RESUMO
Magnetic Resonance Imaging (MRI) typically recruits multiple sequences (defined here as "modalities"). As each modality is designed to offer different anatomical and functional clinical information, there are evident disparities in the imaging content across modalities. Inter- and intra-modality affine and non-rigid image registration is an essential medical image analysis process in clinical imaging, as for example before imaging biomarkers need to be derived and clinically evaluated across different MRI modalities, time phases and slices. Although commonly needed in real clinical scenarios, affine and non-rigid image registration is not extensively investigated using a single unsupervised model architecture. In our work, we present an unsupervised deep learning registration methodology that can accurately model affine and non-rigid transformations, simultaneously. Moreover, inverse-consistency is a fundamental inter-modality registration property that is not considered in deep learning registration algorithms. To address inverse consistency, our methodology performs bi-directional cross-modality image synthesis to learn modality-invariant latent representations, and involves two factorised transformation networks (one per each encoder-decoder channel) and an inverse-consistency loss to learn topology-preserving anatomical transformations. Overall, our model (named "FIRE") shows improved performances against the reference standard baseline method (i.e., Symmetric Normalization implemented using the ANTs toolbox) on multi-modality brain 2D and 3D MRI and intra-modality cardiac 4D MRI data experiments. We focus on explaining model-data components to enhance model explainability in medical image registration. On computational time experiments, we show that the FIRE model performs on a memory-saving mode, as it can inherently learn topology-preserving image registration directly in the training phase. We therefore demonstrate an efficient and versatile registration technique that can have merit in multi-modal image registrations in the clinical setting.
Assuntos
Encéfalo , Processamento de Imagem Assistida por Computador , Algoritmos , Encéfalo/diagnóstico por imagem , Coração/diagnóstico por imagem , Processamento de Imagem Assistida por Computador/métodos , Imageamento por Ressonância Magnética/métodosRESUMO
The world is currently experiencing an ongoing pandemic of an infectious disease named coronavirus disease 2019 (i.e., COVID-19), which is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Computed Tomography (CT) plays an important role in assessing the severity of the infection and can also be used to identify those symptomatic and asymptomatic COVID-19 carriers. With a surge of the cumulative number of COVID-19 patients, radiologists are increasingly stressed to examine the CT scans manually. Therefore, an automated 3D CT scan recognition tool is highly in demand since the manual analysis is time-consuming for radiologists and their fatigue can cause possible misjudgment. However, due to various technical specifications of CT scanners located in different hospitals, the appearance of CT images can be significantly different leading to the failure of many automated image recognition approaches. The multi-domain shift problem for the multi-center and multi-scanner studies is therefore nontrivial that is also crucial for a dependable recognition and critical for reproducible and objective diagnosis and prognosis. In this paper, we proposed a COVID-19 CT scan recognition model namely coronavirus information fusion and diagnosis network (CIFD-Net) that can efficiently handle the multi-domain shift problem via a new robust weakly supervised learning paradigm. Our model can resolve the problem of different appearance in CT scan images reliably and efficiently while attaining higher accuracy compared to other state-of-the-art methods.
RESUMO
BACKGROUND: The future risk of myocardial infarction is commonly assessed using cardiovascular risk scores, coronary artery calcium score, or coronary artery stenosis severity. We assessed whether noncalcified low-attenuation plaque burden on coronary CT angiography (CCTA) might be a better predictor of the future risk of myocardial infarction. METHODS: In a post hoc analysis of a multicenter randomized controlled trial of CCTA in patients with stable chest pain, we investigated the association between the future risk of fatal or nonfatal myocardial infarction and low-attenuation plaque burden (% plaque to vessel volume), cardiovascular risk score, coronary artery calcium score or obstructive coronary artery stenoses. RESULTS: In 1769 patients (56% male; 58±10 years) followed up for a median 4.7 (interquartile interval, 4.0-5.7) years, low-attenuation plaque burden correlated weakly with cardiovascular risk score (r=0.34; P<0.001), strongly with coronary artery calcium score (r=0.62; P<0.001), and very strongly with the severity of luminal coronary stenosis (area stenosis, r=0.83; P<0.001). Low-attenuation plaque burden (7.5% [4.8-9.2] versus 4.1% [0-6.8]; P<0.001), coronary artery calcium score (336 [62-1064] versus 19 [0-217] Agatston units; P<0.001), and the presence of obstructive coronary artery disease (54% versus 25%; P<0.001) were all higher in the 41 patients who had fatal or nonfatal myocardial infarction. Low-attenuation plaque burden was the strongest predictor of myocardial infarction (adjusted hazard ratio, 1.60 (95% CI, 1.10-2.34) per doubling; P=0.014), irrespective of cardiovascular risk score, coronary artery calcium score, or coronary artery area stenosis. Patients with low-attenuation plaque burden greater than 4% were nearly 5 times more likely to have subsequent myocardial infarction (hazard ratio, 4.65; 95% CI, 2.06-10.5; P<0.001). CONCLUSIONS: In patients presenting with stable chest pain, low-attenuation plaque burden is the strongest predictor of fatal or nonfatal myocardial infarction. These findings challenge the current perception of the supremacy of current classical risk predictors for myocardial infarction, including stenosis severity. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT01149590.
