Rotational Profiles
Measurement of lower limb rotation.
Modality:
CT
Pathology:
Lower limb rotation
Status:
Concluded
CSC Lead: Anil
This project has been withdrawn. This project may be revisited at a later date if capacity permits.
Current practice of manually drawing lines over the long bones and the joints on CT images in PACS to figure the degree
of misalignment is time-consuming (15-20 minutes per case). An AI tool would automatically create these lines and
angles, saving time and reducing variability.
Clinical lead: Dr. Christopher Tang
Rationale
Assessment of rotational profiles is time-consuming and complex. An AI algorithm would significantly decrease the amount of time needed to measure the angles of rotation.
Patient pathway
Patients are referred by the orthopaedic team. CT imaging is part of the pre-surgical work up. The report is used to plan the surgical intervention.
Training data
Over 100 patients are referred annually. The data will be retrospective, acquiring a large cohort of patients and segmentation and profile results as annotated on PACS. It is possible annotations will need to be re-done if they cannot be exported from PACS.
Risks
The tool is to help find the accurate rotational profile and will be supervised and signed off by the radiologist, so an error is unlikely to persist far enough to affect a patient. goals: Reduction of radiologist time in reporting these scans as well as increase consistency in calculation.
Success criteria
Accurate, automated rotational profiles.
Alternatives
Currently no commercial products identified
References
Website used to help calculate rotational angles is here.