A new method of modeling the multi-stage decision-making process of CRT using machine learning with uncertainty quantification.

Aims: Current machine learning-based (ML) models usually attempt to utilize all available patient data to predict patient outcomes while ignoring the associated cost and time for data acquisition. The purpose of this study is to create a multi-stage machine learning model to predict cardiac resynchronization therapy (CRT) response for heart failure (HF) patients. This model exploits uncertainty quantification to recommend additional collection of single-photon emission computed tomography myocardial perfusion imaging (SPECT MPI) variables if baseline clinical variables and features from electrocardiogram (ECG) are not sufficient. Methods: 218 patients who underwent rest-gated SPECT MPI were enrolled in this study. CRT response was defined as an increase in left ventricular ejection fraction (LVEF) > 5% at a 6±1 month follow-up. A multi-stage ML model was created by combining two ensemble models: Ensemble 1 was trained with clinical variables and ECG; Ensemble 2 included Ensemble 1 plus SPECT MPI features. Uncertainty quantification from Ensemble 1 allowed for multi-stage decision-making to determine if the acquisition of SPECT data for a patient is necessary. The performance of the multi-stage model was compared with that of Ensemble models 1 and 2. Results: The response rate for CRT was 55.5% (n = 121) with overall male gender 61.0% (n = 133), an average age of 62.0±11.8, and LVEF of 27.7±11.0. The multi-stage model performed similarly to Ensemble 2 (which utilized the additional SPECT data) with AUC of 0.75 vs. 0.77, accuracy of 0.71 vs. 0.69, sensitivity of 0.70 vs. 0.72, and specificity 0.72 vs. 0.65, respectively. However, the multi-stage model only required SPECT MPI data for 52.7% of the patients across all folds. Conclusions: By using rule-based logic stemming from uncertainty quantification, the multi-stage model was able to reduce the need for additional SPECT MPI data acquisition without sacrificing performance.

Classification Prediction of Alzheimer's Disease and Vascular Dementia Using Physiological Data and ECD SPECT Images.

Alzheimer's disease (AD) and vascular dementia (VaD) are the two most common forms of dementia. However, their neuropsychological and pathological features often overlap, making it difficult to distinguish between AD and VaD. In addition to clinical consultation and laboratory examinations, clinical dementia diagnosis in Taiwan will also include Tc-99m-ECD SPECT imaging examination. Through machine learning and deep learning technology, we explored the feasibility of using the above clinical practice data to distinguish AD and VaD. We used the physiological data (33 features) and Tc-99m-ECD SPECT images of 112 AD patients and 85 VaD patients in the Taiwanese Nuclear Medicine Brain Image Database to train the classification model. The results, after filtering by the number of SVM RFE 5-fold features, show that the average accuracy of physiological data in distinguishing AD/VaD is 81.22% and the AUC is 0.836; the average accuracy of training images using the Inception V3 model is 85% and the AUC is 0.95. Finally, Grad-CAM heatmap was used to visualize the areas of concern of the model and compared with the SPM analysis method to further understand the differences. This research method can quickly use machine learning and deep learning models to automatically extract image features based on a small amount of general clinical data to objectively distinguish AD and VaD.

Comparison of Taiwanese and European Calibration Factors for Heart-to-Mediastinum Ratio in Multicenter 123I-mIBG Phantom Studies.

Background: Cross-calibration of 123I-labeled meta-iodobenzylguanidine (mIBG) myocardial-derived indices is essential to extrapolate findings from several clinical centers. Here, we conducted a phantom study to generate conversion coefficients for the calibration of heart-to-mediastinum ratios and compare them between Taiwan and Europe. Methods: We used an acrylic phantom dedicated to 123I-mIBG planar imaging to calculate the conversion coefficients of 136 phantom images derived from 36 Taiwanese institutions. A European phantom image database including 191 images from 27 institutions was used. Conversion coefficients were categorized into five collimator types: low-energy (LE) high-resolution (LEHR), LE general-purpose (LEGP), extended LEGP (ELEGP), medium-energy (ME) GP (MEGP), and ME low-penetration (MELP) collimators. Results: The conversion coefficients were 0.53 ± 0.039, 0.59 ± 0.032, 0.79 ± 0.032, 0.96 ± 0.038, and 0.99 ± 0.050 for LEHR, LEGP, ELEGP, MEGP, and MELP collimators, respectively. The Taiwanese and European conversion coefficients for the LEHR, LEGP, and MELP collimators did not significantly differ. The coefficient of variation was slightly higher for the Taiwanese than the European conversion coefficients (3.7%-7.5% vs. 2.3%-5.6%). Conclusions: We calculated conversion coefficients for various types of collimators used in Taiwan using a 123I-mIBG phantom. In general, the Taiwanese and European conversion coefficients were comparable. These findings further corroborated and highlighted the need for 123I-mIBG standardization using the phantom-determined conversion coefficients.

