Skip to main content

IBEX QDR: A Ground-breaking Technological Upgrade based on Quantitative Digital Radiography to Exploit the Full Diagnostic Potential of Standard Digital X-Ray

Periodic Reporting for period 2 - IBEX QDR (IBEX QDR: A Ground-breaking Technological Upgrade based on Quantitative Digital Radiography to Exploit the Full Diagnostic Potential of Standard Digital X-Ray)

Reporting period: 2018-08-01 to 2020-01-31

Bone Fractures are a very common health problem. Between 2,000 and 3,000 fracture patients are examined using X-rays at this hospital every year. Fractures usually happen when too much force is applied to the bone, usually during a fall or an accident. Fractures can result in a partially or completely broken bone and sometimes broken in several places. Accurate diagnosis of fractures is very important in deciding their treatment.
The likelihood of suffering a fracture can be increased by bone conditions such as osteoporosis, a condition where your bones become fragile and brittle and break easily. Around 3 million people in the UK suffer from osteoporosis. Osteoporotic hip fractures occur in 1 in 3 women and 1 in 5 men over the age of 50 and are the most common cause of injury-related deaths in the UK. Fragility fractures are estimated to cost the NHS £2.3 billion in health and social care provision in the UK alone.
Once diagnosed, there are effective ways to treat Osteoporosis and prevent fractures. However, the X-ray scans performed at fracture clinics are currently unable to diagnose a patient’s bone health, and only a small proportion of patients diagnosed with fractures are referred for measurement of bone health at specialist DEXA clinics. DEXA clinics are heavily utilised and it routinely takes over 3 months from an ‘at risk’ patient (usually patients over 50 presenting with a low-impact fracture) first presenting at a fracture clinic to an accurate diagnosis of their bone-health.
Digital radiography systems are far more prevalent in hospitals than DEXA systems and are utilised for a variety of diagnostic procedures that require the internal body structures to be imaged, however they don't currently offer any form of quantitative information and suffer from the effects of scattered X-rays which degrade image quality.
The IBEX technology has the potential to make these systems more efficient and safer by removing the need for an anti-scatter grid to be used, resulting in the potential to collect images at a lower X-ray dose, and also enabling additional diagnostic information to be collected at the same time as a fracture is assessed.
The main aim of this project was to complete development of the IBEX technology and demonstrate its efficacy in a clinical study to support commercial exploitation and adoption by healthcare providers.
The initial project was to develop the technology based on the patented IBEX multi-absorption plate (MAP) technology but through continuous engagement with potential customers and some interesting development activity it was decided to refocus development on a software only solution that could achieve the objectives of the project in a more straightforward solution.
As such, the main areas of focus for this first period were on the software development and on preparation of documentation for the clinical study.
Software development activities brought significant improvements to the outputs of the technology and also opened up additional development opportunities that could be exploited outside of this project. Key results have shown that the scatter correction algorithms are performing very well in comparison to anti-scatter grids and bone mineral density results on phantoms showed good accuracy in comparison to DEXA.

The second period focused on finalising the Trueview software development, including extensive testing and validation in preparation for the clinical study submission. The clinical study was initiated and recruitment to Phase 1 completed. Unfortunately, due to issues with recruitment and the impact of the Covid-19 pandemic, we were not able to complete the study in the project time-frame. We fully intend to complete it and report out once restrictions in hospitals have been lifted.
Alongside this there has been a significant amount of effort in implementing and improving quality procedures required to meet ISO 13485 accreditation and to collate information required for the Technical File. The result of this work was achieving ISO13485 accreditation, putting us in a good position to pursue commercialisation of the Trueview product.
The other main area of effort has been in commercial engagement with potential customers. We have attended five major radiology conferences as technical exhibitors and the technology has been very well received by everyone we have spoken to. This has also led to a number of successful customer evaluations and demonstrations and we have been able to secure license deals in the Cone Beam CT and Mammography applications, with multiple ongoing discussions that should lead to further license deals in the digital radiography application within the year.
The results so far have shown that even in an incomplete state, the Trueview technology compares favourably with, and in many cases outperforms, traditional anti-scatter grids and other software-based scatter correction methods. This is due to the real physics-based nature of the Trueview algorithms as opposed to image processing-based methods employed by other companies. In addition to this, the quantitative information obtained using the Trueview technology cannot currently be collected by any other digital radiography-based system, indicating true innovation.
The second period of the project focused on finalising the Trueview software and conducting the clinical study. Unfortunately, due to recruitment issues and the lock-down imposed as a a result of the Covid-19 pandemic, we have not been able to fully complete recruitment for the study and are unable to report out the final results. Despite this, we have been able to conduct a preliminary, non regulated, analysis of the data available to us with the following results:

Phase 1 - Comparison to DEXA
Based on the limited data set of 12 patients that we had taken from the site prior to the lock-down, an assessment of the classifier accuracy was conducted (assuming DEXA classifications to be correct).
It was shown that the Trueview classifier had an accuracy of 83%, which is extremely promising given the level of methodological error present in the data. We believe that this represents the lower bound of the Trueview capability and that, when fully integrated in a customer system , this accuracy will increase significantly enabling Trueview to provide a measure of bone health for any patient receiving an X-ray scan and to be used as a triage tool to funnel the most at-risk patients into DEXA and to avoid sending asymptomatic patients unnecessarily, thus reducing the burden on oversubscribed DEXA clinics.
A full analysis of the final results will provide a clearer picture and we expect them to support the hypothesis that Trueview can provide an accurate measure of bone health.

Phase 2 - Image Quality Assessment
Cone Beam CT outputs
Mammography outputs
Digital radiography outputs 1
Digital radiography outputs 2