RESUMO
The diverse composition of mammalian tissues poses challenges for understanding the cell-cell interactions required for organ homeostasis and how spatial relationships are perturbed during disease. Existing methods such as single-cell genomics, lacking a spatial context, and traditional immunofluorescence, capturing only two to six molecular features, cannot resolve these issues. Imaging technologies have been developed to address these problems, but each possesses limitations that constrain widespread use. Here we report a method that overcomes major impediments to highly multiplex tissue imaging. "Iterative bleaching extends multiplexity" (IBEX) uses an iterative staining and chemical bleaching method to enable high-resolution imaging of >65 parameters in the same tissue section without physical degradation. IBEX can be employed with various types of conventional microscopes and permits use of both commercially available and user-generated antibodies in an "open" system to allow easy adjustment of staining panels based on ongoing marker discovery efforts. We show how IBEX can also be used with amplified staining methods for imaging strongly fixed tissues with limited epitope retention and with oligonucleotide-based staining, allowing potential cross-referencing between flow cytometry, cellular indexing of transcriptomes and epitopes by sequencing, and IBEX analysis of the same tissue. To facilitate data processing, we provide an open-source platform for automated registration of iterative images. IBEX thus represents a technology that can be rapidly integrated into most current laboratory workflows to achieve high-content imaging to reveal the complex cellular landscape of diverse organs and tissues.
Assuntos
Células/metabolismo , Imagem Óptica/métodos , Animais , Corantes Fluorescentes/metabolismo , Humanos , Processamento de Imagem Assistida por Computador , Imunização , Linfonodos/diagnóstico por imagem , Camundongos , Especificidade de Órgãos , FenótipoRESUMO
Modern, supervised machine learning approaches to medical image classification, image segmentation, and object detection usually require many annotated images. As manual annotation is usually labor-intensive and time-consuming, a well-designed software program can aid and expedite the annotation process. Ideally, this program should be configurable for various annotation tasks, enable efficient placement of several types of annotations on an image or a region of an image, attribute annotations to individual annotators, and be able to display Digital Imaging and Communications in Medicine (DICOM)-formatted images. No current open-source software program fulfills these requirements. To fill this gap, we developed DicomAnnotator, a configurable open-source software program for DICOM image annotation. This program fulfills the above requirements and provides user-friendly features to aid the annotation process. In this paper, we present the design and implementation of DicomAnnotator. Using spine image annotation as a test case, our evaluation showed that annotators with various backgrounds can use DicomAnnotator to annotate DICOM images efficiently. DicomAnnotator is freely available at https://github.com/UW-CLEAR-Center/DICOM-Annotator under the GPLv3 license.
Assuntos
Curadoria de Dados , Software , HumanosRESUMO
This paper had published originally without open access, but has since been republished with open access.
RESUMO
Many types of medical and scientific experiments acquire raw data in the form of images. Various forms of image processing and image analysis are used to transform the raw image data into quantitative measures that are the basis of subsequent statistical analysis. In this article we describe the SimpleITK R package. SimpleITK is a simplified interface to the insight segmentation and registration toolkit (ITK). ITK is an open source C++ toolkit that has been actively developed over the past 18 years and is widely used by the medical image analysis community. SimpleITK provides packages for many interpreter environments, including R. Currently, it includes several hundred classes for image analysis including a wide range of image input and output, filtering operations, and higher level components for segmentation and registration. Using SimpleITK, development of complex combinations of image and statistical analysis procedures is feasible. This article includes several examples of computational image analysis tasks implemented using SimpleITK, including spherical marker localization, multi-modal image registration, segmentation evaluation, and cell image analysis.
