Cardiac DiseaseDEEPLAIMNCT07888218
Replacing Spatial Filtering With Deep-learning Noise Reduction
An observational study of Cardiac Disease, run by Central Hospital, Nancy, France at 1 site in France.
Trial summary
Spatial filtering is routinely required to reduce the statistical noise of gated SPECT myocardial perfusion imaging, but at the expense of image blurring. This study will evaluate a deep-learning-based noise reduction (DLNR) approach, developed to shorten acquisition times, when used as a replacement for conventional… Read more about this study
- Who is it for?
- Ages 18 to 90 years
- Conditions the study is for
- Cardiac Disease
- How many people
- 49 participants
- Where
- 1 siteFrance
- Study dates
- 25 Nov 2026to 3 Mar 2027
- Phase
- Observational
Key requirements
Who can take part, as listed by the sponsor. Only the study team can confirm whether you are eligible, so take this list to your doctor.
- Ages 18 to 90 years
- All sexes
- No healthy volunteers
- Adult, Older adult
Study population: Patients referred for myocardial scintigraphy due to known or suspected coronary artery disease, who underwent the same one-day "low-dose" protocol.
Who may be able to join
- Patients who underwent myocardial scintigraphy at the Nancy University Hospital (CHRU) under the conditions applicable to the groups described above.
Who cannot join
- Significant arrhythmia, technical reconstruction issues, poor image quality (motion artifacts during acquisition, insufficient count statistics).
- Patients who objected to the use of their data for research purposes.
Trial locations and contacts
1 site in 1 country. Contact the study team or a site near you to ask about taking part.
About this study
Official title: Replacing Spatial Filtering With Deep-learning Noise Reduction Improves Low-dose Gated SPECT Myocardial Perfusion Imaging
Spatial filtering is routinely required to reduce the statistical noise of gated SPECT myocardial perfusion imaging, but at the expense of image blurring. This study will evaluate a deep-learning-based noise reduction (DLNR) approach, developed to shorten acquisition times, when used as a replacement for conventional spatial filters (CSF).
Read the detailed description
SPECT image noise can also be reduced using deep-learning algorithms, which were initially developed to enable shorter acquisition times or lower injected activities while maintaining diagnostic image quality. More recently, when used alone rather than in combination with spatial filters, such an algorithm was shown to outperform the spatial filters previously required for bone SPECT, providing greater noise reduction with substantially less smoothing. This raises the question of whether the same approach could be applied to gated-SPECT images acquired with low-dose protocols.
- Conditions
- Cardiac Disease
- Medical subject headings
- Heart Diseases
What is being measured
The primary outcome is the main question the study is designed to answer.
Compare deep learning-based noise reduction and conventional spatial filters for myocardial scintigraphy image reconstruction.
Experienced physicians will perform a visual analysis of the images reconstructed with and without deep learning and will complete a questionnaire that includes an image quality and interpretation score.
Measured: Day 1
Secondary outcomes (1)
Compare left ventricle end-diastolic and end-systolic volumes and Left Ventricle Ejection Fraction between deep learning and conventional filters
Image quality: visual scale of cine loops for the two observers, categorized as good-to-excellent (low background noise, good edge definition), intermediate, or poor (very high background noise, edge definition making analysis difficult or impossible).
Measured: Day 1
Timeline
From the start of the study to its completion; the mark is today. Estimated dates are the sponsor’s plan.
- First posted22 Oct 2026
- Study start25 Nov 2026 (estimated)
- Primary completion23 Jan 2027 (estimated)
- Study completion3 Mar 2027 (estimated)
Resources
- Official record on ClinicalTrials.govThe registry entry this page is built from, with its full change history.
- Other trials for Cardiac DiseaseStudies of the same condition, open and under way.
- What the trial phases meanHow a treatment moves from first tests in people to approval.
More details
For doctors and researchers: how the study is designed, who runs it, and its registry IDs.
Study design
- Study type
- Observational
- Observational model
- Other
- Time perspective
- Retrospective
- Enrolment
- 49 participants (estimated)
Sponsor and investigators
- Lead sponsor
- Central Hospital, Nancy, France
- Sponsor type
- University, hospital or other
- Responsible party
- Principal investigator, Pierre Yves MARIE, Central Hospital, Nancy, France
IDs, dates and publications
- ClinicalTrials.gov ID
- NCT07888218
- Sponsor's protocol ID
- 2026PI207
- Acronym
- DEEPLAIM
- Status last verified
- Oct 2026
- First submitted
- 14 Oct 2026
- Last updated
- 22 Oct 2026
- Results posted
- No
- Data monitoring committee
- No
- FDA-regulated drug
- No
- FDA-regulated device
- No