Automated Deep Learning Segmentation of Cardiac Inflammatory FDG PET.

BACKGROUND: Fluorodeoxyglucose positron emission tomography (FDG PET) with suppression of myocardial glucose utilization plays a pivotal role in diagnosing cardiac sarcoidosis. Reorientation of images to match perfusion datasets and myocardial segmentation enables consistent image scaling and quantification. However, such manual tasks are cumbersome. We developed a 3D U-Net deep-learning (DL) algorithm for automated myocardial segmentation in cardiac sarcoidosis FDG PET. METHODS: The DL model was trained on FDG PET scans from 316 patients with left ventricular contours derived from paired perfusion datasets. Qualitative analysis of clinical readability was performed to compare DL segmentation with the current automated method on a 50-patient test subset. Additionally, left ventricle displacement and angulation, as well as SUVmax sampling were compared to inter-user reproducibility results. A hybrid workflow was also investigated to accelerate study processing time. RESULTS: DL segmentation enhanced readability scores in over 90% of cases compared to the standard segmentation currently used in the software. DL segmentation performed similarly to a trained technologist, surpassing standard segmentation for left ventricle displacement and angulation, as well as correlation of SUVmax. Using the DL segmentation as initial placement for manual segmentation significantly decreased processing time. CONCLUSION: A novel DL-based automated segmentation tool markedly improves processing of cardiac sarcoidosis FDG PET. This tool yields optimized splash display of sarcoidosis FDG PET datasets with no user input and offers significant processing time improvement for manual segmentation of such datasets.

Choline Metabolites and 15-Year Risk of Incident Diabetes in a Prospective Cohort of Adults: Coronary Artery Risk Development in Young Adults (CARDIA) Study.

OBJECTIVE: The potential for choline metabolism to influence the development of diabetes has received increased attention. Previous studies on circulating choline metabolites and incident diabetes have been conducted in samples of older adults, often with a high prevalence of risk factors. RESEARCH DESIGN AND METHODS: Participants were from year 15 of follow-up (2000-2001) in the Coronary Artery Risk Development in Young Adults (CARDIA) Study (n = 3,133, aged 33-45 years) with plasma choline metabolite (choline, betaine, and trimethylamine N-oxide [TMAO]) data. We quantified associations between choline metabolites and 15-year risk of incident diabetes (n = 387) among participants free of diabetes at baseline using Cox proportional hazards regression models adjusted for sociodemographics, health behaviors, and clinical variables. RESULTS: Betaine was inversely associated with 15-year risk of incident diabetes (hazard ratio 0.76 [95% CI 0.67, 0.88] per 1-SD unit betaine), and TMAO was positively associated with 15-year risk of incident diabetes (1.11 [1.01, 1.22] per 1-SD unit). Choline was not significantly associated with 15-year risk of incident diabetes (1.05 [0.94, 1.16] per 1-SD). CONCLUSIONS: Our findings are consistent with other published literature supporting a role for choline metabolism in diabetes. Our study extends the current literature by analyzing a racially diverse population-based cohort of early middle-aged individuals in whom preventive activities may be most relevant.

Development and validation of a machine learning model to predict myocardial blood flow and clinical outcomes from patients' electrocardiograms.

We develop a machine learning (ML) model using electrocardiography (ECG) to predict myocardial blood flow reserve (MFR) and assess its prognostic value for major adverse cardiovascular events (MACEs). Using 3,639 ECG-positron emission tomography (PET) and 17,649 ECG-single-photon emission computed tomography (SPECT) data pairs, the ML model is trained with a swarm intelligence approach and support vector regression (SVR). The model achieves a receiver-operator curve (ROC) area under the curve (AUC) of 0.83, with a sensitivity and specificity of 0.75. An ECG-MFR value below 2 is significantly associated with MACE, with hazard ratios (HRs) of 3.85 and 3.70 in the discovery and validation phases, respectively. The model's C-statistic is 0.76, with a net reclassification improvement (NRI) of 0.35. Validated in an independent cohort, the ML model using ECG data offers superior MACE prediction compared to baseline clinical models, highlighting its potential for risk stratification in patients with coronary artery disease (CAD) using the accessible 12-lead ECG.