EAGMN: Coronary artery semantic labeling using edge attention graph matching network.

Coronary artery disease (CAD) is one of the primary causes leading deaths worldwide. The presence of atherosclerotic lesions in coronary arteries is the underlying pathophysiological basis of CAD, and accurate extraction of individual arterial branches using invasive coronary angiography (ICA) is crucial for stenosis detection and CAD diagnosis. However, deep-learning-based models face challenges in generating semantic segmentation for coronary arteries due to the morphological similarity among different types of arteries. To address this challenge, we propose an innovative approach called the Edge Attention Graph Matching Network (EAGMN) for coronary artery semantic labeling. Inspired by the learning process of interventional cardiologists in interpreting ICA images, our model compares arterial branches between two individual graphs generated from different ICAs. We begin with extracting individual graphs based on the vascular tree obtained from the ICA. Each node in the individual graph represents an arterial segment, and the EAGMN aims to learn the similarity between nodes from the two individual graphs. By converting the coronary artery semantic segmentation task into a graph node similarity comparison task, identifying the node-to-node correspondence would assign semantic labels for each arterial branch. More specifically, the EAGMN utilizes the association graph constructed from the two individual graphs as input. A graph attention module is employed for feature embedding and aggregation, while a decoder generates the linear assignment for node-to-node semantic mapping. Based on the learned node-to-node relationships, unlabeled coronary arterial segments are classified using the labeled coronary arterial segments, thereby achieving semantic labeling. A dataset with 263 labeled ICAs is used to train and validate the EAGMN. Experimental results indicate the EAGMN achieved a weighted accuracy of 0.8653, a weighted precision of 0.8656, a weighted recall of 0.8653 and a weighted F1-score of 0.8643. Furthermore, we employ ZORRO to provide interpretability and explainability of the graph matching for artery semantic labeling. These findings highlight the potential of the EAGMN for accurate and efficient coronary artery semantic labeling using ICAs. By leveraging the inherent characteristics of ICAs and incorporating graph matching techniques, our proposed model provides a promising solution for improving CAD diagnosis and treatment.

Generative adversarial network-based attenuation correction for 99mTc-TRODAT-1 brain SPECT.

