AI-supported early fracture diagnosis (Phase 2: SeeAI LTD.) V01
Research type
Research Study
Full title
Artificial intelligence-supported early fracture diagnosis (Phase 2: SeeAI LTD.)
IRAS ID
286809
Contact name
Saile Patricia Moreno Villegas
Contact email
Duration of Study in the UK
1 years, 0 months, 0 days
Research summary
Each year in Scotland, the NHS gives some 5,000 patients x-rays of wrists, hands, ankles and feet, most often looking for a fracture after trauma. Although isolated injuries in these areas are often categorised as ‘minor’, misdiagnosis and consequent mismanagement can result in significant impact for patients and financial costs to the NHS.
Artificial Intelligence (AI) or “machine learning” (a set of procedure rules to take in clinical data such as X-rays, assess the risk of a fracture and present this risk and information to a clinical team) could be developed to help clinicians make diagnoses.
To develop AI or machine learning tools and to take these tools to the level of “approved for health care use” and integrated into the appropriate IT and/or equipment for healthcare use requires a partnership between NHS, academia and industry.
NHS Grampian A&E and Radiology clinicians have identified that there is significant clinical need and are eager to work in partnership with those with the technical skills to develop potential solutions.
A pilot study (Phase 1) involving a small (100 patients) fully anonymised x-ray dataset (no patient names, addresses, date of birth or hospital numbers) has already been completed (IRAS no. 271600) using an AI laboratory space within the Grampian Data Safe Haven (DaSH). After successfully demonstrating potential to develop useful clinical solutions to support fracture detection, we have been invited to proceed to Phase 2.
In a second phase, we will access a larger dataset (images for 10,000 patients) to develop our solutions further.
The technical outcome of phase 2 is a working prototype tested on unseen data and ready for formal multi-site clinical trials. We have developed a prototype for detecting fractures and generating a radiology report through phase 1. The system will be trained with a large scale dataset, and further technical improvements will be added to allow the AI system to withstand practical settings. Specifically, the project will seek to achieve three innovations.- 1: Accurately detecting subtle fractures and non-acute fractures
The outcome of phase 1 proved that our AI algorithm detects and classifies various fractures. However, the data provided contained a limited number in the type of fractures present. For instance, we did not encounter enough images with very subtle or non-acute fractures. The former is difficult to detect and classify and the latter needs to be reported if present in the x-ray images. We aim to develop an algorithm that can detect and identify these types of fractures.- 2: A technique for identifying abnormalities
Detecting fractures is not the sole remit of radiologists. They must report other findings, such as soft tissue swelling, joint effusion, and tumours if they are present. We will train an algorithm to crop the area to be checked (such as joints, or soft tissue) and analyse if abnormalities can be detected.- 3: Human-in-the-loop system for faster annotation and continuous improvement
Human-machine interaction is advantageous in achieving the best performance in various settings. For this project, such a strategy will be significantly beneficial during the annotation process of a large scale dataset. Our fracture detection model requires annotated bounding boxes over the fractures. Such a task is labour-intensive, and it is not realistic to employ radiologists to annotate 20,000 images manually. We will use Human-in-the-loop (HITL) strategy to update the AI algorithms and label the data iteratively so the annotation cost is reduced and the annotation process is sped up without sacrificing quality.REC name
London - Stanmore Research Ethics Committee
REC reference
20/PR/0319
Date of REC Opinion
27 Aug 2020
REC opinion
Favourable Opinion