RESUMO
Modern scientific endeavors increasingly require team collaborations to construct and interpret complex computational workflows. This work describes an image-analysis environment that supports the use of computational tools that facilitate reproducible research and support scientists with varying levels of software development skills. The Jupyter notebook web application is the basis of an environment that enables flexible, well-documented, and reproducible workflows via literate programming. Image-analysis software development is made accessible to scientists with varying levels of programming experience via the use of the SimpleITK toolkit, a simplified interface to the Insight Segmentation and Registration Toolkit. Additional features of the development environment include user friendly data sharing using online data repositories and a testing framework that facilitates code maintenance. SimpleITK provides a large number of examples illustrating educational and research-oriented image analysis workflows for free download from GitHub under an Apache 2.0 license: github.com/InsightSoftwareConsortium/SimpleITK-Notebooks .
Assuntos
Diagnóstico por Imagem/métodos , Processamento de Imagem Assistida por Computador/instrumentação , Processamento de Imagem Assistida por Computador/métodos , Radiologia/educação , Pesquisa , Comportamento Cooperativo , Humanos , Reprodutibilidade dos Testes , Fluxo de TrabalhoRESUMO
According to the 2022 World Health Organization's Global Tuberculosis (TB) report, an estimated 10.6 million people fell ill with TB, and 1.6 million died from the disease in 2021. In addition, 2021 saw a reversal of a decades-long trend of declining TB infections and deaths, with an estimated increase of 4.5% in the number of people who fell ill with TB compared to 2020, and an estimated yearly increase of 450,000 cases of drug resistant TB. Estimating the severity of pulmonary TB using frontal chest X-rays (CXR) can enable better resource allocation in resource constrained settings and monitoring of treatment response, enabling prompt treatment modifications if disease severity does not decrease over time. The Timika score is a clinically used TB severity score based on a CXR reading. This work proposes and evaluates three deep learning-based approaches for predicting the Timika score with varying levels of explainability. The first approach uses two deep learning-based models, one to explicitly detect lesion regions using YOLOV5n and another to predict the presence of cavitation using DenseNet121, which are then utilized in score calculation. The second approach uses a DenseNet121-based regression model to directly predict the affected lung percentage and another to predict cavitation presence using a DenseNet121-based classification model. Finally, the third approach directly predicts the Timika score using a DenseNet121-based regression model. The best performance is achieved by the second approach with a mean absolute error of 13-14% and a Pearson correlation of 0.7-0.84 using three held-out datasets for evaluating generalization.
RESUMO
The world health organization's global tuberculosis (TB) report for 2022 identifies TB, with an estimated 1.6 million, as a leading cause of death. The number of new cases has risen since 2020, particularly the number of new drug-resistant cases, estimated at 450,000 in 2021. This is concerning, as treatment of patients with drug resistant TB is complex and may not always be successful. The NIAID TB Portals program is an international consortium with a primary focus on patient centric data collection and analysis for drug resistant TB. The data includes images, their associated radiological findings, clinical records, and socioeconomic information. This work describes a TB Portals' Chest X-ray based image retrieval system which enables precision medicine. An input image is used to retrieve similar images and the associated patient specific information, thus facilitating inspection of outcomes and treatment regimens from comparable patients. Image similarity is defined using clinically relevant biomarkers: gender, age, body mass index (BMI), and the percentage of lung affected per sextant. The biomarkers are predicted using variations of the DenseNet169 convolutional neural network. A multi-task approach is used to predict gender, age and BMI incorporating transfer learning from an initial training on the NIH Clinical Center CXR dataset to the TB portals dataset. The resulting gender AUC, age and BMI mean absolute errors were 0.9854, 4.03years and 1.67kgm2. For the percentage of sextant affected by lesions the mean absolute errors ranged between 7% to 12% with higher error values in the middle and upper sextants which exhibit more variability than the lower sextants. The retrieval system is currently available from https://rap.tbportals.niaid.nih.gov/find_similar_cxr.