Background: Attenuation correction (AC) is an important correction method to improve the quantification accuracy of dopamine transporter (DAT) single photon emission computed tomography (SPECT). Chang's method was developed for AC (Chang-AC) when CT-based AC was not available, assuming uniform attenuation coefficients inside the body contour. This study aims to evaluate Chang-AC and different deep learning (DL)-based AC approaches on 99mTc-TRODAT-1 brain SPECT using clinical patient data on two different scanners. Methods: Two hundred and sixty patients who underwent 99mTc-TRODAT-1 SPECT/CT scans from two different scanners (scanner A and scanner B) were retrospectively recruited. The ordered-subset expectation-maximization (OS-EM) method reconstructed 120 projections with dual-energy scatter correction, with or without CT-AC. We implemented a 3D conditional generative adversarial network (cGAN) for the indirect deep learning-based attenuation correction (DL-ACµ) and direct deep learning-based attenuation correction (DL-AC) methods, estimating attenuation maps (µ-maps) and attenuation-corrected SPECT images from non-attenuation-corrected (NAC) SPECT, respectively. We further applied cross-scanner training (cross-scanner indirect deep learning-based attenuation correction [cull-ACµ] and cross-scanner direct deep learning-based attenuation correction [call-AC]) and merged the datasets from two scanners for ensemble training (ensemble indirect deep learning-based attenuation correction [eDL-ACµ] and ensemble direct deep learning-based attenuation correction [eDL-AC]). The estimated µ-maps from (c/e)DL-ACµ were then used in reconstruction for AC purposes. Chang's method was also implemented for comparison. Normalized mean square error (NMSE), structural similarity index (SSIM), specific uptake ratio (SUR), and asymmetry index (%ASI) of the striatum were calculated for different AC methods. Results: The NMSE for Chang's method, DL-ACµ, DL-AC, cDL-ACµ, cDL-AC, eDL-ACµ, and eDL-AC is 0.0406 ± 0.0445, 0.0059 ± 0.0035, 0.0099 ± 0.0066, 0.0253 ± 0.0102, 0.0369 ± 0.0124, 0.0098 ± 0.0035, and 0.0162 ± 0.0118 for scanner A and 0.0579 ± 0.0146, 0.0055 ± 0.0034, 0.0063 ± 0.0028, 0.0235 ± 0.0085, 0.0349 ± 0.0086, 0.0115 ± 0.0062, and 0.0117 ± 0.0038 for scanner B, respectively. The SUR and %ASI results for DL-ACµ are closer to CT-AC, Followed by DL-AC, eDL-ACµ, cDL-ACµ, cDL-AC, eDL-AC, Chang's method, and NAC. Conclusion: All DL-based AC methods are superior to Chang-AC. DL-ACµ is superior to DL-AC. Scanner-specific training is superior to cross-scanner and ensemble training. DL-based AC methods are feasible and robust for 99mTc-TRODAT-1 brain SPECT.

AGMN: Association Graph-based Graph Matching Network for Coronary Artery Semantic Labeling on Invasive Coronary Angiograms.

Semantic labeling of coronary arterial segments in invasive coronary angiography (ICA) is important for automated assessment and report generation of coronary artery stenosis in computer-aided coronary artery disease (CAD) diagnosis. However, separating and identifying individual coronary arterial segments is challenging because morphological similarities of different branches on the coronary arterial tree and human-to-human variabilities exist. Inspired by the training procedure of interventional cardiologists for interpreting the structure of coronary arteries, we propose an association graph-based graph matching network (AGMN) for coronary arterial semantic labeling. We first extract the vascular tree from invasive coronary angiography (ICA) and convert it into multiple individual graphs. Then, an association graph is constructed from two individual graphs where each vertex represents the relationship between two arterial segments. Thus, we convert the arterial segment labeling task into a vertex classification task; ultimately, the semantic artery labeling becomes equivalent to identifying the artery-to-artery correspondence on graphs. More specifically, the AGMN extracts the vertex features by the embedding module using the association graph, aggregates the features from adjacent vertices and edges by graph convolution network, and decodes the features to generate the semantic mappings between arteries. By learning the mapping of arterial branches between two individual graphs, the unlabeled arterial segments are classified by the labeled segments to achieve semantic labeling. A dataset containing 263 ICAs was employed to train and validate the proposed model, and a five-fold cross-validation scheme was performed. Our AGMN model achieved an average accuracy of 0.8264, an average precision of 0.8276, an average recall of 0.8264, and an average F1-score of 0.8262, which significantly outperformed existing coronary artery semantic labeling methods. In conclusion, we have developed and validated a new algorithm with high accuracy, interpretability, and robustness for coronary artery semantic labeling on ICAs.

Real-World Testing of a Machine Learning-Derived Visual Scale for Tc99m TRODAT-1 for Diagnosing Lewy Body Disease: Comparison with a Traditional Approach Using Semiquantification.

Objectives: Abnormal dopamine transporter (DAT) uptake is an important biomarker for diagnosing Lewy body disease (LBD), including Parkinson's disease (PD) and dementia with Lewy bodies (DLB). We evaluated a machine learning-derived visual scale (ML-VS) for Tc99m TRODAT-1 from one center and compared it with the striatal/background ratio (SBR) using semiquantification for diagnosing LBD in two other centers. Patients and Methods: This was a retrospective analysis of data from a history-based computerized dementia diagnostic system. MT-VS and SBR among normal controls (NCs) and patients with PD, PD with dementia (PDD), DLB, or Alzheimer's disease (AD) were compared. Results: We included 715 individuals, including 122 NCs, 286 patients with PD, 40 with AD, 179 with DLB, and 88 with PDD. Compared with NCs, patients with PD exhibited a significantly higher prevalence of abnormal DAT uptake using all methods. Compared with the AD group, PDD and DLB groups exhibited a significantly higher prevalence of abnormal DAT uptake using all methods. The distribution of ML-VS was significantly different between PD and NC, DLB and AD, and PDD and AD groups (all p < 0.001). The correlation coefficient of ML-VS/SBR in all participants was 0.679. Conclusions: The ML-VS designed in one center is useful for differentiating PD from NC, DLB from AD, and PDD from AD in other centers. Its correlation with traditional approaches using different scanning machines is also acceptable. Future studies should develop models using data pools from multiple centers for increasing diagnostic accuracy.

