Fellowship Bio

Dr Rob Miller graduated from Bristol University with a medical and neuroscience degree. He completed his foundation training in Thames Valley and subsequently moved to London for surgical training. His dedication to this work will persist as a Topol fellow, allowing him to further advance the application of machine learning in hand surgery.

Fellowship Project

Development & deployment of artificial intelligence radiology platforms

King’s College Hospitals NHS Foundation Trust

The project goals were to understand the theory, development, implementation, and evaluation of autonomous imaging AI in clinical radiology. This focused on the use of machine learning techniques through an active learning platform to train an intracranial haemorrhage segmentation model. During the fellowship Rob contributed to the development of an intra-cranial haemorrhage segmentation model using the MONAI label platform. In doing so, a model was created to facilitate increased rate of intra-cranial haemorrhage CT head images, which can be used to enhance training of an intra-cranial haemorrhage phenotyping model. Performance of the segmentation model was evaluated against a novice clinician regarding accuracy and speed of segmentations and usability of the MONAI label system.

Fellowship Testimonial

With a subspecialist interest in hand surgery, I have been actively contributing to developing machine learning models in this field during my fellowship. This fellowship allowed me to compile a successful application to the Topol digital transformation fellowship to develop a hand x-ray segmentation model and continue work on autonomous hand function analysis, which is aligned to my clinical sub- specialist area of interest.

Links

Topol Fellowship profile