News

The misdiagnosis of cardiovascular disease can lead to missed or unnecessary treatments and patient distress. New metrology through EMPIR projects aims to reduce this risk

Positron emission tomography (PET) can be considered to be the gold standard for non-invasively measuring cardiac blood flow, or ‘perfusion’, a measurement that can be used for diagnosing cardiovascular disease (CVD). This involves injecting a radionuclide into a patient and monitoring the perfusion of the affected organ.

Then, using the perfusion images from the PET scanner and the resulting perfusion maps generated after analysing such images, clinicians predict whether the patient has the disease or not. However, PET is not fully quantifiable or traceable in a metrological sense with potential variation in results between scanner types or between PET centres and hospitals. The expertise of the clinicians interpreting image maps can also add an additional source of error and, in clinical practice, these factors can lead to false diagnoses, patient distress and unnecessary treatments.

In the EMPIR project ‘Metrology for multi-modality of impaired tissue perfusion’ (15HLT05, PerfusImaging) a metrological framework was established for measuring cardiac perfusion by different medical imaging modalities. This also includes PET-scans as the current gold standard in clinical imaging of cardiac perfusion.  Much of the PET-related work involved the UK’s National Metrology Institute, the National Physical Laboratory (NPL), and expertise and clinical data from the project’s consortium partner the Turku PET Centre in Finland. In particular, together both partners have developed the first ever calibration standard for PET scanners as well as a risk-based decision-making framework for clinical diagnosis. 

A new EMPIR project Software for evaluating PET cardiac perfusion imaging uncertainties for more accurate diagnosis (19SIP04, TracPETperf) will formalise the results and establish the recommended standardisation measures from the previous project. To do this new software is being designed that can automatically assign causes of measurement uncertainties in image data and display them as an interpretable and actionable visual map. The software will also display automatic classification of suspected regions, which would greatly enhance the reliability of the diagnosis in cases which are borderline between healthy and diseased, thus ultimately assisting clinical decision-making.

This work, led by NPL, will also be performed with the Turku PET Centre who will be supplying their Carimas image processing software, currently distributed for research purposes by Turku PET centre. When completed in 2022 results from the TracPETperf project will allow more effective comparisons between individual doctors, systems and institutes.

Nadia Smith who was involved in the PerfusImaging project and is coordinating the TracPETperf project said about the work

“This project will provide standardised approaches and software solutions to assess and minimise the effects of measurement uncertainties across various imaging devices or sites in myocardial PET perfusion measurements. This software will be the first of its kind, incorporating uncertainties into the diagnostic pipeline. This will result in more efficient and cost-effective healthcare services, and ultimately lead to more personalised and timely treatments for patients”.

These EMPIR projects are co-funded by the European Union's Horizon 2020 research and innovation programme and the EMPIR Participating States.

Want to hear more about EURAMET?
Sign up for EURAMET newsletters and other information
Follow us on LinkedIn and Twitter