Synuclein Motor Dysfunction Composite Scale for the Discrimination of Dementia With Lewy Bodies From Alzheimer's Disease.

Background: An abnormal increase of α-synuclein in the brain is the hallmark of dementia with Lewy bodies (DLB). However, the diagnostic power of plasma α-synuclein in DLB is not yet confirmed. Parkinsonism is highly associated with and is one of the core clinical features of DLB. We studied plasma α-synuclein and developed a novel tool that combined plasma α-synuclein level and Motor Dysfunction Questionnaire (MDQ), namely Synuclein Motor Dysfunction Composite Scale (SMDCS), for the clinical discrimination of DLB from Alzheimer's disease (AD). Methods: This cross-sectional study analyzed participants' demographical data, plasma α-synuclein level, MDQ, structured clinical history questionnaire, neuropsychological and motor function tests, and neuroimaging studies. The power of plasma α-synuclein level, MDQ, and SMDCS for discriminating DLB from non-demented controls (NC) or AD were compared. Results: Overall, 121 participants diagnosed as 58 DLB, 31 AD, and 31 NC were enrolled. Patients with DLB had significantly higher mean plasma α-synuclein level (0.24 ± 0.32 pg/ml) compared to the NC group (0.08 ± 0.05 pg/ml) and the AD group (0.08 ± 0.05 pg/ml). The DLB group demonstrated higher MDQ (2.95 ± 1.60) compared to the NC (0.42 ± 0.98) or AD (0.44 ± 0.99) groups. The sensitivity/specificity of plasma α-synuclein level, MDQ, and SMDCS for differentiating DLB from non-DLB were 0.80/0.64, 0.83/0.89, and 0.88/0.93, respectively. Conclusion: Both plasma α-synuclein and MDQ were significantly higher in patients with DLB compared to the NC or AD groups. The novel SMDCS, significantly improved accuracy for the clinical differentiation of DLB from AD or NC.

Emergency department visits among people with predementia highly predicts conversion to dementia.

Emergency department visits (EDV) are common among older adults with and without dementia. The risk factors and demands of EDVs for people with dementia have been well studied; however, the association between EDVs and conversion to dementia among people with predementia has not been thoroughly explored. To study the predictive value of EDVs in predementia's progression to dementia. The baseline predementia cohort registered from September 2015 to August 2017, with longitudinal follow-up in the History-based Artificial Intelligent Clinical Dementia Diagnostic System database, was retrospectively analyzed. The rates of conversion among the different EDVs were compared. Multivariate logistic regression and Cox proportional hazards analyses were applied to study the influence of EDVs on progression. Age, education, sex, neuropsychological tests, activities of daily living, neuropsychiatric symptoms, parkinsonism, and multiple vascular risk factors were adjusted for. A total of 512 participants were analyzed, including 339 (66.2%) non-converters and 173 (33.8%) converters with a mean follow-up of 3.3 (range 0.4-6.1) and 2.8 (range 0.5-5.9) years, respectively. Compared to people without EDV (EDV 0), the hazard ratios for conversion to dementia were 3.6, 5.9, and 6.9 in those with EDV once (EDV 1), twice (EDV 2), and more than twice (EDV >2), respectively. In addition, older age, lower education, poorer cognition, poorer ADL performance, and longer follow-up periods also increased the conversion rates. EDVs in the predementia stages highly predict progression to dementia. Therefore, a sound public health as well as primary healthcare system that provide strategies for better management of mental and physical condition might help prevention of EDVs among older people in the predementia stages.

