Diagnostic performance of a novel deep learning attenuation correction software for MPI using a cardio dedicated CZT camera. Experience in the clinical practice.

PURPOSE: To evaluate the diagnostic performance of a novel deep learning attenuation correction software (DLACS) for myocardial perfusion imaging (MPI) using a cadmium-zinc-telluride (CZT) cardio dedicated camera with invasive coronary angiography (ICA) correlation for the diagnosis of coronary artery disease (CAD) in a high-risk population. METHODS: Retrospective study of 300 patients (196 males [65%], mean age 68 years) from September 2014 to October 2019 undergoing MPI, followed by ICA and evaluated by means of quantitative angiography software, within six months after the MPI. The mean pre-test probability score for coronary disease according to the European Society of Cardiology criteria was 37% for the whole cohort. The MPI was performed in a dedicated CZT cardio camera (D-SPECT Spectrum Dynamics) with a two-day protocol, according to the European Association of Nuclear Medicine guidelines. MPI was retrospectively evaluated with and without the DLACS. RESULTS: The overall diagnostic accuracy of MPI without DLACS to identify patients with any obstructive CAD at ICA was 87%, sensitivity 94%, specificity 57%, Positive Predictive Value 91% and Negative Predictive Value 64%. Using DLACS the overall diagnostic accuracy was 90%, sensitivity 91%, specificity 86%, Positive Predictive Value 97% and Negative Predictive Value 66%. CONCLUSION: Use of the novel DLACS enhances performance of the MPI using the CZT D-SPECT camera and achieves improved results, especially avoiding artefacts and reducing the number of false positive results.

Metabolic network alterations as a supportive biomarker in dementia with Lewy bodies with preserved dopamine transmission.

PURPOSE: Metabolic network analysis of FDG-PET utilizes an index of inter-regional correlation of resting state glucose metabolism and has been proven to provide complementary information regarding the disease process in parkinsonian syndromes. The goals of this study were (i) to evaluate pattern similarities of glucose metabolism and network connectivity in dementia with Lewy bodies (DLB) subjects with subthreshold dopaminergic loss compared to advanced disease stages and to (ii) investigate metabolic network alterations of FDG-PET for discrimination of patients with early DLB from other neurodegenerative disorders (Alzheimer's disease, Parkinson's disease, multiple system atrophy) at individual patient level via principal component analysis (PCA). METHODS: FDG-PETs of subjects with probable or possible DLB (n = 22) without significant dopamine deficiency (z-score < 2 in putamen binding loss on DaT-SPECT compared to healthy controls (HC)) were scaled by global-mean, prior to volume-of-interest-based analyses of relative glucose metabolism. Single region metabolic changes and network connectivity changes were compared against HC (n = 23) and against DLB subjects with significant dopamine deficiency (n = 86). PCA was applied to test discrimination of patients with DLB from disease controls (n = 101) at individual patient level. RESULTS: Similar patterns of hypo- (parietal- and occipital cortex) and hypermetabolism (basal ganglia, limbic system, motor cortices) were observed in DLB patients with and without significant dopamine deficiency when compared to HC. Metabolic connectivity alterations correlated between DLB patients with and without significant dopamine deficiency (R2 = 0.597, p < 0.01). A PCA trained by DLB patients with dopamine deficiency and HC discriminated DLB patients without significant dopaminergic loss from other neurodegenerative parkinsonian disorders at individual patient level (area-under-the-curve (AUC): 0.912). CONCLUSION: Disease-specific patterns of altered glucose metabolism and altered metabolic networks are present in DLB subjects without significant dopaminergic loss. Metabolic network alterations in FDG-PET can act as a supporting biomarker in the subgroup of DLB patients without significant dopaminergic loss at symptoms onset.

Deep learning prediction of quantitative coronary angiography values using myocardial perfusion images with a CZT camera.