Assuntos
Angina Estável/etiologia , Angiografia por Tomografia Computadorizada , Angiografia Coronária , Doença da Artéria Coronariana/diagnóstico por imagem , Estenose Coronária/diagnóstico por imagem , Infarto do Miocárdio/etiologia , Placa Aterosclerótica , Calcificação Vascular/diagnóstico por imagem , Idoso , Angina Estável/diagnóstico , Angina Estável/mortalidade , Doença da Artéria Coronariana/complicações , Doença da Artéria Coronariana/mortalidade , Estenose Coronária/complicações , Estenose Coronária/mortalidade , Feminino , Fatores de Risco de Doenças Cardíacas , Humanos , Masculino , Pessoa de Meia-Idade , Infarto do Miocárdio/diagnóstico , Infarto do Miocárdio/mortalidade , Valor Preditivo dos Testes , Prognóstico , Medição de Risco , Escócia , Fatores de Tempo , Calcificação Vascular/complicações , Calcificação Vascular/mortalidadeRESUMO
PURPOSE: Cerebral amyloid angiopathy (CAA) is a cerebral small vessel disease associated with perivascular ß-amyloid deposition. CAA is also associated with strokes due to lobar intracerebral haemorrhage (ICH). 18F-flutemetamol amyloid ligand PET may improve the early detection of CAA. We performed pharmacokinetic modelling using both full (0-30, 90-120 min) and reduced (30 min) 18F-flutemetamol PET-MR acquisitions, to investigate regional cerebral perfusion and amyloid deposition in ICH patients. METHODS: Dynamic18F-flutemetamol PET-MR was performed in a pilot cohort of sixteen ICH participants; eight lobar ICH cases with probable CAA and eight deep ICH patients. A model-based input function (mIF) method was developed for compartmental modelling. mIF 1-tissue (1-TC) and 2-tissue (2-TC) compartmental modelling, reference tissue models and standardized uptake value ratios were assessed in the setting of probable CAA detection. RESULTS: The mIF 1-TC model detected perfusion deficits and 18F-flutemetamol uptake in cases with probable CAA versus deep ICH patients, in both full and reduced PET acquisition time (all P < 0.05). In the reduced PET acquisition, mIF 1-TC modelling reached the highest sensitivity and specificity in detecting perfusion deficits (0.87, 0.77) and 18F-flutemetamol uptake (0.83, 0.71) in cases with probable CAA. Overall, 52 and 48 out of the 64 brain areas with 18F-flutemetamol-determined amyloid deposition showed reduced perfusion for 1-TC and 2-TC models, respectively. CONCLUSION: Pharmacokinetic (1-TC) modelling using a 30 min PET-MR time frame detected impaired haemodynamics and increased amyloid load in probable CAA. Perfusion deficits and amyloid burden co-existed within cases with CAA, demonstrating a distinct imaging pattern which may have merit in elucidating the pathophysiological process of CAA.
Assuntos
Compostos de Anilina/metabolismo , Compostos de Anilina/farmacocinética , Benzotiazóis/metabolismo , Benzotiazóis/farmacocinética , Angiopatia Amiloide Cerebral/metabolismo , Circulação Cerebrovascular , Tomografia por Emissão de Pósitrons/métodos , Idoso , Idoso de 80 Anos ou mais , Encéfalo/metabolismo , Feminino , Humanos , Masculino , Pessoa de Meia-IdadeRESUMO
Cycle-consistent generative adversarial network (CycleGAN) has been widely used for cross-domain medical image synthesis tasks particularly due to its ability to deal with unpaired data. However, most CycleGAN-based synthesis methods cannot achieve good alignment between the synthesized images and data from the source domain, even with additional image alignment losses. This is because the CycleGAN generator network can encode the relative deformations and noises associated to different domains. This can be detrimental for the downstream applications that rely on the synthesized images, such as generating pseudo-CT for PET-MR attenuation correction. In this paper, we present a deformation invariant cycle-consistency model that can filter out these domain-specific deformation. The deformation is globally parameterized by thin-plate-spline (TPS), and locally learned by modified deformable convolutional layers. Robustness to domain-specific deformations has been evaluated through experiments on multi-sequence brain MR data and multi-modality abdominal CT and MR data. Experiment results demonstrated that our method can achieve better alignment between the source and target data while maintaining superior image quality of signal compared to several state-of-the-art CycleGAN-based methods.