RESUMO
Follicular lymphoma (FL) is a generally incurable malignancy that evolves from developmentally blocked germinal center (GC) B cells. To promote survival and immune escape, tumor B cells undergo significant genetic changes and extensively remodel the lymphoid microenvironment. Dynamic interactions between tumor B cells and the tumor microenvironment (TME) are hypothesized to contribute to the broad spectrum of clinical behaviors observed among FL patients. Despite the urgent need, existing clinical tools do not reliably predict disease behavior. Using a multi-modal strategy, we examined cell-intrinsic and -extrinsic factors governing progression and therapeutic outcomes in FL patients enrolled onto a prospective clinical trial. By leveraging the strengths of each platform, we identify several tumor-specific features and microenvironmental patterns enriched in individuals who experience early relapse, the most high-risk FL patients. These features include stromal desmoplasia and changes to the follicular growth pattern present 20 months before first progression and first relapse.
Assuntos
Linfoma Folicular , Humanos , Linfócitos B , Linfoma Folicular/genética , Multiômica , Estudos Prospectivos , Recidiva , Microambiente Tumoral , Ensaios Clínicos como AssuntoRESUMO
Together with the molecular knowledge of genes and proteins, biological images promise to significantly enhance the scientific understanding of complex cellular systems and to advance predictive and personalized therapeutic products for human health. For this potential to be realized, quality-assured image data must be shared among labs at a global scale to be compared, pooled, and reanalyzed, thus unleashing untold potential beyond the original purpose for which the data was generated. There are two broad sets of requirements to enable image data sharing in the life sciences. One set of requirements is articulated in the companion White Paper entitled "Enabling Global Image Data Sharing in the Life Sciences," which is published in parallel and addresses the need to build the cyberinfrastructure for sharing the digital array data (arXiv:2401.13023 [q-bio.OT], https://doi.org/10.48550/arXiv.2401.13023). In this White Paper, we detail a broad set of requirements, which involves collecting, managing, presenting, and propagating contextual information essential to assess the quality, understand the content, interpret the scientific implications, and reuse image data in the context of the experimental details. We start by providing an overview of the main lessons learned to date through international community activities, which have recently made considerable progress toward generating community standard practices for imaging Quality Control (QC) and metadata. We then provide a clear set of recommendations for amplifying this work. The driving goal is to address remaining challenges, and democratize access to common practices and tools for a spectrum of biomedical researchers, regardless of their expertise, access to resources, and geographical location.
RESUMO
Tuberculosis (TB) drug resistance is a worldwide public health problem. It decreases the likelihood of a positive outcome for the individual patient and increases the likelihood of disease spread. Therefore, early detection of TB drug resistance is crucial for improving outcomes and controlling disease transmission. While drug-sensitive tuberculosis cases are declining worldwide because of effective treatment, the threat of drug-resistant tuberculosis is growing, and the success rate of drug-resistant tuberculosis treatment is only around 60%. The TB Portals program provides a publicly accessible repository of TB case data with an emphasis on collecting drug-resistant cases. The dataset includes multi-modal information such as socioeconomic/geographic data, clinical characteristics, pathogen genomics, and radiological features. The program is an international collaboration whose participants are typically under a substantial burden of drug-resistant tuberculosis, with data collected from standard clinical care provided to the patients. Consequentially, the TB Portals dataset is heterogenous in nature, with data representing multiple treatment centers in different countries and containing cross-domain information. This study presents the challenges and methods used to address them when working with this real-world dataset. Our goal was to evaluate whether combining radiological features derived from a chest X-ray of the host and genomic features from the pathogen can potentially improve the identification of the drug susceptibility type, drug-sensitive (DS-TB) or drug-resistant (DR-TB), and the length of the first successful drug regimen. To perform these studies, significantly imbalanced data needed to be processed, which included a much larger number of DR-TB cases than DS-TB, many more cases with radiological findings than genomic ones, and the sparse high dimensional nature of the genomic information. Three evaluation studies were carried out. First, the DR-TB/DS-TB classification model achieved an average accuracy of 92.4% when using genomic features alone or when combining radiological and genomic features. Second, the regression model for the length of the first successful treatment had a relative error of 53.5% using radiological features, 25.6% using genomic features, and 22.0% using both radiological and genomic features. Finally, the relative error of the third regression model predicting the length of the first treatment using the most common drug combination varied depending on the feature type used. When using radiological features alone, the relative error was 17.8%. For genomic features alone, the relative error increased to 19.9%. The model had a relative error of 19.0% when both radiological and genomic features were combined. Although combining radiological and genomic features did not improve upon the use of genomic features when classifying DR-TB/DS-TB, the combination of the two feature types improved the relative error of the predictive model for the length of the first successful treatment. Furthermore, the regression model trained on radiological features achieved the best performance when predicting the treatment length of the most common drug combination.