PURPOSE: Evaluate the prediction of quantitative coronary angiography (QCA) values from MPI, by means of deep learning. METHODS: 546 patients (67% men) undergoing stress 99mTc-tetrofosmin MPI in a CZT camera in the upright and supine position were included (1092 MPIs). Patients were divided into two groups: ICA group included 271 patients who performed an ICA within 6 months of MPI and a control group with 275 patients with low pre-test probability for CAD and a normal MPI. QCA analyses were performed using radiologic software and verified by an expert reader. Left ventricular myocardium was segmented using clinical nuclear cardiology software and verified by an expert reader. A deep learning model was trained using a double cross-validation scheme such that all data could be used as test data as well. RESULTS: Area under the receiver-operating characteristic curve for the prediction of QCA, with > 50% narrowing of the artery, by deep learning for the external test cohort: per patient 85% [95% confidence interval (CI) 84%-87%] and per vessel; LAD 74% (CI 72%-76%), RCA 85% (CI 83%-86%), LCx 81% (CI 78%-84%), and average 80% (CI 77%-83%). CONCLUSION: Deep learning can predict the presence of different QCA percentages of coronary artery stenosis from MPIs.

Diagnostic performance of different cardiac stress protocols for myocardial perfusion imaging for the diagnosis of coronary artery disease using a cadmium-zinc-telluride camera with invasive coronary angiography correlation.

PURPOSE: To evaluate the diagnostic performance of three different cardiac stress protocols for myocardial perfusion imaging (MPI) using a cadmium-zinc-telluride (CZT) camera with invasive coronary angiography (ICA) correlation for the diagnosis of coronary artery disease in a high risk population. METHODS: Retrospective study of 263 patients (96 women and 167 males, mean age 68 years) from which 119 patients performed a bicycle stress test (BST), 113 pharmacological stress test (PST) and 31 a combination of the two (CST) between September 2014 and December 2018. The patients then underwent myocardial perfusion imaging (MPI), followed by ICA and evaluated by means of quantitative angiography software, within six months after the MPI. The mean pre-test probability score for coronary disease according to the European Society of Cardiology criteria was 36% for the whole population. The MPI was performed in a dedicated CZT cardio camera (D-SPECT Spectrum Dynamics) with a two-day protocol, according to the European Association of Nuclear Medicine guidelines. RESULTS: No significant difference was observed between the three stress protocols in terms of diagnostic accuracy (BST 85%, PST 88%, CST 84%). The overall diagnostic accuracy of MPI to identify patients with any obstructive CAD at ICA was 86%, Sensitivity 93%, Specificity 54%, PPV 90% and NPV 63%. CONCLUSION: The CZT D-SPECT camera achieves overall satisfactory results in the diagnosis of CAD, observing no significant differences in the diagnostic performance when the stress test was performed as a BST, PST or CST.

Adopting transfer learning for neuroimaging: a comparative analysis with a custom 3D convolution neural network model.

BACKGROUND: In recent years, neuroimaging with deep learning (DL) algorithms have made remarkable advances in the diagnosis of neurodegenerative disorders. However, applying DL in different medical domains is usually challenged by lack of labeled data. To address this challenge, transfer learning (TL) has been applied to use state-of-the-art convolution neural networks pre-trained on natural images. Yet, there are differences in characteristics between medical and natural images, also image classification and targeted medical diagnosis tasks. The purpose of this study is to investigate the performance of specialized and TL in the classification of neurodegenerative disorders using 3D volumes of 18F-FDG-PET brain scans. RESULTS: Results show that TL models are suboptimal for classification of neurodegenerative disorders, especially when the objective is to separate more than two disorders. Additionally, specialized CNN model provides better interpretations of predicted diagnosis. CONCLUSIONS: TL can indeed lead to superior performance on binary classification in timely and data efficient manner, yet for detecting more than a single disorder, TL models do not perform well. Additionally, custom 3D model performs comparably to TL models for binary classification, and interestingly perform better for diagnosis of multiple disorders. The results confirm the superiority of the custom 3D-CNN in providing better explainable model compared to TL adopted ones.

MRI of nigrosome-1: A potential triage tool for patients with suspected parkinsonism.