RESUMO
Medical imaging is a key component in clinical diagnosis, treatment planning and clinical trial design, accounting for almost 90% of all healthcare data. CNNs achieved performance gains in medical image analysis (MIA) over the last years. CNNs can efficiently model local pixel interactions and be trained on small-scale MI data. Despite their important advances, typical CNN have relatively limited capabilities in modelling "global" pixel interactions, which restricts their generalisation ability to understand out-of-distribution data with different "global" information. The recent progress of Artificial Intelligence gave rise to Transformers, which can learn global relationships from data. However, full Transformer models need to be trained on large-scale data and involve tremendous computational complexity. Attention and Transformer compartments ("Transf/Attention") which can well maintain properties for modelling global relationships, have been proposed as lighter alternatives of full Transformers. Recently, there is an increasing trend to co-pollinate complementary local-global properties from CNN and Transf/Attention architectures, which led to a new era of hybrid models. The past years have witnessed substantial growth in hybrid CNN-Transf/Attention models across diverse MIA problems. In this systematic review, we survey existing hybrid CNN-Transf/Attention models, review and unravel key architectural designs, analyse breakthroughs, and evaluate current and future opportunities as well as challenges. We also introduced an analysis framework on generalisation opportunities of scientific and clinical impact, based on which new data-driven domain generalisation and adaptation methods can be stimulated.
Assuntos
Inteligência Artificial , Processamento de Imagem Assistida por Computador , Simulação por ComputadorRESUMO
Background: Bone marrow adipose tissue (BMAT) represents > 10% fat mass in healthy humans and can be measured by magnetic resonance imaging (MRI) as the bone marrow fat fraction (BMFF). Human MRI studies have identified several diseases associated with BMFF but have been relatively small scale. Population-scale studies therefore have huge potential to reveal BMAT's true clinical relevance. The UK Biobank (UKBB) is undertaking MRI of 100,000 participants, providing the ideal opportunity for such advances. Objective: To establish deep learning for high-throughput multi-site BMFF analysis from UKBB MRI data. Materials and methods: We studied males and females aged 60-69. Bone marrow (BM) segmentation was automated using a new lightweight attention-based 3D U-Net convolutional neural network that improved segmentation of small structures from large volumetric data. Using manual segmentations from 61-64 subjects, the models were trained to segment four BM regions of interest: the spine (thoracic and lumbar vertebrae), femoral head, total hip and femoral diaphysis. Models were tested using a further 10-12 datasets per region and validated using datasets from 729 UKBB participants. BMFF was then quantified and pathophysiological characteristics assessed, including site- and sex-dependent differences and the relationships with age, BMI, bone mineral density, peripheral adiposity, and osteoporosis. Results: Model accuracy matched or exceeded that for conventional U-Nets, yielding Dice scores of 91.2% (spine), 94.5% (femoral head), 91.2% (total hip) and 86.6% (femoral diaphysis). One case of severe scoliosis prevented segmentation of the spine, while one case of Non-Hodgkin Lymphoma prevented segmentation of the spine, femoral head and total hip because of T2 signal depletion; however, successful segmentation was not disrupted by any other pathophysiological variables. The resulting BMFF measurements confirmed expected relationships between BMFF and age, sex and bone density, and identified new site- and sex-specific characteristics. Conclusions: We have established a new deep learning method for accurate segmentation of small structures from large volumetric data, allowing high-throughput multi-site BMFF measurement in the UKBB. Our findings reveal new pathophysiological insights, highlighting the potential of BMFF as a novel clinical biomarker. Applying our method across the full UKBB cohort will help to reveal the impact of BMAT on human health and disease.
RESUMO
Percutaneous coronary intervention (PCI) is a minimally invasive technique for treating vascular diseases. PCI requires precise and real-time visualization and guidance to ensure surgical safety and efficiency. Existing mainstream guiding methods rely on hemodynamic parameters. However, these methods are less intuitive than images and pose some challenges to the decision-making of cardiologists. This paper proposes a novel PCI guiding assistance system by combining a novel vascular segmentation network and a heuristic intervention path planning algorithm, providing cardiologists with clear and visualized information. A dataset of 1077 DSA images from 288 patients is also collected in clinical practice. A Likert Scale is also designed to evaluate system performance in user experiments. Results of user experiments demonstrate that the system can generate satisfactory and reasonable paths for PCI. Our proposed method outperformed the state-of-the-art baselines based on three metrics (Jaccard: 0.4091, F1: 0.5626, Accuracy: 0.9583). The proposed system can effectively assist cardiologists in PCI by providing a clear segmentation of vascular structures and optimal real-time intervention paths, thus demonstrating great potential for robotic PCI autonomy.
