Nuclear Medicine

Subtle PET by Subtle Medical

AI-powered software solution to enhance images from up to 4x faster PET scans

Nuclear Medicine Image Quality Improvement

With more than 300 000 exams performed every year in France, PET/CT has become the go-to solution for diagnosing and monitoring cancers. But the high acquisition time required by the tool makes it tedious both for the physicians and the patients. SubtlePET brings a solution by dividing by four the recording time of a PET-scan exam.

What if an IA solution could drastically decrease the acquisition time required for a PET-scan exam ?

Developed from proprietary deep learning algorithms, SubtlePET allows you to divide by four the required acquisition time for a PET-Scan. Using the result of a much shorter exam, SubtlePET technology enhances and denoises the picture to obtain a clinically-equivalent image, making sure the final diagnosis stays the same. 

Expected benefits

  • Improve the patient experience by reducing the scanning time.
  • Increase your efficiency by scanning more patients a day.
  • Sign up for a virtual upgrade of your scanners whatever their brand, without needing to invest in new equipements.
  • Integrate an AI tool to streamline your workflow without slowing you down.

What Specialist are saying

In addition to the quality of the SubtlePET-enhanced images being indistinguishable from the standard images, there were no artifacts and the lesion detectability was excellent. I can absolutely see this improving both radiology workflow and the patient experience.

Peter Giuliano, MD Department Chair and Director of the Department of Nuclear Medicine at Hoag Hospital (California)


  1. Quantitative Standardized Uptake Value Evaluation of 4x Faster PET Scans Enhanced using Deep Learning – A Chaudhari, PhD; P Gulaka, PhD; T Zhang; S Srinivas, MD, PhD; G Zaharchuk, MD, PhD; E Gong, PhD, Stanford.
  2. 200x Low-dose PET Reconstruction using Deep Learning – Junshen Xu*, Enhao Gong*, John Pauly, Greg Zaharchuk.
  3. Accelerating Whole-Body PET Acquisitions Using Deep Learning: External validation on foreign country data – Jose Leite, Gustavo Tukamoto1, Akshay Chaudhari, Praveen Gulaka, Enhao Gong, GregZaharchuk, Igor Rafael Martins dos Santos, Flávia Paiva Proença Lobo Lopes, Felipe Campos, Kitamura