Clinical impacts of scar reduction on gated myocardial perfusion SPECT after cardiac resynchronization therapy.

BACKGROUND: It had not been reported that myocardial scar shown on gated myocardial perfusion SPECT (GMPS) might reduce after cardiac resynchronization therapy (CRT). In this study, we aim to investigate the clinical impact and characteristic of scar reduction (SR) after CRT. METHODS AND RESULTS: Sixty-one heart failure patients following standard indication for CRT received twice GMPS as pre- and post-CRT evaluations. The patients with an absolute reduction of scar ≥ 10% after CRT were classified as the SR group while the rest were classified as the non-SR group. The SR group (N = 22, 36%) showed more improvement on LV function (∆LVEF: 18.1 ± 12.4 vs 9.4 ± 9.9 %, P = 0.007, ∆ESV: - 91.6 ± 52.6 vs - 38.1 ± 46.5 mL, P < 0.001) and dyssynchrony (ΔPSD: - 26.19 ± 18.42 vs - 5.8 ± 23.0°, P < 0.001, Δ BW: - 128.7 ± 82.8 vs - 25.2 ± 109.0°, P < 0.001) than non-SR group (N = 39, 64%). Multivariate logistic regression analysis showed baseline QRSd (95% CI 1.019-1.100, P = 0.006) and pre-CRT Reduced Wall Thickening (RWT) (95% CI 1.016-1.173, P = 0.028) were independent predictors for the development of SR. CONCLUSION: More than one third of patients showed SR after CRT who had more post-CRT improvement on LV function and dyssynchrony than those without SR. Wider QRSd and higher RWT before CRT were related to the development of SR after CRT.

The Feasibility of Differentiating Lewy Body Dementia and Alzheimer's Disease by Deep Learning Using ECD SPECT Images.

The correct differential diagnosis of dementia has an important impact on patient treatment and follow-up care strategies. Tc-99m-ECD SPECT imaging, which is low cost and accessible in general clinics, is used to identify the two common types of dementia, Alzheimer's disease (AD) and Lewy body dementia (LBD). Two-stage transfer learning technology and reducing model complexity based on the ResNet-50 model were performed using the ImageNet data set and ADNI database. To improve training accuracy, the three-dimensional image was reorganized into three sets of two-dimensional images for data augmentation and ensemble learning, then the performance of various deep learning models for Tc-99m-ECD SPECT images to distinguish AD/normal cognition (NC), LBD/NC, and AD/LBD were investigated. In the AD/NC, LBD/NC, and AD/LBD tasks, the AUC values were around 0.94, 0.95, and 0.74, regardless of training models, with an accuracy of 90%, 87%, and 71%, and F1 scores of 89%, 86%, and 76% in the best cases. The use of transfer learning and a modified model resulted in better prediction results, increasing the accuracy by 32% for AD/NC. The proposed method is practical and could rapidly utilize a deep learning model to automatically extract image features based on a small number of SPECT brain perfusion images in general clinics to objectively distinguish AD and LBD.

Motor Dysfunction Questionnaire and Dopamine Transporter Imaging Composite Scale Improve Differentiating Dementia With Lewy Bodies From Alzheimer's Disease With Motor Dysfunction.

Objective: Characteristic parkinsonism is the major comorbidity of dementia with Lewy bodies (DLB). We aimed to differentiate DLB from Alzheimer's disease (AD) with motor dysfunction using a composite scale with a characteristic motor dysfunction questionnaire (MDQ) and dopamine transporter (DAT) imaging. It could help detect DLB easily in healthcare settings without movement disorder specialists. Methods: This is a two-phase study. In the design phase, seven questions were selected and composed of a novel MDQ. In the test phase, all participants with DLB, AD, or non-dementia (ND) control completed dementia and parkinsonism survey, the novel designed questionnaire, DAT imaging, and composite scales of MDQ and DAT. The cutoff scores of the MDQ, semiquantitative analysis of the striatal-background ratio (SBR) and visual rating of DAT, and the composite scale of MDQ and DAT for discriminating DLB from AD or ND were derived and compared. Results: A total of 277 participants were included in this study (126 with DLB, 86 with AD, and 65 with ND). Compared with the AD or ND groups, the DLB group showed a significantly higher frequency in all seven items in the MDQ and a significantly lower SBR. For discrimination of DLB from non-DLB with MDQ, SBR, and composite scale, the cutoff scores of 3/2, 1.37/1.38, and 6/5 were suggested for the diagnosis of DLB with the sensitivities/specificities of 0.91/0.72, 0.91/0.80, and 0.87/0.93, respectively. The composite scale significantly improved the accuracy of discrimination compared with either the MDQ or SBR. Conclusion: This study showed that the novel designed simple questionnaire was a practical screening tool and had similar power to DAT scanning to detect DLB. The questionnaire can be applied in clinical practice and population studies for screening DLB. In addition, the composite scale of MDQ and DAT imaging further improved the diagnostic accuracy, indicating the superiority of the dual-model diagnostic tool.