Prediction of Cumulative Exposure to Atherogenic Lipids During Early Adulthood.

BACKGROUND: The ability of a 1-time measurement of non-high-density lipoprotein cholesterol (non-HDL-C) or low-density lipoprotein cholesterol (LDL-C) to predict the cumulative exposure to these lipids during early adulthood (age 18-40 years) and the associated atherosclerotic cardiovascular disease (ASCVD) risk after age 40 years is not clear. OBJECTIVES: The objectives of this study were to evaluate whether a 1-time measurement of non-HDL-C or LDL-C in a young adult can predict cumulative exposure to these lipids during early adulthood, and to quantify the association between cumulative exposure to non-HDL-C or LDL-C during early adulthood and the risk of ASCVD after age 40 years. METHODS: We included CARDIA (Coronary Artery Risk Development in Young Adults Study) participants who were free of cardiovascular disease before age 40 years, were not taking lipid-lowering medications, and had ≥3 measurements of LDL-C and non-HDL-C before age 40 years. First, we assessed the ability of a 1-time measurement of LDL-C or non-HDL-C obtained between age 18 and 30 years to predict the quartile of cumulative lipid exposure from ages 18 to 40 years. Second, we assessed the associations between quartiles of cumulative lipid exposure from ages 18 to 40 years with ASCVD events (fatal and nonfatal myocardial infarction and stroke) after age 40 years. RESULTS: Of 4,104 CARDIA participants who had multiple lipid measurements before and after age 30 years, 3,995 participants met our inclusion criteria and were in the final analysis set. A 1-time measure of non-HDL-C and LDL-C had excellent discrimination for predicting membership in the top or bottom quartiles of cumulative exposure (AUC: 0.93 for the 4 models). The absolute values of non-HDL-C and LDL-C that predicted membership in the top quartiles with the highest simultaneous sensitivity and specificity (highest Youden's Index) were >135 mg/dL for non-HDL-C and >118 mg/dL for LDL-C; the values that predicted membership in the bottom quartiles were <107 mg/dL for non-HDL-C and <96 mg/dL for LDL-C. Individuals in the top quartile of non-HDL-C and LDL-C exposure had demographic-adjusted HRs of 4.6 (95% CI: 2.84-7.29) and 4.0 (95% CI: 2.50-6.33) for ASCVD events after age 40 years, respectively, when compared with each bottom quartile. CONCLUSIONS: Single measures of non-HDL-C and LDL-C obtained between ages 18 and 30 years are highly predictive of cumulative exposure before age 40 years, which in turn strongly predicts later-life ASCVD events.

Impact of Social Determinants of Health and Lifestyle on Association Between Lipoprotein(a) and Cardiovascular Events.