BACKGROUND AND PURPOSE: Susceptibility-weighted imaging (SWI) of nigrosome-1 is an emerging and clinically applicable imaging marker for parkinsonism, which can be derived from routinely performed brain MRI. The purpose of the study was to assess whether SWI can be used as a triage tool for more efficient selection of subsequent Dopamine Transporter Scan (DaTSCAN) single-photon emission computed tomography (SPECT). METHODS: We examined 72 consecutive patients with suspected parkinsonism with both DaTSCAN SPECT and SWI (48 in Philips Ingenia, 24 in GE Signa). Additionally, we examined 24 healthy controls with SWI (14 in Philips Ingenia, 10 in GE Signa). Diagnostic performance of SWI and DaTSCAN SPECT was assessed on the basis of clinical diagnosis, in terms of sensitivity, specificity, and diagnostic accuracy. RESULTS: A total of 54 parkinsonism patients (69 years ± 9, 32 men), 18 nonparkinsonism patients (69.4 years ± 9, 10 men), and 24 healthy controls (62 years ± 8, 10 men) were recruited. SWI had a specificity of 92% and a sensitivity of 74%, whereas DaTSCAN SPECT had 83% and 94%, respectively. By preselecting patients with abnormal or inconclusive SWI, the diagnostic performance of DaTSCAN SPECT improved (specificity 100%, sensitivity 95%). Scans from Philips were associated with significantly lower image quality compared to GE (p < .001). The experienced rater outperformed the less experienced one in diagnostic accuracy (82% vs. 68%). CONCLUSIONS: SWI can be used as triage tool because normal SWI can in most cases rule out parkinsonism. However, the performance of SWI depends on acquisition parameters and rater's experience.

A 3D deep learning model to predict the diagnosis of dementia with Lewy bodies, Alzheimer's disease, and mild cognitive impairment using brain 18F-FDG PET.

PURPOSE: The purpose of this study is to develop and validate a 3D deep learning model that predicts the final clinical diagnosis of Alzheimer's disease (AD), dementia with Lewy bodies (DLB), mild cognitive impairment due to Alzheimer's disease (MCI-AD), and cognitively normal (CN) using fluorine 18 fluorodeoxyglucose PET (18F-FDG PET) and compare model's performance to that of multiple expert nuclear medicine physicians' readers. MATERIALS AND METHODS: Retrospective 18F-FDG PET scans for AD, MCI-AD, and CN were collected from Alzheimer's disease neuroimaging initiative (556 patients from 2005 to 2020), and CN and DLB cases were from European DLB Consortium (201 patients from 2005 to 2018). The introduced 3D convolutional neural network was trained using 90% of the data and externally tested using 10% as well as comparison to human readers on the same independent test set. The model's performance was analyzed with sensitivity, specificity, precision, F1 score, receiver operating characteristic (ROC). The regional metabolic changes driving classification were visualized using uniform manifold approximation and projection (UMAP) and network attention. RESULTS: The proposed model achieved area under the ROC curve of 96.2% (95% confidence interval: 90.6-100) on predicting the final diagnosis of DLB in the independent test set, 96.4% (92.7-100) in AD, 71.4% (51.6-91.2) in MCI-AD, and 94.7% (90-99.5) in CN, which in ROC space outperformed human readers performance. The network attention depicted the posterior cingulate cortex is important for each neurodegenerative disease, and the UMAP visualization of the extracted features by the proposed model demonstrates the reality of development of the given disorders. CONCLUSION: Using only 18F-FDG PET of the brain, a 3D deep learning model could predict the final diagnosis of the most common neurodegenerative disorders which achieved a competitive performance compared to the human readers as well as their consensus.

Associations among education, age, and the dementia with Lewy bodies (DLB) metabolic pattern: A European-DLB consortium project.

INTRODUCTION: We assessed the influence of education as a proxy of cognitive reserve and age on the dementia with Lewy bodies (DLB) metabolic pattern. METHODS: Brain 18F-fluorodeoxyglucose positron emission tomography and clinical/demographic information were available in 169 probable DLB patients included in the European DLB-consortium database. Principal component analysis identified brain regions relevant to local data variance. A linear regression model was applied to generate age- and education-sensitive maps corrected for Mini-Mental State Examination score, sex (and either education or age). RESULTS: Age negatively covaried with metabolism in bilateral middle and superior frontal cortex, anterior and posterior cingulate, reducing the expression of the DLB-typical cingulate island sign (CIS). Education negatively covaried with metabolism in the left inferior parietal cortex and precuneus (making the CIS more prominent). DISCUSSION: These findings point out the importance of tailoring interpretation of DLB biomarkers considering the concomitant effect of individual, non-disease-related variables such as age and cognitive reserve.