RESUMO
Classification of drug-resistant tuberculosis (DR-TB) and drug-sensitive tuberculosis (DS-TB) from chest radiographs remains an open problem. Our previous cross validation performance on publicly available chest X-ray (CXR) data combined with image augmentation, the addition of synthetically generated and publicly available images achieved a performance of 85% AUC with a deep convolutional neural network (CNN). However, when we evaluated the CNN model trained to classify DR-TB and DS-TB on unseen data, significant performance degradation was observed (65% AUC). Hence, in this paper, we investigate the generalizability of our models on images from a held out country's dataset. We explore the extent of the problem and the possible reasons behind the lack of good generalization. A comparison of radiologist-annotated lesion locations in the lung and the trained model's localization of areas of interest, using GradCAM, did not show much overlap. Using the same network architecture, a multi-country classifier was able to identify the country of origin of the X-ray with high accuracy (86%), suggesting that image acquisition differences and the distribution of non-pathological and non-anatomical aspects of the images are affecting the generalization and localization of the drug resistance classification model as well. When CXR images were severely corrupted, the performance on the validation set was still better than 60% AUC. The model overfitted to the data from countries in the cross validation set but did not generalize to the held out country. Finally, we applied a multi-task based approach that uses prior TB lesions location information to guide the classifier network to focus its attention on improving the generalization performance on the held out set from another country to 68% AUC.
RESUMO
BACKGROUND: Tuberculosis (TB) drug resistance is a worldwide public health problem that threatens progress made in TB care and control. Early detection of drug resistance is important for disease control, with discrimination between drug-resistant TB (DR-TB) and drug-sensitive TB (DS-TB) still being an open problem. The objective of this work is to investigate the relevance of readily available clinical data and data derived from chest X-rays (CXRs) in DR-TB prediction and to investigate the possibility of applying machine learning techniques to selected clinical and radiological features for discrimination between DR-TB and DS-TB. We hypothesize that the number of sextants affected by abnormalities such as nodule, cavity, collapse and infiltrate may serve as a radiological feature for DR-TB identification, and that both clinical and radiological features are important factors for machine classification of DR-TB and DS-TB. METHODS: We use data from the NIAID TB Portals program (https://tbportals.niaid.nih.gov), 1,455 DR-TB cases and 782 DS-TB cases from 11 countries. We first select three clinical features and 26 radiological features from the dataset. Then, we perform Pearson's chi-squared test to analyze the significance of the selected clinical and radiological features. Finally, we train machine classifiers based on different features and evaluate their ability to differentiate between DR-TB and DS-TB. RESULTS: Pearson's chi-squared test shows that two clinical features and 23 radiological features are statistically significant regarding DR-TB vs. DS-TB. A ten-fold cross-validation using a support vector machine shows that automatic discrimination between DR-TB and DS-TB achieves an average accuracy of 72.34% and an average AUC value of 78.42%, when combing all 25 statistically significant features. CONCLUSIONS: Our study suggests that the number of affected lung sextants can be used for predicting DR-TB, and that automatic discrimination between DR-TB and DS-TB is possible, with a combination of clinical features and radiological features providing the best performance.
