Technical Name One Model Fit All: Revolutionary One-Stop Shop System for Predicting Coronary Stenosis without Normal Database in Myocardial Perfusion Imaging
Project Operator Far Eastern Memorial Hospital
Project Host 吳彥雯
Summary
1. A novel, rapid one-stop predictive system for obstructive coronary artery disease (CAD), with automatic preprocessing techniques including segmentation, alignment with standard templates, normalization, standardization,3D spherical conversion.2. With deep learning method to accurately predict obstructive CAD, the treatment responseprognosis.3. This technology can be applied in any 3D myocardial images such as SPECT, PET, CT, MRI without normal database, with considerable accuracy.
Scientific Breakthrough
1. A one-stop predictive system of myocardial perfusion imaging for coronary artery disease.2. The automatic preprocessing includes segmentation, alignment with standard templates, normalization,3D map conversion.3. The deep learning algorithm could be used in the diagnosis of CAD, prediction of treatment responseprognosis, with excellent performance comparing with the conventional algorithms.4. This technology can be applied in SPECT, PET, CTMRI, without the normal database.
Industrial Applicability
The applications include rapidstandardized diagnosis process,evaluationprediction of treatment responsecardiovascular outcomes. We could cooperate with the equipmentsoftware analysis manufacturers,Nationalcommercial health insurance/companies whether it is expected to be used to evaluate the patients’ riskthe necessity of invasive procedureshospitalization.
Keyword Artificial intelligence machine learning deep learning neural network algorithm coronary artery heart, vessel, vascular, image myocardial perfusion imaging (MPI) single photon emission computed tomography (SPECT) spherical coordinate system
Notes
  • Contact
  • Yen-Wen Wu
other people also saw