The effect of acute exercise on interleukin-6 and hypothalamic-pituitary-adrenal axis responses in patients with coronary artery disease.

Vulnerability to stress-induced inflammation has been linked to a dysfunctional hypothalamus-pituitary-adrenal (HPA) axis. In the present study, patients with known or suspected coronary artery disease (CAD) were assessed with respect to inflammatory and HPA axis response to acute physical exercise. An exercise stress test was combined with SPECT myocardial perfusion imaging. Plasma and saliva samples were collected before and 30 min after exercise. Interleukin (IL)-6 and adrenocorticotropic hormone (ACTH) were measured in plasma, while cortisol was measured in both plasma and saliva. In total, 124 patients were included of whom 29% had a prior history of CAD and/or a myocardial perfusion deficit. The levels of exercise intensity and duration were comparable in CAD and non-CAD patients. However, in CAD patients, IL-6 increased after exercise (p = 0.019) while no differences were seen in HPA axis variables. Conversely, patients without CAD exhibited increased levels of ACTH (p = 0.003) and cortisol (p = 0.004 in plasma, p = 0.006 in saliva), but no change in IL-6. We conclude that the IL-6 response to acute physical exercise is exaggerated in CAD patients and may be out of balance due to HPA axis hypoactivity. It remains to be further investigated whether this imbalance is a potential diagnostic and therapeutic target in CAD.

Metabolic Correlates of Dopaminergic Loss in Dementia with Lewy Bodies.

BACKGROUND: Striatal dopamine deficiency and metabolic changes are well-known phenomena in dementia with Lewy bodies and can be quantified in vivo by 123 I-Ioflupane brain single-photon emission computed tomography of dopamine transporter and 18 F-fluorodesoxyglucose PET. However, the linkage between both biomarkers is ill-understood. OBJECTIVE: We used the hitherto largest study cohort of combined imaging from the European consortium to elucidate the role of both biomarkers in the pathophysiological course of dementia with Lewy bodies. METHODS: We compared striatal dopamine deficiency and glucose metabolism of 84 dementia with Lewy body patients and comparable healthy controls. After normalization of data, we tested their correlation by region-of-interest-based and voxel-based methods, controlled for study center, age, sex, education, and current cognitive impairment. Metabolic connectivity was analyzed by inter-region coefficients stratified by dopamine deficiency and compared to healthy controls. RESULTS: There was an inverse relationship between striatal dopamine availability and relative glucose hypermetabolism, pronounced in the basal ganglia and in limbic regions. With increasing dopamine deficiency, metabolic connectivity showed strong deteriorations in distinct brain regions implicated in disease symptoms, with greatest disruptions in the basal ganglia and limbic system, coincident with the pattern of relative hypermetabolism. CONCLUSIONS: Relative glucose hypermetabolism and disturbed metabolic connectivity of limbic and basal ganglia circuits are metabolic correlates of dopamine deficiency in dementia with Lewy bodies. Identification of specific metabolic network alterations in patients with early dopamine deficiency may serve as an additional supporting biomarker for timely diagnosis of dementia with Lewy bodies. © 2019 The Authors. Movement Disorders published by Wiley Periodicals, Inc. on behalf of International Parkinson and Movement Disorder Society.

Influence of reconstruction algorithms on image quality in SPECT myocardial perfusion imaging.