Background: In European cohorts, healthier lifestyle either attenuated or associated with lower cardiovascular risk despite elevated lipoprotein(a) [Lp(a)]. Objectives: The purpose of this study was to test if social determinants of health (SDOH) and Life's Simple 7 (LS7) scores impact the association of Lp(a) with cardiovascular events in U.S. cohorts. Methods: We performed a sequential multivariable Cox proportional hazard analysis using the ARIC (Atherosclerosis Risk In Communities) and MESA (Multi-Ethnic Study of Atherosclerosis) cohorts. We first adjusted for age, gender, non-high-density lipoprotein-cholesterol, race, and ethnicity, then sequentially added SDOH and LS7 scores. The primary outcomes were time until first myocardial infarction (MI) or stroke. Results: ARIC (n = 15,072; median Lp(a) = 17.3 mg/dL) had 16.2 years and MESA (n = 6,822; median Lp(a) = 18.3 mg/dL) had 12.3 years of average follow-up. In age, gender, race, and ethnicity, and non-high-density lipoprotein-cholesterol adjusted analyses, Lp(a) was associated with MI in ARIC (HR: 1.10, P < 0.001) and MESA (HR: 1.11, P = 0.001), and stroke in ARIC (HR: 1.07, P < 0.001) but not MESA (HR: 0.97, P = 0.53). In models with SDOH and LS7, associations of Lp(a) remained similar with MI (ARIC, HR: 1.08, P < 0.001; MESA, HR: 1.10, P = 0.001) and stroke (ARIC, HR: 1.06, P = 0.002; MESA, HR: 0.96, P = 0.37). Each additional SDOH correlated positively with MI (ARIC, HR: 1.04, P = 0.01; MESA, HR: 1.08, P = 0.003) and stroke in ARIC (HR: 1.08, P = 0.00) but not MESA (HR: 1.03, P = 0.41). Each additional LS7 point correlated negatively with MI (ARIC, HR: 0.88, P < 0.001; MESA, HR: 0.85, P < 0.001) and stroke (ARIC, HR: 0.91, P < 0.001; MESA, HR: 0.86, P < 0.001). Conclusions: SDOH and lifestyle factors associated with risk for MI and stroke but did not largely impact the association between Lp(a) and cardiovascular events.

Projected Changes in Statin and Antihypertensive Therapy Eligibility With the AHA PREVENT Cardiovascular Risk Equations.

Importance: Since 2013, the American College of Cardiology (ACC) and American Heart Association (AHA) have recommended the pooled cohort equations (PCEs) for estimating the 10-year risk of atherosclerotic cardiovascular disease (ASCVD). An AHA scientific advisory group recently developed the Predicting Risk of cardiovascular disease EVENTs (PREVENT) equations, which incorporated kidney measures, removed race as an input, and improved calibration in contemporary populations. PREVENT is known to produce ASCVD risk predictions that are lower than those produced by the PCEs, but the potential clinical implications have not been quantified. Objective: To estimate the number of US adults who would experience changes in risk categorization, treatment eligibility, or clinical outcomes when applying PREVENT equations to existing ACC and AHA guidelines. Design, Setting, and Participants: Nationally representative cross-sectional sample of 7765 US adults aged 30 to 79 years who participated in the National Health and Nutrition Examination Surveys of 2011 to March 2020, which had response rates ranging from 47% to 70%. Main Outcomes and Measures: Differences in predicted 10-year ASCVD risk, ACC and AHA risk categorization, eligibility for statin or antihypertensive therapy, and projected occurrences of myocardial infarction or stroke. Results: In a nationally representative sample of 7765 US adults aged 30 to 79 years (median age, 53 years; 51.3% women), it was estimated that using PREVENT equations would reclassify approximately half of US adults to lower ACC and AHA risk categories (53.0% [95% CI, 51.2%-54.8%]) and very few US adults to higher risk categories (0.41% [95% CI, 0.25%-0.62%]). The number of US adults receiving or recommended for preventive treatment would decrease by an estimated 14.3 million (95% CI, 12.6 million-15.9 million) for statin therapy and 2.62 million (95% CI, 2.02 million-3.21 million) for antihypertensive therapy. The study estimated that, over 10 years, these decreases in treatment eligibility could result in 107 000 additional occurrences of myocardial infarction or stroke. Eligibility changes would affect twice as many men as women and a greater proportion of Black adults than White adults. Conclusion and Relevance: By assigning lower ASCVD risk predictions, application of the PREVENT equations to existing treatment thresholds could reduce eligibility for statin and antihypertensive therapy among 15.8 million US adults.

Prospective Study to Evaluate the Role of Dual Point Contrast-enhanced Magnetic Resonance Imaging in Differentiation of Brain Tumoral from Nontumoral Tissue: A Magnetic Resonance/PET Study.