RESUMO
BACKGROUND: Primary immunodeficiency (PI) is a group of heterogeneous disorders resulting from immune system defects. Over 70% of PI is undiagnosed, leading to increased mortality, co-morbidity and healthcare costs. Among PI disorders, combined immunodeficiencies (CID) are characterized by complex immune defects. Common variable immunodeficiency (CVID) is among the most common types of PI. In light of available treatments, it is critical to identify adult patients at risk for CID and CVID, before the development of serious morbidity and mortality. METHODS: We developed a deep learning-based method (named "TabMLPNet") to analyze clinical history from nationally representative medical claims from electronic health records (Optum® data, covering all US), evaluated in the setting of identifying CID/CVID in adults. Further, we revealed the most important CID/CVID-associated antecedent phenotype combinations. Four large cohorts were generated: a total of 47,660 PI cases and (1:1 matched) controls. RESULTS: The sensitivity/specificity of TabMLPNet modeling ranges from 0.82-0.88/0.82-0.85 across cohorts. Distinctive combinations of antecedent phenotypes associated with CID/CVID are identified, consisting of respiratory infections/conditions, genetic anomalies, cardiac defects, autoimmune diseases, blood disorders and malignancies, which can possibly be useful to systematize the identification of CID and CVID. CONCLUSIONS: We demonstrated an accurate method in terms of CID and CVID detection evaluated on large-scale medical claims data. Our predictive scheme can potentially lead to the development of new clinical insights and expanded guidelines for identification of adult patients at risk for CID and CVID as well as be used to improve patient outcomes on population level.
Primary immunodeficiencies (PI) are disorders that weaken the immune system, increasing the incident of life-threatening infections, organ damage and the development of cancer and autoimmune diseases. Although PI is estimated to affect 1-2% of the global population, 70-90% of these patients remain undiagnosed. Many patients are diagnosed during adulthood, after other serious diseases have already developed. We developed a computational method to analyze the clinical history from a large group of people with and without PI. We focused on combined (CID) and common variable immunodeficiency (CVID), which are among the least studied and most common PI subtypes, respectively. We could identify people with CID or CVID and combinations of diseases and symptoms which could make it easier to identify CID or CVID. Our method could be used to more readily identify adults at risk of CID or CVID, enabling treatment to start earlier and their long-term health to be improved.
RESUMO
A drug delivery nanosystem of noble bimetallic nanoparticles (NPs) which consists of Au NPs capped with Pt NPs (Au@Pt NPs) is constructed and functionalised with a quinazoline based small molecule (Au@Pt@Q NPs), acting as a theranostic agent against glioblastoma. Two different hydrothermal synthetic procedures for bimetallic Au@Pt NPs are presented and the resulting nanostructures are fully characterised by means of spectroscopic and microscopic methods. The imaging and targeting capacity of the new drug delivery system is assessed through fluorescent optical microscopy and cytotoxicity evaluations. The constructed Au@Pt NPs consist a monodispersed colloidal solution of 25 nm with photoluminescent, fluorescent and X-Ray absorption properties that confirm their diagnostic potential. Haemolysis testing demonstrated that Au@Pt NPs are biocompatible and fluorescent microscopy confirmed their entering the cells. Cytological evaluation of the NPs through MTT assay showed that they do not inhibit the proliferation of control cell line HEK293, whereas they are toxic in U87MG, U251 and D54 glioblastoma cell lines; rendering them selective targeting agents for treating glioblastoma.
