Improving detection accuracy of perfusion defect in standard dose SPECT-myocardial perfusion imaging by deep-learning denoising

• Published in: Journal of nuclear cardiology Vol. 29; no. 5; pp. 2340 - 2349
• Main Authors: Liu, Junchi, Yang, Yongyi, Wernick, Miles N., Pretorius, P. Hendrik, Slomka, Piotr J., King, Michael A.
• Format: Journal Article
• Language: English
• Published: Cham Elsevier Inc 01-10-2022; Springer International Publishing; Springer Nature B.V
• Subjects: Cardiology; Deep Learning; Humans; Image Processing, Computer-Assisted - methods; Imaging; Medicine; Medicine & Public Health; Myocardial Perfusion Imaging - methods; noise-to-noise training; Nuclear Medicine; Original Article; Perfusion; post-reconstruction filtering; Radiology; ROC Curve; SPECT-MPI; Tomography, Emission-Computed, Single-Photon - methods; AC; MPI; DL; OSEM; TPD; LV; SPECT; AUC; noise-to-noise training; SC; post-reconstruction filtering; deep learning; RC; SPECT-MPI
• ISSN: 1071-3581; 1532-6551
• DOI: 10.1007/s12350-021-02676-w

We previously developed a deep-learning (DL) network for image denoising in SPECT-myocardial perfusion imaging (MPI). Here we investigate whether this DL network can be utilized for improving detection of perfusion defects in standard-dose clinical acquisitions.
To quantify perfusion-defect detection accuracy, we conducted a receiver-operating characteristic (ROC) analysis on reconstructed images with and without processing by the DL network using a set of clinical SPECT-MPI data from 190 subjects. For perfusion-defect detection hybrid studies were used as ground truth, which were created from clinically normal studies with simulated realistic lesions inserted. We considered ordered-subset expectation-maximization (OSEM) reconstruction with corrections for attenuation, resolution, and scatter and with 3D Gaussian post-filtering. Total perfusion deficit (TPD) scores, computed by Quantitative Perfusion SPECT (QPS) software, were used to evaluate the reconstructed images.
Compared to reconstruction with optimal Gaussian post-filtering (sigma = 1.2 voxels), further DL denoising increased the area under the ROC curve (AUC) from 0.80 to 0.88 (P-value < 10−4). For reconstruction with less Gaussian post-filtering (sigma = 0.8 voxels), thus better spatial resolution, DL denoising increased the AUC value from 0.78 to 0.86 (P-value < 10−4) and achieved better spatial resolution in reconstruction.
DL denoising can effectively improve the detection of abnormal defects in standard-dose SPECT-MPI images over conventional reconstruction.