Background and Purpose: Follow-up imaging of gliomas is crucial to look for residual or recurrence and to differentiate them from nontumoral tissue. Positron emission tomography (PET)-magnetic resonance imaging (MRI) is the problem-solving tool in such cases. We investigated the role of dual point contrast (DPC)-enhanced MRI to discriminate tumoral from the nontumoral tissue compared to PET-MRI taken as the gold standard. Materials and Methods: The institutional ethics committee approved the study, and consent was obtained from all the patients included in the study. We prospectively did immediate and 75-min delayed contrast MRI in glioma cases who came for follow-up as a part of PET-MRI study in our institute. Subtracted images were obtained using immediate and 75-min delayed contrast images. Color-coded subtracted images were compared with PET-MRI images. 75-min delayed contrast MRI and diffusion-weighted imaging (DWI) images with Gray Scale inversion were compared with PET attenuation-corrected images. Results: We included 23 PET MRI cases done with different radiotracers in our study. Overall, we found PET-DPC correlation in (20/20 ~ 100%) cases of enhancing tumors. In two cases (DOPA and fluorodeoxyglucose), since they were nonenhancing low-grade gliomas and the other one was melanoma with intrinsic T1 hyperintensity and the DPC technique could not be used. DWI-PET correlated in 17/19 (~89.4%) cases, and perfusion-weighted imaging (PWI)-PET dynamic susceptibility contrast (DSC)/ASL correlated in 14/18 (~77.7%) cases after cases with hemorrhage were excluded. Conclusion: DPC MRI showed a good correlation with PET MRI in discriminating tumoral from the nontumoral tissue. DPC MRI can act as a potential alternative to PET MRI in peripheral hospitals where PET is not available. However, the DPC technique is limited in low-grade nonenhancing gliomas.

Proteomic analysis of cardiorespiratory fitness for prediction of mortality and multisystem disease risks.

Despite the wide effects of cardiorespiratory fitness (CRF) on metabolic, cardiovascular, pulmonary and neurological health, challenges in the feasibility and reproducibility of CRF measurements have impeded its use for clinical decision-making. Here we link proteomic profiles to CRF in 14,145 individuals across four international cohorts with diverse CRF ascertainment methods to establish, validate and characterize a proteomic CRF score. In a cohort of around 22,000 individuals in the UK Biobank, a proteomic CRF score was associated with a reduced risk of all-cause mortality (unadjusted hazard ratio 0.50 (95% confidence interval 0.48-0.52) per 1 s.d. increase). The proteomic CRF score was also associated with multisystem disease risk and provided risk reclassification and discrimination beyond clinical risk factors, as well as modulating high polygenic risk of certain diseases. Finally, we observed dynamicity of the proteomic CRF score in individuals who undertook a 20-week exercise training program and an association of the score with the degree of the effect of training on CRF, suggesting potential use of the score for personalization of exercise recommendations. These results indicate that population-based proteomics provides biologically relevant molecular readouts of CRF that are additive to genetic risk, potentially modifiable and clinically translatable.

Shared plasma metabolomic profiles of cognitive and mobility decline predict future dementia.

Experiencing decline in both cognition and mobility is associated with a substantially higher dementia risk than cognitive decline only. Metabolites associated with both cognitive and mobility declines may be early predictors of dementia and reveal specific pathways to dementia. We analyzed data from 2450 participants initially free of dementia who had 613 metabolites measured in plasma in 1998-1999 (mean age = 75.2 ± 2.9 years old, 37.8% Black, 50% women) from the Health, Aging and Body Composition study. Dementia diagnosis was determined by race-specific decline in 3MS scores, medication use, and hospital records through 2014. Cognition and mobility were repeatedly measured using 3MS and a 20-m walking test up to 10 years, respectively. We examined metabolite associations with changes in 3MS (n = 2046) and gait speed (n = 2019) using multivariable linear regression adjusted for age, sex, race, and baseline performance and examined metabolite associations with dementia risk using Cox regression. During a mean follow-up of 9.3 years, 534 (21.8%) participants developed dementia. On average, 3MS declined 0.47/year and gait declined 0.04 m/sec/year. After covariate adjustment, 75 metabolites were associated with cognitive decline, and 111 metabolites were associated with gait decline (FDR-adjusted p < 0.05). Twenty-six metabolites were associated with both cognitive and gait declines. Eighteen of 26 metabolites were associated with dementia risk (p < 0.05), notably amino acids, glycerophospholipids (lysoPCs, PCs, PEs), and sphingolipids. Results remained similar after adjusting for cardiovascular disease or apolipoprotein E ɛ4 carrier status. During aging, metabolomic profiles of cognitive decline and mobility decline show distinct and shared signatures. Shared metabolomic profiles suggest that inflammation and deficits in mitochondria and the urea cycle in addition to the central nervous system may play key roles in both cognitive and mobility declines and predict dementia. Future studies are warranted to investigate longitudinal metabolite changes and metabolomic markers with dementia pathologies.