Assuntos
Glioblastoma , Nanopartículas Metálicas , Sistemas de Liberação de Medicamentos , Glioblastoma/tratamento farmacológico , Ouro/química , Células HEK293 , Humanos , Nanopartículas Metálicas/química , Platina/químicaRESUMO
AIMS: Coronary artery calcification is a marker of cardiovascular risk, but its association with qualitatively and quantitatively assessed plaque subtypes is unknown. METHODS AND RESULTS: In this post-hoc analysis, computed tomography (CT) images and 5-year clinical outcomes were assessed in SCOT-HEART trial participants. Agatston coronary artery calcium score (CACS) was measured on non-contrast CT and was stratified as zero (0 Agatston units, AU), minimal (1-9 AU), low (10-99 AU), moderate (100-399 AU), high (400-999 AU), and very high (≥1000 AU). Adverse plaques were investigated by qualitative (visual categorization of positive remodelling, low-attenuation plaque, spotty calcification, and napkin ring sign) and quantitative (calcified, non-calcified, low-attenuation, and total plaque burden; Autoplaque) assessments. Of 1769 patients, 36% had a zero, 9% minimal, 20% low, 17% moderate, 10% high, and 8% very high CACS. Amongst patients with a zero CACS, 14% had non-obstructive disease, 2% had obstructive disease, 2% had visually assessed adverse plaques, and 13% had low-attenuation plaque burden >4%. Non-calcified and low-attenuation plaque burden increased between patients with zero, minimal, and low CACS (P < 0.001), but there was no statistically significant difference between those with medium, high, and very high CACS. Myocardial infarction occurred in 41 patients, 10% of whom had zero CACS. CACS >1000 AU and low-attenuation plaque burden were the only predictors of myocardial infarction, independent of obstructive disease, and 10-year cardiovascular risk score. CONCLUSION: In patients with stable chest pain, zero CACS is associated with a good but not perfect prognosis, and CACS cannot rule out obstructive coronary artery disease, non-obstructive plaque, or adverse plaque phenotypes, including low-attenuation plaque.
Assuntos
Doença da Artéria Coronariana , Infarto do Miocárdio , Placa Aterosclerótica , Calcificação Vascular , Cálcio , Angiografia por Tomografia Computadorizada/métodos , Angiografia Coronária/métodos , Doença da Artéria Coronariana/complicações , Doença da Artéria Coronariana/diagnóstico por imagem , Humanos , Infarto do Miocárdio/complicações , Placa Aterosclerótica/complicações , Placa Aterosclerótica/diagnóstico por imagem , Valor Preditivo dos Testes , Medição de Risco , Fatores de Risco , Tomografia Computadorizada por Raios X , Calcificação Vascular/complicações , Calcificação Vascular/diagnóstico por imagemRESUMO
BACKGROUND: Pericoronary adipose tissue (PCAT) attenuation and low-attenuation noncalcified plaque (LAP) burden can both predict outcomes. OBJECTIVES: This study sought to assess the relative and additive values of PCAT attenuation and LAP to predict future risk of myocardial infarction. METHODS: In a post hoc analysis of the multicenter SCOT-HEART (Scottish Computed Tomography of the Heart) trial, the authors investigated the relationships between the future risk of fatal or nonfatal myocardial infarction and PCAT attenuation measured from coronary computed tomography angiography (CTA) using multivariable Cox regression models including plaque burden, obstructive coronary disease, and cardiac risk score (incorporating age, sex, diabetes, smoking, hypertension, hyperlipidemia, and family history). RESULTS: In 1,697 evaluable participants (age: 58 ± 10 years), there were 37 myocardial infarctions after a median follow-up of 4.7 years. Mean PCAT was -76 ± 8 HU and median LAP burden was 4.20% (IQR: 0%-6.86%). PCAT attenuation of the right coronary artery (RCA) was predictive of myocardial infarction (HR: 1.55; P = 0.017, per 1 SD increment) with an optimum threshold of -70.5 HU (HR: 2.45; P = 0.01). In multivariable analysis, adding PCAT-RCA of ≥-70.5 HU to an LAP burden of >4% (the optimum threshold for future myocardial infarction; HR: 4.87; P < 0.0001) led to improved prediction of future myocardial infarction (HR: 11.7; P < 0.0001). LAP burden showed higher area under the curve compared to PCAT attenuation for the prediction of myocardial infarction (AUC = 0.71 [95% CI: 0.62-0.80] vs AUC = 0.64 [95% CI: 0.54-0.74]; P < 0.001), with increased area under the curve when the 2 metrics are combined (AUC = 0.75 [95% CI: 0.65-0.85]; P = 0.037). CONCLUSION: Coronary CTA-defined LAP burden and PCAT attenuation have marked and complementary predictive value for the risk of fatal or nonfatal myocardial infarction.