Detection of Alzheimer's disease using ECD SPECT images by transfer learning from FDG PET.

OBJECTIVE: To develop a practical method to rapidly utilize a deep learning model to automatically extract image features based on a small number of SPECT brain perfusion images in general clinics to objectively evaluate Alzheimer's disease (AD). METHODS: For the properties of low cost and convenient access in general clinics, Tc-99-ECD SPECT imaging data in brain perfusion detection was used in this study for AD detection. Two-stage transfer learning based on the Inception v3 network model was performed using the ImageNet dataset and ADNI database. To improve training accuracy, the three-dimensional image was reorganized into three sets of two-dimensional images for data augmentation and ensemble learning. The effect of pre-training parameters for Tc-99m-ECD SPECT image to distinguish AD from normal cognition (NC) was investigated, as well as the effect of the sample size of F-18-FDG PET images used in pre-training. The same model was also fine-tuned for the prediction of the MMSE score from the Tc-99m-ECD SPECT image. RESULTS: The AUC values of w/wo pre-training parameters for Tc-99m-ECD SPECT image to distinguish AD from NC were 0.86 and 0.90. The sensitivity, specificity, precision, accuracy, and F1 score were 100%, 75%, 76%, 86%, and 86%, respectively for the training model with 1000 cases of F-18-FDG PET image for pre-training. The AUC values for various sample sizes of the training dataset (100, 200, 400, 800, 1000 cases) for pre-training were 0.86, 0.91, 0.95, 0.97, and 0.97. Regardless of the pre-training condition ECD dataset used, the AUC value was greater than 0.85. Finally, predicting cognitive scores and MMSE scores correlated (R2 = 0.7072). CONCLUSIONS: With the ADNI pre-trained model, the sensitivity and accuracy of the proposed deep learning model using SPECT ECD perfusion images to differentiate AD from NC were increased by approximately 30% and 10%, respectively. Our study indicated that the model trained on PET FDG metabolic imaging for the same disease could be transferred to a small sample of SPECT cerebral perfusion images. This model will contribute to the practicality of SPECT cerebral perfusion images using deep learning technology to objectively recognize AD.

Left-ventricular dyssynchrony in viable myocardium by myocardial perfusion SPECT is predictive of mechanical response to CRT.

OBJECTIVES: Gated myocardial perfusion SPECT (GMPS) provides a one-stop-shop evaluation for cardiac resynchronization therapy (CRT). However, conflicting results have been observed regarding whether the baseline left-ventricular (LV) mechanical dyssynchrony as assessed by phase analysis on GMPS was predictive of therapeutic response to CRT. Since dyssynchrony parameters by phase analysis spuriously increased by scarred myocardium, the purpose of this study was to explore the value of dyssynchrony after stripping off the scar region in correlation to mechanical response to CRT. METHODS: Forty-seven patients following standard indications for CRT received GMPS with phase analysis as pre-CRT evaluation. A decrease of end-systolic volume (ESV) > 15% on follow-up echocardiography after CRT was considered as a mechanical response to CRT. Myocardial regions with less than 50% of maximal activity on GMPS were considered as a scar. The phase standard deviation (PSD) and histogram bandwidth (BW) without or with stripping off scar were assessed by phase analysis of GMPS and were used for evaluation of LV dyssynchrony of all myocardium or only the viable myocardium, respectively. RESULTS: No significant difference was noted between mechanical responders (31 of 47 patients, 66%) and nonresponders ( 16 of 47 patients, 34%) for PSD (48.6° ± 19.4° vs 43.9° ± 20.7°, p = 0.46) and BW (225° ± 91.1° vs 163.5° ± 94.6°, p = 0.38) of the entire myocardium. However, responders had significantly larger PSD (40.5° ± 15.7° vs 30.5° ± 13.2°, p = 0.03) and borderlinely larger BW (215° ± 91.2° vs. 139.5° ± 78.2°, p = 0.05) than non-responders after stripping off scar. Logistic regression analysis showed that scar area and PSD after stripping off scar were independent predictors of mechanical response. CONCLUSIONS: Our result showed that LV dyssynchrony of the entire myocardium did not predict response to CRT. However, LV dyssynchrony only in the viable myocardium was a significant predictor of CRT mechanical response.

