomy in 3D radiological images. In a collaboration with oncology experts, machine learning specialists and interaction designers, I designed the user experience and UI for the prototype software.
Image contouring is typically done manually, where a clinician will manually draw around a tumour on each MRI slice. It can be a time-consuming and expensive process. Because of that, it is also often only done once, at the beginning of the treatment.
The InnerEye technology may enable: 1. extraction of targeted Radiomics measurements for quantitative radiology, 2. efficient contouring for radiotherapy planning, 3. precise surgery planning and navigation. In practice, Project InnerEye turns multi-dimensional radiological images into measuring devices.
Project InnerEye builds upon many years of research in computer vision and machine learning. It employs algorithms such as Deep Decision Forests (as used already in Kinect and Hololens) as well as Convolutional Neural Networks (as available in CNTK) for the automatic, voxel-wise segmentation of medical images.
Demonstration UI walkthroughs of prototype system.
Demonstrating the basic UI
Editing contours with brushes
Contouring co-registered image pairs
Morphological operations on contours
The technology is being designed with the guidance of expert medical practitioners. The results of our machine learning operations can be readily refined and adjusted by expert clinical researchers until they are completely satisfied with the results. They maintain full control of the results at all times.