Metabolic liability for weight gain in early adulthood.

While weight gain is associated with a host of chronic illnesses, efforts in obesity have relied on single "snapshots" of body mass index (BMI) to guide genetic and molecular discovery. Here, we study >2,000 young adults with metabolomics and proteomics to identify a metabolic liability to weight gain in early adulthood. Using longitudinal regression and penalized regression, we identify a metabolic signature for weight liability, associated with a 2.6% (2.0%-3.2%, p = 7.5 × 10-19) gain in BMI over ≈20 years per SD higher score, after comprehensive adjustment. Identified molecules specified mechanisms of weight gain, including hunger and appetite regulation, energy expenditure, gut microbial metabolism, and host interaction with external exposure. Integration of longitudinal and concurrent measures in regression with Mendelian randomization highlights the complexity of metabolic regulation of weight gain, suggesting caution in interpretation of epidemiologic or genetic effect estimates traditionally used in metabolic research.

One nose but two nostrils: Learn to align with sparse connections between two olfactory cortices.

The integration of neural representations in the two hemispheres is an important problem in neuroscience. Recent experiments revealed that odor responses in cortical neurons driven by separate stimulation of the two nostrils are highly correlated. This bilateral alignment points to structured inter-hemispheric connections, but detailed mechanism remains unclear. Here, we hypothesized that continuous exposure to environmental odors shapes these projections and modeled it as online learning with local Hebbian rule. We found that Hebbian learning with sparse connections achieves bilateral alignment, exhibiting a linear trade-off between speed and accuracy. We identified an inverse scaling relationship between the number of cortical neurons and the inter-hemispheric projection density required for desired alignment accuracy, i.e., more cortical neurons allow sparser inter-hemispheric projections. We next compared the alignment performance of local Hebbian rule and the global stochastic-gradient-descent (SGD) learning for artificial neural networks. We found that although SGD leads to the same alignment accuracy with modestly sparser connectivity, the same inverse scaling relation holds. We showed that their similar performance originates from the fact that the update vectors of the two learning rules align significantly throughout the learning process. This insight may inspire efficient sparse local learning algorithms for more complex problems.

Genotype-based "virtual" metabolomics in a clinical biobank identifies novel metabolite-disease associations.

Introduction: Circulating metabolites act as biomarkers of dysregulated metabolism and may inform disease pathophysiology. A portion of the inter-individual variability in circulating metabolites is influenced by common genetic variation. We evaluated whether a genetics-based "virtual" metabolomics approach can identify novel metabolite-disease associations. Methods: We examined the association between polygenic scores for 724 metabolites with 1,247 clinical phenotypes in the BioVU DNA biobank, comprising 57,735 European ancestry and 15,754 African ancestry participants. We applied Mendelian randomization (MR) to probe significant relationships and validated significant MR associations using independent GWAS of candidate phenotypes. Results and Discussion: We found significant associations between 336 metabolites and 168 phenotypes in European ancestry and 107 metabolites and 56 phenotypes in African ancestry. Of these metabolite-disease pairs, MR analyses confirmed associations between 73 metabolites and 53 phenotypes in European ancestry. Of 22 metabolitephenotype pairs evaluated for replication in independent GWAS, 16 were significant (false discovery rate p < 0.05). These included associations between bilirubin and X-21796 with cholelithiasis, phosphatidylcholine (16:0/22:5n3,18:1/20:4) and arachidonate with inflammatory bowel disease and Crohn's disease, and campesterol with coronary artery disease and myocardial infarction. These associations may represent biomarkers or potentially targetable mediators of disease risk.