RESUMO
High-content imaging is needed to catalog the variety of cellular phenotypes and multicellular ecosystems present in metazoan tissues. We recently developed iterative bleaching extends multiplexity (IBEX), an iterative immunolabeling and chemical bleaching method that enables multiplexed imaging (>65 parameters) in diverse tissues, including human organs relevant for international consortia efforts. IBEX is compatible with >250 commercially available antibodies and 16 unique fluorophores, and can be easily adopted to different imaging platforms using slides and nonproprietary imaging chambers. The overall protocol consists of iterative cycles of antibody labeling, imaging and chemical bleaching that can be completed at relatively low cost in 2-5 d by biologists with basic laboratory skills. To support widespread adoption, we provide extensive details on tissue processing, curated lists of validated antibodies and tissue-specific panels for multiplex imaging. Furthermore, instructions are included on how to automate the method using competitively priced instruments and reagents. Finally, we present a software solution for image alignment that can be executed by individuals without programming experience using open-source software and freeware. In summary, IBEX is a noncommercial method that can be readily implemented by academic laboratories and scaled to achieve high-content mapping of diverse tissues in support of a Human Reference Atlas or other such applications.
Assuntos
EcossistemaRESUMO
Tuberculosis (TB) is a serious infectious disease that mainly affects the lungs. Drug resistance to the disease makes it more challenging to control. Early diagnosis of drug resistance can help with decision making resulting in appropriate and successful treatment. Chest X-rays (CXRs) have been pivotal to identifying tuberculosis and are widely available. In this work, we utilize CXRs to distinguish between drug-resistant and drug-sensitive tuberculosis. We incorporate Convolutional Neural Network (CNN) based models to discriminate the two types of TB, and employ standard and deep learning based data augmentation methods to improve the classification. Using labeled data from NIAID TB Portals and additional non-labeled sources, we were able to achieve an Area Under the ROC Curve (AUC) of up to 85% using a pretrained InceptionV3 network.
Assuntos
Tuberculose Resistente a Múltiplos Medicamentos , Tuberculose , Área Sob a Curva , Humanos , Redes Neurais de Computação , Radiografia , Tuberculose Resistente a Múltiplos Medicamentos/diagnóstico por imagem , Tuberculose Resistente a Múltiplos Medicamentos/tratamento farmacológicoRESUMO
Light-sheet microscopy has become indispensable for imaging developing organisms, and imaging from multiple directions (views) is essential to improve its spatial resolution. We combine multi-view light-sheet microscopy with microfluidics using adaptive optics (deformable mirror) which corrects aberrations introduced by the 45o-tilted glass coverslip. The optimal shape of the deformable mirror is computed by an iterative algorithm that optimizes the point-spread function in two orthogonal views. Simultaneous correction in two optical arms is achieved via a knife-edge mirror that splits the excitation path and combines the detection paths. Our design allows multi-view light-sheet microscopy with microfluidic devices for precisely controlled experiments and high-content screening.
RESUMO
Boosting immune cell function by targeting the coinhibitory receptor PD-1 may have applications in the treatment of chronic infections. Here, we examine the role of PD-1 during Mycobacterium tuberculosis (Mtb) infection of rhesus macaques. Animals treated with anti-PD-1 monoclonal antibody developed worse disease and higher granuloma bacterial loads compared with isotype control-treated monkeys. PD-1 blockade increased the number and functionality of granuloma Mtb-specific CD8 T cells. In contrast, Mtb-specific CD4 T cells in anti-PD-1-treated macaques were not increased in number or function in granulomas, expressed increased levels of CTLA-4, and exhibited reduced intralesional trafficking in live imaging studies. In granulomas of anti-PD-1-treated animals, multiple proinflammatory cytokines were elevated, and more cytokines correlated with bacterial loads, leading to the identification of a role for caspase 1 in the exacerbation of tuberculosis after PD-1 blockade. Last, increased Mtb bacterial loads after PD-1 blockade were found to associate with the composition of the intestinal microbiota before infection in individual macaques. Therefore, PD-1-mediated coinhibition is required for control of Mtb infection in macaques, perhaps because of its role in dampening detrimental inflammation and allowing for normal CD4 T cell responses.