INTRODUCTION: We investigated if image- and diagnostic quality in SPECT MPI could be maintained despite a reduced acquisition time adding Depth Dependent Resolution Recovery (DDRR) for image reconstruction. Images were compared with filtered back projection (FBP) and iterative reconstruction using Ordered Subsets Expectation Maximization with (IRAC) and without (IRNC) attenuation correction (AC). MATERIALS AND METHODS: Stress- and rest imaging for 15 min was performed on 21 subjects with a dual head gamma camera (Infinia Hawkeye; GE Healthcare), ECG-gating with 8 frames/cardiac cycle and a low-dose CT-scan. A 9 min acquisition was generated using five instead of eight gated frames and was reconstructed with DDRR, with (IRACRR) and without AC (IRNCRR) as well as with FBP. Three experienced nuclear medicine specialists visually assessed anonymized images according to eight criteria on a four point scale, three related to image quality and five to diagnostic confidence. Statistical analysis was performed using Visual Grading Regression (VGR). RESULTS: Observer confidence in statements on image quality was highest for the images that were reconstructed using DDRR (P<0·01 compared to FBP). Iterative reconstruction without DDRR was not superior to FBP. Interobserver variability was significant for statements on image quality (P<0·05) but lower in the diagnostic statements on ischemia and scar. The confidence in assessing ischemia and scar was not different between the reconstruction techniques (P = n.s.). CONCLUSION: SPECT MPI collected in 9 min, reconstructed with DDRR and AC, produced better image quality than the standard procedure. The observers expressed the highest diagnostic confidence in the DDRR reconstruction.

The diagnostic value of dopamine transporter imaging and olfactory testing in patients with parkinsonian syndromes.

The aim of the study was to compare the efficacy of olfactory testing and presynaptic dopamine imaging in diagnosing Parkinson's disease (PD) and atypical parkinsonian syndromes (APS); to evaluate if the combination of these two diagnostic tools can improve their diagnostic value. A prospective investigation of 24 PD patients, 16 APS patients and 15 patients with non-parkinsonian syndromes was performed during an 18-month period. Single photon emission computed tomography with the presynaptic radioligand (123)I-FP-CIT (DaTSCAN(®)) and olfactory testing with the Brief 12-item Smell Identification Test (B-SIT) were performed in all patients. DaTSCAN was analysed semi-quantitatively, by calculating two different striatal uptake ratios, and visually according to a predefined ranking scale. B-SIT score was significantly lower for PD patients, but not significantly different between APS and non-parkinsonism. The visual assessment of DaTSCAN had higher sensitivity, specificity and diagnostic accuracy compared to olfactory testing. Most PD patients (75%) had visually predominant dopamine depletion in putamen, while most APS patients (56%) had visually severe dopamine depletion both in putamen and in caudate nucleus. The combination of DaTSCAN and B-SIT led to a higher rate of correctly classified patients. Olfactory testing can distinguish PD from non-parkinsonism, but not PD from APS or APS from non-parkinsonism. DaTSCAN is more efficient than olfactory testing and can be valuable in differentiating PD from APS. However, combining olfactory testing and DaTSCAN imaging has a higher predictive value than these two methods separately.

Comparison between visual assessment of dopaminergic degeneration pattern and semi-quantitative ratio calculations in patients with Parkinson's disease and Atypical Parkinsonian syndromes using DaTSCAN® SPECT.

OBJECTIVE: To verify if (123)I-FP-CIT, DaTSCAN(®) can differentiate early stages of Parkinson's disease (PD) as well as patients with Atypical Parkinsonian syndromes (APS) from manifest Parkinson's disease. METHODS: 128 consecutive patients were investigated with (123)I-FP-CIT SPECT during a 4-year period. All patients were diagnosed according to the established consensus criteria for diagnosis of PD (n = 53) and APS (n = 19). Remaining patients were grouped early PD (before onset of L-DOPA medication), (n = 20), vascular PD (n = 6), and non-PD syndromes (n = 30) and SWEDD (n = 1). SPECT images were analyzed visually according to a predefined ranking scale of dopaminergic nerve cell degeneration, distinguishing a posterior-anterior degeneration pattern (egg shape) from a more global and severe degeneration pattern (burst striatum). Striatum uptake ratios were quantitatively analyzed with the 3D software, EXINI. RESULTS: In the group of APS patients, the burst striatum pattern was most frequent and found in 61 % (11/18 patients). In PD patients, the egg shape pattern was dominating, especially in early PD where it was present in 95 % (19/20 patients). The positive predictive value for the egg shape pattern to diagnose PD was 92 % in this material (APS and all PD patients) and the specificity 90 % for the burst striatum pattern to exclude APS. The uptake ratios were reduced in both PD and APS patients and closely related to the image ranking. CONCLUSION: In this study, we found that in more than half of the patients it was possible to differentiate between PD and APS by visual interpretation only. Similar results were obtained using semi-quantitative uptake ratios. Combining visual assessment with uptake ratios did not add to the discriminating power of DaTSCAN(®) SPECT in this material.