Self-supervised learning for interventional image analytics: toward robust device trackers.

Purpose: The accurate detection and tracking of devices, such as guiding catheters in live X-ray image acquisitions, are essential prerequisites for endovascular cardiac interventions. This information is leveraged for procedural guidance, e.g., directing stent placements. To ensure procedural safety and efficacy, there is a need for high robustness/no failures during tracking. To achieve this, one needs to efficiently tackle challenges, such as device obscuration by the contrast agent or other external devices or wires and changes in the field-of-view or acquisition angle, as well as the continuous movement due to cardiac and respiratory motion. Approach: To overcome the aforementioned challenges, we propose an approach to learn spatio-temporal features from a very large data cohort of over 16 million interventional X-ray frames using self-supervision for image sequence data. Our approach is based on a masked image modeling technique that leverages frame interpolation-based reconstruction to learn fine inter-frame temporal correspondences. The features encoded in the resulting model are fine-tuned downstream in a light-weight model. Results: Our approach achieves state-of-the-art performance, in particular for robustness, compared to ultra optimized reference solutions (that use multi-stage feature fusion or multi-task and flow regularization). The experiments show that our method achieves a 66.31% reduction in the maximum tracking error against the reference solutions (23.20% when flow regularization is used), achieving a success score of 97.95% at a 3× faster inference speed of 42 frames-per-second (on GPU). In addition, we achieve a 20% reduction in the standard deviation of errors, which indicates a much more stable tracking performance. Conclusions: The proposed data-driven approach achieves superior performance, particularly in robustness and speed compared with the frequently used multi-modular approaches for device tracking. The results encourage the use of our approach in various other tasks within interventional image analytics that require effective understanding of spatio-temporal semantics.

Proteomics, Human Environmental Exposure, and Cardiometabolic Risk.

BACKGROUND: The biological mechanisms linking environmental exposures with cardiovascular disease pathobiology are incompletely understood. We sought to identify circulating proteomic signatures of environmental exposures and examine their associations with cardiometabolic and respiratory disease in observational cohort studies. METHODS: We tested the relations of >6500 circulating proteins with 29 environmental exposures across the built environment, green space, air pollution, temperature, and social vulnerability indicators in ≈3000 participants of the CARDIA study (Coronary Artery Risk Development in Young Adults) across 4 centers using penalized and ordinary linear regression. In >3500 participants from FHS (Framingham Heart Study) and JHS (Jackson Heart Study), we evaluated the prospective relations of proteomic signatures of the envirome with cardiovascular disease and mortality using Cox models. RESULTS: Proteomic signatures of the envirome identified novel/established cardiovascular disease-relevant pathways including DNA damage, fibrosis, inflammation, and mitochondrial function. The proteomic signatures of the envirome were broadly related to cardiometabolic disease and respiratory phenotypes (eg, body mass index, lipids, and left ventricular mass) in CARDIA, with replication in FHS/JHS. A proteomic signature of social vulnerability was associated with a composite of cardiovascular disease/mortality (1428 events; FHS: hazard ratio, 1.16 [95% CI, 1.08-1.24]; P=1.77×10-5; JHS: hazard ratio, 1.25 [95% CI, 1.14-1.38]; P=6.38×10-6; hazard ratio expressed as per 1 SD increase in proteomic signature), robust to adjustment for known clinical risk factors. CONCLUSIONS: Environmental exposures are related to an inflammatory-metabolic proteome, which identifies individuals with cardiometabolic disease and respiratory phenotypes and outcomes. Future work examining the dynamic impact of the environment on human cardiometabolic health is warranted.