Assuntos
Linfócitos T CD4-Positivos/efeitos dos fármacos , Inibidores de Checkpoint Imunológico/efeitos adversos , Receptor de Morte Celular Programada 1/antagonistas & inibidores , Tuberculose/tratamento farmacológico , Animais , Carga Bacteriana/efeitos dos fármacos , Linfócitos T CD4-Positivos/imunologia , Linfócitos T CD4-Positivos/metabolismo , Antígeno CTLA-4/metabolismo , Modelos Animais de Doenças , Humanos , Inibidores de Checkpoint Imunológico/administração & dosagem , Macaca mulatta , Masculino , Camundongos , Camundongos Knockout , Mycobacterium tuberculosis/imunologia , Receptor de Morte Celular Programada 1/genética , Receptor de Morte Celular Programada 1/metabolismo , Índice de Gravidade de Doença , Exacerbação dos Sintomas , Tuberculose/diagnóstico , Tuberculose/imunologia , Tuberculose/microbiologiaRESUMO
PURPOSE: C-arm based cone-beam CT (CBCT) has been recently introduced as an in-situ 3D soft tissue imaging modality. When combined with image-guided navigation, it provides a streamlined clinical workflow with, potentially, improved interventional accuracy. A key component in these systems is image to patient registration. The most common registration method relies on fiducial markers placed on the patient's skin. The fiducials are localized in the volumetric image and in the interventional environment. When using C-arm CBCT, the small spatial extent of the volumetric reconstruction makes this registration approach challenging, as the volume must include both the anatomy of interest and the fiducials. The authors have previously proposed a semiautomatic localization approach that addresses this challenge, with evaluation carried out using anthropomorphic phantoms. To truly evaluate the algorithm's utility, the evaluation must be carried out using clinical data. In this article, the authors extend the evaluation of the approach to data sets acquired in a clinical trial. METHODS: Nine CBCT data sets were obtained in three interventional radiology procedures as part of a clinical trial evaluating a commercial navigation system. Fiducials were localized in the volumetric coordinate system directly from the projection images using the evaluated localization approach. Localization was assessed using two quality measures fiducial registration error to quantify precision and fiducial localization error to quantify accuracy. The fiducials used in this study are 6 mm spheres embedded in a custom registration phantom used by the navigation system. RESULTS: In all cases, the proposed approach was able to localize all five fiducial markers embedded in the registration phantom. The approach's mean (std) fiducial registration error was 0.29 (0.13) mm. The mean (std) localization difference as compared to direct volumetric localization was 0.82 (0.34) mm. CONCLUSIONS: Based on the current evaluation using data from clinical cases, the authors conclude that the localization approach is sufficiently accurate for use in thoracic-abdominal interventions, and that it can simplify the current workflow while reducing cumulative radiation to the patient due to repeated CBCT scans.