Assuntos
Doença da Artéria Coronariana , Infarto do Miocárdio , Placa Aterosclerótica , Tecido Adiposo/diagnóstico por imagem , Idoso , Angiografia por Tomografia Computadorizada/métodos , Angiografia Coronária/métodos , Doença da Artéria Coronariana/diagnóstico por imagem , Humanos , Pessoa de Meia-Idade , Infarto do Miocárdio/diagnóstico por imagem , Infarto do Miocárdio/etiologia , Valor Preditivo dos TestesRESUMO
Purpose: To assess the association between nonalcoholic fatty liver disease (NAFLD) and quantitative atherosclerotic plaque at CT. Materials and Methods: In this post hoc analysis of the prospective Scottish Computed Tomography of the HEART trial (November 2010 to September 2014), hepatosteatosis and coronary artery calcium score were measured at noncontrast CT. Presence of stenoses, visually assessed high-risk plaque, and quantitative plaque burden were assessed at coronary CT angiography. Multivariable models were constructed to assess the impact of hepatosteatosis and cardiovascular risk factors on coronary artery disease. Results: Images from 1726 participants (mean age, 58 years ± 9 [SD]; 974 men) were included. Participants with hepatosteatosis (155 of 1726, 9%) had a higher body mass index, more hypertension and diabetes mellitus, and higher cardiovascular risk scores (P < .001 for all) compared with those without hepatosteatosis. They had increased coronary artery calcium scores (median, 43 Agatston units [AU] [interquartile range, 0-273] vs 19 AU [0-225], P = .046), more nonobstructive disease (48% vs 37%, P = .02), and higher low-attenuation plaque burden (5.11% [0-7.16] vs 4.07% [0-6.84], P = .04). However, these associations were not independent of cardiovascular risk factors. Over a median of 4.7 years, there was no evidence of a difference in myocardial infarction between those with and without hepatosteatosis (1.9% vs 2.4%, P = .92). Conclusion: Hepatosteatosis at CT was associated with an increased prevalence of coronary artery disease at CT, but this was not independent of the presence of cardiovascular risk factors.Keywords: CT, Cardiac, Nonalcoholic Fatty Liver Disease, Coronary Artery Disease, Hepatosteatosis, Plaque QuantificationClinical trial registration no. NCT01149590 Supplemental material is available for this article. © RSNA, 2022See also commentary by Abohashem and Blankstein in this issue.
RESUMO
How will my face look when I get older? Or, for a more challenging question: How will my brain look when I get older? To answer this question one must devise (and learn from data) a multivariate auto-regressive function which given an image and a desired target age generates an output image. While collecting data for faces may be easier, collecting longitudinal brain data is not trivial. We propose a deep learning-based method that learns to simulate subject-specific brain ageing trajectories without relying on longitudinal data. Our method synthesises images conditioned on two factors: age (a continuous variable), and status of Alzheimer's Disease (AD, an ordinal variable). With an adversarial formulation we learn the joint distribution of brain appearance, age and AD status, and define reconstruction losses to address the challenging problem of preserving subject identity. We compare with several benchmarks using two widely used datasets. We evaluate the quality and realism of synthesised images using ground-truth longitudinal data and a pre-trained age predictor. We show that, despite the use of cross-sectional data, our model learns patterns of gray matter atrophy in the middle temporal gyrus in patients with AD. To demonstrate generalisation ability, we train on one dataset and evaluate predictions on the other. In conclusion, our model shows an ability to separate age, disease influence and anatomy using only 2D cross-sectional data that should be useful in large studies into neurodegenerative disease, that aim to combine several data sources. To facilitate such future studies by the community at large our code is made available at https://github.com/xiat0616/BrainAgeing.
Assuntos
Doenças Neurodegenerativas , Envelhecimento , Encéfalo/diagnóstico por imagem , Estudos Transversais , Humanos , Imageamento por Ressonância MagnéticaRESUMO
Magnetic resonance (MR) protocols rely on several sequences to assess pathology and organ status properly. Despite advances in image analysis, we tend to treat each sequence, here termed modality, in isolation. Taking advantage of the common information shared between modalities (an organ's anatomy) is beneficial for multi-modality processing and learning. However, we must overcome inherent anatomical misregistrations and disparities in signal intensity across the modalities to obtain this benefit. We present a method that offers improved segmentation accuracy of the modality of interest (over a single input model), by learning to leverage information present in other modalities, even if few (semi-supervised) or no (unsupervised) annotations are available for this specific modality. Core to our method is learning a disentangled decomposition into anatomical and imaging factors. Shared anatomical factors from the different inputs are jointly processed and fused to extract more accurate segmentation masks. Image misregistrations are corrected with a Spatial Transformer Network, which non-linearly aligns the anatomical factors. The imaging factor captures signal intensity characteristics across different modality data and is used for image reconstruction, enabling semi-supervised learning. Temporal and slice pairing between inputs are learned dynamically. We demonstrate applications in Late Gadolinium Enhanced (LGE) and Blood Oxygenation Level Dependent (BOLD) cardiac segmentation, as well as in T2 abdominal segmentation. Code is available at https://github.com/vios-s/multimodal_segmentation.