Function-based dementia severity assessment for vascular cognitive impairment.

BACKGROUND/PURPOSES: Unimpaired activities of daily living (ADL) is essential for the diagnosis of normal cognition and mild cognitive impairment. However, diagnosis according to this concept is difficult to apply to patients comorbid with motor dysfunction. We aim to use a novel ADL questionnaire for operationally diagnosing unimpaired ADL in vascular cognitive impairment with no dementia (VCIND). METHODS AND PARTICIPANTS: This was a retrospective cohort study with both cross-sectional and long-term follow-up analysis. Patients with cerebrovascular disease with normal cognition (CVDNC), VCIND, and vascular dementia (VaD) were analyzed. Cutoff scores for differentiating different stages of cognitive impairment were compared between the new History-based Artificial Intelligent ADL questionnaire (HAI-ADL) and other tools. RESULTS: A total of 596 individuals were analyzed, including 40 CVDNC, 167 VCIND, 218 mild, 119 moderate, and 52 severe-dementia patients. The cutoff scores for determining unimpaired ADL in VCIND were 8.5, 3.5, 5, 100, and 60 in HAI-ADL, CDR-SB, IADL, BI, and CASI, respectively. HAI-ADL had the highest correlations with CDR-SB and the CDR staging system compared to other tools. Four models of progression rates from CVDNC/VCIND to VaD revealed it was much higher in the group with HAI-ADL > 8.5 compared to those with HAI-ADL≦8.5 with odds ratios of 3.75, 3.66, 3.31, and 2.77, respectively. CONCLUSION: Our study showed that HAI-ADL provides an operational determinates unimpaired ADL which is necessary for the diagnosis of VCIND. The predictive value for progression to dementia was proved by a long-term follow-up analysis of the research cohort.

Comparison of ejection fraction calculation between CT and SPECT at high heart rate: A dynamic cardiac phantom study.

The purpose of this study is to compare the ejection fraction (EF) calculation of CT and SPECT at high heart rate. A dynamic cardiac phantom with programmable end-systolic volume (ESV), end-diastolic volume (EDV), and heart rate was used to compare CT, which has high spatial resolution (< 1 mm) and modest temporal resolution of 175 msec, and SPECT, which has high temporal resolution of 16 bins per cardiac cycle but poor spatial resolution (> 1 cm) in EF, ESV, and EDV at the heart rates ≤ 100 bpm for EF = 30 (disease state) and EF = 60 (healthy state). EF calculations for SPECT were accurate in 2% for 40 to 100 bpm for both EF = 30 and EF = 60, and were not heart rate dependent although both ESV and EDV could be underestimated by 18-20%. EF calculations for CT were accurate in 2.2% for 40 and 60 bpm. Inaccuracy in EF calculations, ESV and EDV estimates increased when the heart rate or EF increased. SPECT was accurate for EF calculation for the heart rates ≤ 100 bpm and CT was accurate for the heart rates of ≤ 60 bpm. CT was less accurate for the high heart rates of 80 and 100 bpm, or high EF = 60.

Tremor in Vascular Cognitive Impairment Raises the Possibility of Mixed Pathology With Lewy Body Disease.