Assuntos
Tomografia Computadorizada de Feixe Cônico/estatística & dados numéricos , Algoritmos , Fenômenos Biofísicos , Tomografia Computadorizada de Feixe Cônico/instrumentação , Fenômenos Eletromagnéticos , Humanos , Imageamento Tridimensional , Imagens de Fantasmas , Interpretação de Imagem Radiográfica Assistida por ComputadorRESUMO
PURPOSE: C-arm based cone-beam CT (CBCT) imaging enables the in situ acquisition of three dimensional images. In the context of image-guided interventions, this technology potentially reduces the complexity of a procedure's workflow. Instead of acquiring the preoperative volumetric images in a separate location and transferring the patient to the interventional suite, both imaging and intervention are carried out in the same location. A key component in image-guided interventions is image to patient registration. The most common registration approach, in clinical use, is based on fiducial markers placed on the patient's skin which are then localized in the volumetric image and in the interventional environment. When using C-arm CBCT, this registration approach is challenging as in many cases the small size of the volumetric reconstruction cannot include both the skin fiducials and the organ of interest. METHODS: In this article the author shows that fiducial localization outside the reconstructed volume is possible if the projection images from which the reconstruction was obtained are available. By replacing direct fiducial localization in the volumetric images with localization in the projection images, the author obtains the fiducial coordinates in the volume's coordinate system even when the fiducials are outside the reconstructed region. RESULTS: The approach was evaluated using two types of spherical fiducials, clinically used 4 mm diameter markers and a custom phantom embedded with 6 mm diameter markers that is part of a commercial navigation system. In all cases, the method localized all fiducials, including those that were outside the reconstructed volume. The method's mean (std) localization error as evaluated using fiducials that were directly localized in the CBCT reconstruction was 0.55 (0.22) mm for the 4 mm markers and 0.51(0.18) mm for the 6 mm markers. CONCLUSIONS: Based on the evaluations the author concludes that the proposed localization approach is sufficiently accurate to augment or replace direct volumetric fiducial localization for thoracic-abdominal interventions. This allows the physician to position fiducials in a more flexible manner, relaxing the requirement that both the organ of interest and skin surface be contained in the volumetric reconstruction.
Assuntos
Tomografia Computadorizada de Feixe Cônico/métodos , Imageamento Tridimensional/métodos , Algoritmos , Humanos , Processamento de Imagem Assistida por Computador/métodos , Análise dos Mínimos Quadrados , Modelos Biológicos , Imagens de Fantasmas , Fatores de TempoRESUMO
When choosing an electromagnetic tracking system (EMTS) for image-guided procedures several factors must be taken into consideration. Among others these include the system's refresh rate, the number of sensors that need to be tracked, the size of the navigated region, the system interaction with the environment, whether the sensors can be embedded into the tools and provide the desired transformation data, and tracking accuracy and robustness. To date, the only factors that have been studied extensively are the accuracy and the susceptibility of EMTSs to distortions caused by ferromagnetic materials. In this paper the authors shift the focus from analysis of system accuracy and stability to the broader set of factors influencing the utility of EMTS in the clinical environment. The authors provide an analysis based on all of the factors specified above, as assessed in three clinical environments. They evaluate two commercial tracking systems, the Aurora system from Northern Digital Inc., and the 3D Guidance system with three different field generators from Ascension Technology Corp. The authors show that these systems are applicable to specific procedures and specific environments, but that currently, no single system configuration provides a comprehensive solution across procedures and environments.
Assuntos
Fenômenos Eletromagnéticos , Imageamento Tridimensional/instrumentação , Fenômenos Biofísicos , Humanos , Imageamento Tridimensional/estatística & dados numéricos , Imagens de Fantasmas , Pneumologia/instrumentação , Radiologia Intervencionista/instrumentação , Tomografia Computadorizada por Raios X/instrumentaçãoRESUMO
Segmentation of the left ventricle and quantification of various cardiac contractile functions is crucial for the timely diagnosis and treatment of cardiovascular diseases. Traditionally, the two tasks have been tackled independently. Here we propose a convolutional neural network based multi-task learning approach to perform both tasks simultaneously, such that, the network learns better representation of the data with improved generalization performance. Probabilistic formulation of the problem enables learning the task uncertainties during the training, which are used to automatically compute the weights for the tasks. We performed a five fold cross-validation of the myocardium segmentation obtained from the proposed multi-task network on 97 patient 4-dimensional cardiac cine-MRI datasets available through the STA-COM LV segmentation challenge against the provided gold-standard myocardium segmentation, obtaining a Dice overlap of 0.849 ± 0.036 and mean surface distance of 0.274 ± 0.083 mm, while simultaneously estimating the myocardial area with mean absolute difference error of 205 ± 198 mm2.