Assuntos
Processamento de Imagem Assistida por Computador , Aprendizado de Máquina Supervisionado , Coração/diagnóstico por imagem , Imageamento por Ressonância MagnéticaRESUMO
The rapid spread of coronavirus 2019 disease (COVID-19) has manifested a global public health crisis, and chest CT has been proven to be a powerful tool for screening, triage, evaluation and prognosis in COVID-19 patients. However, CT is not only costly but also associated with an increased incidence of cancer, in particular for children. This study will question whether clinical symptoms and laboratory results can predict the CT outcomes for the pediatric patients with positive RT-PCR testing results in order to determine the necessity of CT for such a vulnerable group. Clinical data were collected from 244 consecutive pediatric patients (16 years of age and under) treated at Wuhan Children's Hospital with positive RT-PCR testing, and the chest CT were performed within 3 days of clinical data collection, from January 21 to March 8, 2020. This study was approved by the local ethics committee of Wuhan Children's Hospital. Advanced decision tree based machine learning models were developed for the prediction of CT outcomes. Results have shown that age, lymphocyte, neutrophils, ferritin and C-reactive protein are the most related clinical indicators for predicting CT outcomes for pediatric patients with positive RT-PCR testing. Our decision support system has managed to achieve an AUC of 0.84 with 0.82 accuracy and 0.84 sensitivity for predicting CT outcomes. Our model can effectively predict CT outcomes, and our findings have indicated that the use of CT should be reconsidered for pediatric patients, as it may not be indispensable.
RESUMO
OBJECTIVES: This study was designed to investigate whether coronary computed tomography angiography assessments of coronary plaque might explain differences in the prognosis of men and women presenting with chest pain. BACKGROUND: Important sex differences exist in coronary artery disease. Women presenting with chest pain have different risk factors, symptoms, prevalence of coronary artery disease and prognosis compared to men. METHODS: Within a multicenter randomized controlled trial, we explored sex differences in stenosis, adverse plaque characteristics (positive remodeling, low-attenuation plaque, spotty calcification, or napkin ring sign) and quantitative assessment of total, calcified, noncalcified and low-attenuation plaque burden. RESULTS: Of the 1,769 participants who underwent coronary computed tomography angiography, 772 (43%) were female. Women were more likely to have normal coronary arteries and less likely to have adverse plaque characteristics (p < 0.001 for all). They had lower total, calcified, noncalcified, and low-attenuation plaque burdens (p < 0.001 for all) and were less likely to have a low-attenuation plaque burden >4% (41% vs. 59%; p < 0.001). Over a median follow-up of 4.7 years, myocardial infarction (MI) occurred in 11 women (1.4%) and 30 men (3%). In those who had MI, women had similar total, noncalcified, and low-attenuation plaque burdens as men, but men had higher calcified plaque burden. Low-attenuation plaque burden predicted MI (hazard ratio: 1.60; 95% confidence interval: 1.10 to 2.34; p = 0.015), independent of calcium score, obstructive disease, cardiovascular risk score, and sex. CONCLUSIONS: Women presenting with stable chest pain have less atherosclerotic plaque of all subtypes compared to men and a lower risk of subsequent MI. However, quantitative low-attenuation plaque is as strong a predictor of subsequent MI in women as in men. (Scottish Computed Tomography of the HEART Trial [SCOT-HEART]; NCT01149590).
Assuntos
Doença da Artéria Coronariana , Infarto do Miocárdio , Placa Aterosclerótica , Angiografia por Tomografia Computadorizada , Angiografia Coronária , Doença da Artéria Coronariana/diagnóstico por imagem , Doença da Artéria Coronariana/epidemiologia , Vasos Coronários/diagnóstico por imagem , Feminino , Humanos , Masculino , Infarto do Miocárdio/diagnóstico por imagem , Infarto do Miocárdio/epidemiologia , Valor Preditivo dos Testes , Fatores de RiscoRESUMO
Multiview based learning has generally returned dividends in performance because additional information can be extracted for the representation of the diversity of different views. The advantage of multiview based learning fits the purpose of segmenting cardiac anatomy from multiview echocardiography, which is a non-invasive, low-cost and low-risk imaging modality. Nevertheless, it is still challenging because of limited training data, a poor signal-to-noise ratio of the echocardiographic data, and large variances across views for a joint learning. In addition, for a better interpretation of pathophysiological processes, clinical decision-making and prognosis, such cardiac anatomy segmentation and quantitative analysis of various clinical indices should ideally be performed for the data covering the full cardiac cycle. To tackle these challenges, a multiview recurrent aggregation network (MV-RAN) has been developed for the echocardiographic sequences segmentation with the full cardiac cycle analysis. Experiments have been carried out on multicentre and multi-scanner clinical studies consisting of spatio-temporal (2D + t) datasets. Compared to other state-of-the-art deep learning based methods, the MV-RAN method has achieved significantly superior results (0.92 ± 0.04 Dice scores) for the segmentation of the left ventricle on the independent testing datasets. For the estimation of clinical indices, our MV-RAN method has also demonstrated great promise and will undoubtedly propel forward the understanding of pathophysiological processes, computer-aided diagnosis and personalised prognosis using echocardiography.
