James McCouat
Biography
I work in the field of Medical Image Analysis specifically the use of machine learning, such as deep learning, to automatically extract clinically useful information from patient MRIs or X-rays. I am jointly supervised by Irina Voiculescu in Computer Science and Sion Glyn-Jones in Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences (NDORMS) a department I am also affiliated with. I work mainly on scans or X-rays of the hip and pelvis. I am particularly interested in the following areas:
- How prior knowledge of the shape and structure of the human anatomy can be encoded into machine learning algorithms.
- Techniques to improve the performance of machine learning algorithms on limited data.
- Unsupervised and semi-unsupervised machine learning techniques that are viable in situations where not all the dataset is labeled.
- How machine learning can be practically implemented within the NHS to give information that is genuinely useful to doctors.
Selected Publications
-
Vertebrae Detection and Localization in CT with Two−Stage CNNs and Dense Annotations
James McCouat and Ben Glocker
2020.
Details about Vertebrae Detection and Localization in CT with Two−Stage CNNs and Dense Annotations | BibTeX data for Vertebrae Detection and Localization in CT with Two−Stage CNNs and Dense Annotations