OBJECTIVES: Tremor is common in patients with Lewy body disease (LBD) and not rare in normal individuals. Prevalence of tremor in patients with vascular cognitive impairment (VCI) and its association with other comorbidities are seldom studied. The aim of this study was to investigate the patient characteristics of VCI associated with tremor and to evaluate the possibility of mixed pathology with LBD in these patients. METHODS: Retrospective analysis of a large population with VCI registered in the database of a regional healthcare system was performed. VCI patients were divided into tremor and non-tremor groups. The associated characteristics including demographics, clinical features in motor and non-motor domains, vascular risk factors, and neuroimaging features were compared between the tremor group and the non-tremor group. RESULTS: Among 1337 patients with VCI, 292 (21.8%) had tremor, while 1045 (78.2%) did not have tremor. The tremor group had significantly higher prevalence of all motor and non-motor LBD clinical features than the non-tremor group. The tremor group also demonstrated more severe neuropsychiatric symptoms. Among patients with tremor, patients having tremor onset earlier than stroke onset showed significantly higher prevalence of rapid eye movement sleep behavior disorder. All comparisons were adjusted for age and severity of dementia. CONCLUSION: Tremor is a common comorbidity of VCI. VCI patients with tremor had a higher prevalence of motor and non-motor LBD features. These findings raised the possibility of VCI patients with tremor having high possibility of mixed pathology with LBD.

Freezing of Speech Single Questionnaire as a Screening Tool for Cognitive Dysfunction in Patients With Dementia With Lewy Bodies.

INTRODUCTION: Freezing phenomenon is a striking feature of Parkinson's disease. However, it has never been studied in people with dementia with Lewy bodies (DLB). We designed a freezing of speech single questionnaire (FOSSQ) and investigated the frequency and association of freezing of speech (FOS) in patients with DLB and other types of dementia. METHODS: This is a retrospective analysis of data from the project of history-based artificial intelligent computerized dementia diagnostic system. We compared the frequencies of FOS among non-demented (ND) participants, patients with Alzheimer's disease (AD), vascular dementia (VaD), and DLB. Further, we explored the association factors of FOS in all the participants. RESULTS: We enrolled 666 individuals with the following disease distribution: 190, ND; 230, AD; 183, VaD; and 63, DLB. Compared to individuals with ND (2.1%), patients with AD (6.1%), or VaD (18.0%), DLB (54.0%) showed a significantly higher frequency of positive FOS (all p < 0.001). The association factors of FOS were older age, more severe dementia, more severe motor dysfunction, fluctuating cognition, visual hallucinations, parkinsonism, rapid eye movement sleep behavior disorder, attention, mental manipulation, and language. CONCLUSION: Our study showed that the informant-based FOSSQ may be a practical screening tool for discriminating DLB from individuals with ND or other forms of dementia. The FOSSQ can be applied in clinical practice as well as on the artificial intelligent platform.

An Informant-Based Simple Questionnaire for Visuospatial Dysfunction Assessment in Dementia.

OBJECTIVES: Visuospatial dysfunction (VSD) is one of the most important symptoms for the diagnosis of dementia with Lewy bodies (DLB). The aim of this study was to validate a novel VSD questionnaire and determine the cutoff score for the screening for VSD in DLB. METHODS: This is a retrospective analysis of data from a project of the History-based Artificial Intelligent Clinical Dementia Diagnostic System (HAICDDS). VSD of non-demented control (NDC), Alzheimer's disease (AD), and DLB participants were analyzed and compared using the visuospatial questionnaire in the HAICDDS (HAI-VSQ), the Draw subscale in the Cognitive Abilities Screening Instrument (CASI-Draw), and the visuospatial subscale in Montreal Cognitive Assessment (MoCA-VS). RESULTS: A total of 440 individuals were studied, including 154 NDC, 229 AD, and 57 DLB participants. Compared to NDC or AD participants, DLB participants showed a higher total score on HAI-VSQ after adjustment for age. Using HAI-VSQ, a cutoff score ≥ 2 was useful for the screening for VSD in DLB with a sensitivity of 0.77 and a specificity of 0.94. Compared with CASI-Draw or MoCA-VS, HAI-VSQ was least influenced by gender, age, and education and had the highest correlation with the sum of boxes of the Clinical Dementia Rating scale. After adjustment for age, education, gender, and global cognitive function, HAI-VSQ significantly discriminated DLB from AD and NDC whereas MoCA-VS or CASI-Draw did not. CONCLUSION: Our study showed that the newly designed simple questionnaire was a practical screening tool for VSD in DLB that can be applied in clinical practice as well as on a registration platform.