The role of the deep convolutional neural network as an aid to interpreting brain [18F]DOPA PET/CT in the diagnosis of Parkinson's disease.

OBJECTIVES: To test the performance of a 3D convolutional neural network (CNN) in analysing brain [18F]DOPA PET/CT in order to identify patients with nigro-striatal neurodegeneration. We evaluated the robustness of the 3D CNN by testing it against a manual regional analysis of the striata by using a striatal-to-occipital ratio (SOR). METHODS: We analyzed patients who had undergone [18F]DOPA PET/CT from 2016 to 2018. Two examiners interpreted PET/CT images as positive or negative. Only patients with at least 2 years of follow-up and an ascertained neurological diagnosis were included. A 3D CNN was developed to evaluate [18F]DOPA PET/CT and refine the diagnosis of movement disorder. This system required training and testing, which were carried out on 2/3 and 1/3 of patients, respectively. A regional analysis was also conducted by drawing region of interest on T1-weighted 3D MRI scans, on which the [18F]DOPA PET images were first co-registered. RESULTS: Ninety-eight patients were enrolled: 43 presented nigro-striatal degeneration and 55 negative cases used as controls. After training on 69 patients, the diagnostic performance of the 3D CNN was then calculated in 29 patients. Sensitivity, specificity, negative predictive value, positive predictive value and accuracy were 100%, 89%, 100%, 85% and 93%, respectively. When we compared the 3D CNN results with the SOR analysis, we found that the two patients falsely classified as positive by the 3D CNN procedure showed SOR values ≤ 5th percentile of the negative cases' distribution. CONCLUSIONS: 3D CNNs are able to interpret [18F]DOPA PET/CT properly, revealing patients affected by Parkinson's disease. KEY POINTS: • [18F]DOPA PET/CT is a sensitive diagnostic tool to identify patients with nigro-striatal neurodegeneration. • A semiquantitative evaluation of the images allows a more confident interpretation of the PET findings. • 3D convolutional neural network allows an accurate interpretation of 18F-DOPA PET/CT images, revealing patients affected by Parkinson's disease.

Impact of attenuation correction and gated acquisition in SPECT myocardial perfusion imaging: results of the multicentre SPAG (SPECT Attenuation Correction vs Gated) study.

PURPOSE: In clinical myocardial single photon emission computed tomography (SPECT), attenuation artefacts may cause a loss of specificity in the identification of diseased vessels that can be corrected by means of gated SPECT (GSPECT) acquisition or CT attenuation correction (AC). The purpose of this multicentre study was to assess the impact of GSPECT and AC on the diagnostic performance of myocardial scintigraphy, according to patient's sex, body mass index (BMI) and site of coronary artery disease (CAD). METHODS: We studied a group of 104 patients who underwent coronary angiography within 1 month before or after the SPECT study. Patients with a BMI>27 were considered "overweight". Attenuation-corrected and standard GSPECT early images were randomly interpreted by three readers blinded to the clinical data. RESULTS: In the whole group, GSPECT and AC showed a diagnostic accuracy of 86.5% (sensitivity 82%, specificity 93%) and 77% (sensitivity 75.4%, specificity 81.4%), respectively (p<0.05). In women, when anterior ischaemia was matched with CAD, AC failed to show any increase in specificity (AC 63.6% vs GSPECT 63.6%) with evident loss of sensitivity (AC 72.7% vs GSPECT 90.9%). AC significantly improved SPECT specificity in the identification of right CAD in overweight men (AC 100% vs GSPECT 66.7%, p<0.05). CONCLUSION: AC improved specificity in the evaluation of right CAD in overweight men. In the other evaluable subgroups specificity was not significantly affected while sensitivity was frequently reduced.

Feasibility and diagnostic accuracy of a gated SPECT early-imaging protocol: a multicenter study of the Myoview Imaging Optimization Group.

UNLABELLED: The aim of this study was to investigate whether early (time 1, or T1) myocardial tetrofosmin imaging is feasible and as accurate in detecting coronary artery disease as is standard delayed (time 2, or T2) imaging. METHODS: One hundred twenty patients (100 men and 20 women; mean age +/- SD, 61 +/- 10 y) with anginal symptoms underwent tetrofosmin gated SPECT. Stress/rest T1 imaging was performed at 15 min and T2 at 45 min after injection. Image quality was visually evaluated using a 4-point scale (from 0 = poor to 3 = optimal). Myocardial perfusion analysis was performed on a 20-segment model using quantitative perfusion SPECT software, and reversible ischemia was scored as a summed difference score (SDS). Coronary angiography was performed within 1 mo on all patients, and stenosis of more than 50% of the diameter was considered significant. RESULTS: Overall, quality was scored as optimal or good for 94% of T1 images and 95% of T2 images (P = not statistically significant). Heart, lung, liver, and subdiaphragmatic counts did not differ for stress and rest T1 and T2 imaging. A good linear relationship was seen between T1 and T2 SDS (r = 0.69; P < 0.0001), and Bland-Altman analysis showed good agreement between the 2 conditions. In terms of global diagnostic accuracy, areas under the receiver-operating-characteristic curve were comparable between T1 and T2 (0.80 vs. 0.81, P = not statistically significant). Discrepancies between T1 and T2 SDS were observed in 44% of patients (T1 - T2 SDS > 2). Linear regression analysis showed a good correlation between T1 and T2 SDS (r = 0.67; P < 0.0001), whereas the Bland-Altman method showed a shift in the mean value of the difference of +2.67 +/- 2.73. In patients with a T1 - T2 SDS of more than 2, areas under the receiver-operating-characteristic curves were significantly higher for T1 than for T2 images (0.79 vs. 0.70, P < 0.001). CONCLUSION: T1 imaging is feasible and as accurate as T2 imaging in identifying coronary artery disease. However, in a discrete subset of patients, early acquisition strengthens the clinical message of defect reversibility by permitting earlier, more accurate identification of more severe myocardial ischemia.