Assuntos
Ecocardiografia , Coração , Diagnóstico por Computador , Coração/diagnóstico por imagem , Ventrículos do Coração/diagnóstico por imagem , Razão Sinal-RuídoRESUMO
Aim: To examine the multimodal contrasting ability of gold-dotted magnetic nanoparticles (Au*MNPs) for magnetic resonance (MR), computed tomography (CT) and intravascular ultrasound (IVUS) imaging. Materials & methods: Au*MNPs were prepared by adapting an impregnation method, without using surface capping reagents and characterized (transmission electron microscopy, x-ray diffraction and Fourier-transform infrared spectroscopy) with their in vitro cytotoxicity assessed, followed by imaging assessments. Results: The contrast-enhancing ability of Au*MNPs was shown to be concentration-dependent across MR, CT and IVUS imaging. The Au content of the Au*MNP led to evident increases of the IVUS signal. Conclusion: We demonstrated that Au*MNPs showed concentration-dependent contrast-enhancing ability in MRI and CT imaging, and for the first-time in IVUS imaging due to the Au content. These Au*MNPs are promising toward solidifying tri-modal imaging-based theragnostics.
Assuntos
Ouro , Nanopartículas de Magnetita , Linhagem Celular Tumoral , Humanos , Imageamento por Ressonância Magnética , Nanopartículas Metálicas , Tomografia Computadorizada por Raios X , Ultrassonografia de IntervençãoRESUMO
Patient motion degrades image quality, affecting the quantitative assessment of PET images. This problem affects studies of coronary lesions in which microcalcification processes are targeted. Coronary PET imaging protocols require scans of up to 30 min, introducing the risk of gross patient motion (GPM) during the acquisition. Here, we investigate the feasibility of an automated data-driven method for the detection of GPM during PET acquisition. Methods: Twenty-eight patients with stable coronary disease underwent a 30-min PET acquisition 1 h after the injection of 18F-sodium fluoride (18F-NaF) at 248 ± 10 MBq (mean ± SD) and then a coronary CT angiography scan. An automated data-driven GPM detection technique tracking the center of mass of the count rates for every 200 ms in the PET list-mode data was devised and evaluated. Two patient motion patterns were considered: sudden repositioning (motion of >0.5 mm within 3 s) and general repositioning (motion of >0.3 mm over 15 s or more). After the reconstruction of diastolic images, individual GPM frames with focal coronary uptake were coregistered in 3 dimensions, creating a GPM-compensated (GPMC) image series. Lesion motion was reported for all lesions with focal uptake. Relative differences in SUVmax and target-to-background ratio (TBR) between GPMC and non-GPMC (standard electrocardiogram-gated data) diastolic PET images were compared in 3 separate groups defined by the maximum motion observed in the lesion (<5, 5-10, and >10 mm). Results: A total of 35 18F-NaF-avid lesions were identified in 28 patients. An average of 3.5 ± 1.5 GPM frames were considered for each patient, resulting in an average frame duration of 7 ± 4 (range, 3-21) min. The mean per-patient motion was: 7 ± 3 mm (maximum, 13.7 mm). GPM correction increased SUVmax and TBR in all lesions with greater than 5 mm of motion. In lesions with 5-10 mm of motion (n = 15), SUVmax and TBR increased by 4.6% ± 5.6% (P = 0.02) and 5.8% ± 6.4% (P < 0.002), respectively. In lesions with greater than 10 mm of motion (n = 15), the SUVmax and TBR increased by 5.0% ± 5.3% (P = 0.009) and 11.5% ± 10.1% (P = 0.001), respectively. GPM correction led to the diagnostic reclassification of 3 patients (11%). Conclusion: GPM during coronary 18F-NaF PET imaging is common and may affect quantitative accuracy. Automated retrospective compensation of this motion is feasible and should be considered for coronary PET imaging.