Chlorine in breath condensate--a measure of airway affection in pollinosis?

BACKGROUND: Infiltration of inflammatory cells in bronchial mucosa and glandular hypersecretion are hallmarks of asthma. It has been postulated that exhaled breath condensate (EBC) mirrors events in epithelial lining fluid of airways, such as presence of local inflammation as well as glandular hypersecretion. It is also well known that eosinophil cationic protein (ECP) and cysteinyl-leukotrienes (cys-LT) are released by circulating inflammatory cells when triggered by antigen stimulation in asthma patients. OBJECTIVES: The aim of this study was to evaluate whether chlorine and/or cys-LT in EBC would reflect changes of exposure of airborne pollen in patients with asthma. METHODS: EBC and serum were collected from 23 patients with allergic asthma during a pollen season and repeated 5 months later during a period with no aeroallergens. Chlorine was measured by means of a sensitive coulometric technique and cys-LT by an EIA technique. Serum ECP was measured and lung function tests were performed and symptoms noted during both occasions. RESULTS: Significantly higher concentrations of chlorine in EBC (p = 0.007) and ECP in serum (p = 0.003) were found during the pollen season compared to post-season. Chlorine levels tended to be higher in patients who reported of chest symptoms compared to those who denied symptoms during the pollen season (p = 0.06). Areas under the receiver-operated characteristic curves (AUC(ROC)) were compared and similar discriminative power to identify exacerbations of asthma was recorded by chlorine in EBC (range 0.67-0.78) and ECP in serum (range 0.64-0.78). CONCLUSION: It is concluded that chlorine in EBC and ECP in serum decreased significantly post-season, and this is suggested to mirror the decrement in airborne antigen. It is furthermore proposed that chlorine in EBC and ECP in serum tend to have a similar capacity to identify seasonal variations in airborne pollen in patients with asthma.

Single-isomer R-salbutamol is not superior to racemate regarding protection for bronchial hyperresponsiveness.

Bronchial hyper-reactivity (BHR) has been suggested to follow cessation of regular medication with racemic salbutamol. This study aimed at investigating the effects from medication with R,S- and R-salbutamol on bronchial response to provocation with isocapnic hyperventilation of cold air (IHCA). Twenty-six patients with mild to moderate asthma were enrolled in a double-blind, randomised, cross-over study. Bronchial response to provocation was measured before and after 1 week's medication. Doses of 0.63 mg R-salbutamol or 1.25 mg R/S-salbutamol were inhaled three times daily during medication-weeks and a wash-out week intervened. Tests were performed 6 h after the last dose of test drug. Impulse oscillometry and forced expiratory volume during one second were methods used to identify bronchial response to provocation. Two patients withdrew from the investigation due to side-effects, one from R- the other from R,S-salbutamol. Comparable resting bronchial conditions were indicated by differences in baseline lung function values of <2% between study days. No statistically significant medication-dependent differences in BHR could be demonstrated between treatment groups. However, 15 patients exhibited higher (P = 0.03) post-treatment BHR after pure R-salbutamol than after R,S-salbutamol. Furthermore, plasma concentrations of R-salbutamol tended to be lower (P = 0.08) after medication with R- than after R,S-salbutamol despite equal doses of R-salbutamol given during the two separate treatment periods. We also found that considerable amounts of S-salbutamol were retrieved in plasma after medication with pure R-salbutamol. We conclude that we were unable to demonstrate favourable effects of R-salbutamol over R,S-salbutamol regarding response to provocation with IHCA after regular medication of 1 week's duration.