Human metabolic chambers reveal a coordinated metabolic-physiologic response to nutrition.

The emerging field of precision nutrition is based on the notion that inter-individual responses across diets of different calorie-macronutrient content may contribute to inter-individual differences in metabolism, adiposity, and weight gain. Free-living diet studies have been traditionally challenged by difficulties in controlling adherence to prescribed calories and macronutrient content and rarely allow a period of metabolic stability prior to metabolic measures (to minimize influences of weight changes). In this context, key physiologic measures central to precision nutrition responses may be most precisely quantified via whole room indirect calorimetry over 24-h, in which precise control of activity and nutrition can be achieved. In addition, these studies represent unique "N of 1" human crossover metabolic-physiologic experiments during which specific molecular pathways central to nutrient metabolism may be discerned. Here, we quantified 263 circulating metabolites during a ≈40-day inpatient admission in which up to 94 participants underwent seven monitored 24-h nutritional interventions of differing macronutrient composition in a whole-room indirect calorimeter to capture precision metabolic responses. Broadly, we observed heterogenous responses in metabolites across dietary chambers, with the exception of carnitines which tracked with 24-h respiratory quotient. We identified excursions in shared metabolic species (e.g., carnitines, glycerophospholipids, amino acids) that mapped onto gold-standard calorimetric measures of substrate oxidation preference and lipid availability. These findings support a coordinated metabolic-physiologic response to nutrition, highlighting the relevance of these controlled settings to uncover biological pathways of energy utilization during precision nutrition studies.

Distinct information conveyed to the olfactory bulb by feedforward input from the nose and feedback from the cortex.

Sensory systems are organized hierarchically, but feedback projections frequently disrupt this order. In the olfactory bulb (OB), cortical feedback projections numerically match sensory inputs. To unravel information carried by these two streams, we imaged the activity of olfactory sensory neurons (OSNs) and cortical axons in the mouse OB using calcium indicators, multiphoton microscopy, and diverse olfactory stimuli. Here, we show that odorant mixtures of increasing complexity evoke progressively denser OSN activity, yet cortical feedback activity is of similar sparsity for all stimuli. Also, representations of complex mixtures are similar in OSNs but are decorrelated in cortical axons. While OSN responses to increasing odorant concentrations exhibit a sigmoidal relationship, cortical axonal responses are complex and nonmonotonic, which can be explained by a model with activity-dependent feedback inhibition in the cortex. Our study indicates that early-stage olfactory circuits have access to local feedforward signals and global, efficiently formatted information about odor scenes through cortical feedback.

Automated Deep Learning Segmentation of Cardiac Inflammatory FDG PET.

Background: Fluorodeoxyglucose positron emission tomography (FDG PET) with glycolytic metabolism suppression plays a pivotal role in diagnosing cardiac sarcoidosis. Reorientation of images to match perfusion datasets is critical and myocardial segmentation enables consistent image scaling and quantification. However, both are challenging and labor intensive. We developed a 3D U-Net deep learning (DL) algorithm for automated myocardial segmentation in cardiac sarcoidosis FDG PET. Methods: The DL model was trained on 316 patients' FDG PET scans, and left ventricular contours derived from perfusion datasets. Qualitative analysis of clinical readability was performed to compare DL segmentation with the current automated method on a 50-patient test subset. Additionally, left ventricle displacement and angulation, as well as SUVmax sampling were compared to inter-user reproducibility results. Results: DL segmentation enhanced readability scores in over 90% of cases compared to the standard segmentation currently used in the software. DL segmentation performed similarly to a trained technologist, surpassing standard segmentation for left ventricle displacement and angulation, as well as correlation of SUVmax. Conclusion: The DL-based automated segmentation tool presents a marked improvement in the processing of cardiac sarcoidosis FDG PET, promising enhanced clinical workflow. This tool holds significant potential for accelerating clinical practice and improving consistency and quality. Further research with varied datasets is warranted to broaden its applicability.