RESUMO
PURPOSE: Patient-to-image registration is a preliminary step required in surgical navigation based on preoperative images. Human intervention and fiducial markers hamper this task as they are time-consuming and introduce potential errors. We aimed to develop a fully automatic 2D registration system for augmented reality in ear surgery. METHODS: CT-scans and corresponding oto-endoscopic videos were collected from 41 patients (58 ears) undergoing ear examination (vestibular schwannoma before surgery, profound hearing loss requiring cochlear implant, suspicion of perilymphatic fistula, contralateral ears in cases of unilateral chronic otitis media). Two to four images were selected from each case. For the training phase, data from patients (75% of the dataset) and 11 cadaveric specimens were used. Tympanic membranes and malleus handles were contoured on both video images and CT-scans by expert surgeons. The algorithm used a U-Net network for detecting the contours of the tympanic membrane and the malleus on both preoperative CT-scans and endoscopic video frames. Then, contours were processed and registered through an iterative closest point algorithm. Validation was performed on 4 cases and testing on 6 cases. Registration error was measured by overlaying both images and measuring the average and Hausdorff distances. RESULTS: The proposed registration method yielded a precision compatible with ear surgery with a 2D mean overlay error of 0.65 ± 0.60 mm for the incus and 0.48 ± 0.32 mm for the round window. The average Hausdorff distance for these 2 targets was 0.98 ± 0.60 mm and 0.78 ± 0.34 mm respectively. An outlier case with higher errors (2.3 mm and 1.5 mm average Hausdorff distance for incus and round window respectively) was observed in relation to a high discrepancy between the projection angle of the reconstructed CT-scan and the video image. The maximum duration for the overall process was 18 s. CONCLUSIONS: A fully automatic 2D registration method based on a convolutional neural network and applied to ear surgery was developed. The method did not rely on any external fiducial markers nor human intervention for landmark recognition. The method was fast and its precision was compatible with ear surgery.
Assuntos
Redes Neurais de Computação , Tomografia Computadorizada por Raios X , Humanos , Tomografia Computadorizada por Raios X/métodos , Osso Temporal/diagnóstico por imagem , Osso Temporal/cirurgia , Realidade Aumentada , Otoscopia/métodos , Feminino , Gravação em Vídeo , Masculino , Otopatias/cirurgia , Otopatias/diagnóstico por imagem , Procedimentos Cirúrgicos Otológicos/métodos , Pessoa de Meia-Idade , Algoritmos , Cirurgia Assistida por Computador/métodos , Adulto , Membrana Timpânica/diagnóstico por imagem , Membrana Timpânica/cirurgia , Martelo/diagnóstico por imagem , Martelo/cirurgia , Endoscopia/métodosRESUMO
BACKGROUND: Heart failure- (HF) and arrhythmia-related complications are the main causes of morbidity and mortality in patients with nonischemic dilated cardiomyopathy (NIDCM). Cardiovascular magnetic resonance (CMR) imaging is a noninvasive tool for risk stratification based on fibrosis assessment. Diffuse interstitial fibrosis in NIDCM may be a limitation for fibrosis assessment through late gadolinium enhancement (LGE), which might be overcome through quantitative T1 and extracellular volume (ECV) assessment. T1 and ECV prognostic value for arrhythmia-related events remain poorly investigated. We asked whether T1 and ECV have a prognostic value in NIDCM patients. METHODS: This prospective multicenter study analyzed 225 patients with NIDCM confirmed by CMR who were followed up for 2 years. CMR evaluation included LGE, native T1 mapping and ECV values. The primary endpoint was the occurrence of a major adverse cardiovascular event (MACE) which was divided in two groups: HF-related events and arrhythmia-related events. Optimal cutoffs for prediction of MACE occurrence were calculated for all CMR quantitative values. RESULTS: Fifty-eight patients (26%) developed a MACE during follow-up, 42 patients (19%) with HF-related events and 16 patients (7%) arrhythmia-related events. T1 Z-score (p = 0.008) and global ECV (p = 0.001) were associated with HF-related events occurrence, in addition to left ventricular ejection fraction (p < 0.001). ECV > 32.1% (optimal cutoff) remained the only CMR independent predictor of HF-related events occurrence (HR 2.15 [1.14-4.07], p = 0.018). In the arrhythmia-related events group, patients had increased native T1 Z-score and ECV values, with both T1 Z-score > 4.2 and ECV > 30.5% (optimal cutoffs) being independent predictors of arrhythmia-related events occurrence (respectively, HR 2.86 [1.06-7.68], p = 0.037 and HR 2.72 [1.01-7.36], p = 0.049). CONCLUSIONS: ECV was the sole independent predictive factor for both HF- and arrhythmia-related events in NIDCM patients. Native T1 was also an independent predictor in arrhythmia-related events occurrence. The addition of ECV and more importantly native T1 in the decision-making algorithm may improve arrhythmia risk stratification in NIDCM patients. Trial registration NCT02352129. Registered 2nd February 2015-Retrospectively registered, https://clinicaltrials.gov/ct2/show/NCT02352129.
Assuntos
Cardiomiopatia Dilatada , Insuficiência Cardíaca , Humanos , Cardiomiopatia Dilatada/patologia , Prognóstico , Volume Sistólico , Miocárdio/patologia , Meios de Contraste , Estudos Prospectivos , Função Ventricular Esquerda , Imagem Cinética por Ressonância Magnética/métodos , Valor Preditivo dos Testes , Gadolínio , Espectroscopia de Ressonância Magnética , FibroseRESUMO
OBJECTIVE: In the management of the aortic aneurysm, 4D flow magnetic resonance Imaging provides valuable information for the computation of new biomarkers using computational fluid dynamics (CFD). However, accurate segmentation of the aorta is required. Thus, our objective is to evaluate the performance of two automatic segmentation methods on the calculation of aortic wall pressure. METHODS: Automatic segmentation of the aorta was performed with methods based on deep learning and multi-atlas using the systolic phase in the 4D flow MRI magnitude image of 36 patients. Using mesh morphing, isotopological meshes were generated, and CFD was performed to calculate the aortic wall pressure. Node-to-node comparisons of the pressure results were made to identify the most robust automatic method respect to the pressures obtained with a manually segmented model. RESULTS: Deep learning approach presented the best segmentation performance with a mean Dice similarity coefficient and a mean Hausdorff distance (HD) equal to 0.92+/- 0.02 and 21.02+/- 24.20 mm, respectively. At the global level HD is affected by the performance in the abdominal aorta. Locally, this distance decreases to 9.41+/- 3.45 and 5.82+/- 6.23 for the ascending and descending thoracic aorta, respectively. Moreover, with respect to the pressures from the manual segmentations, the differences in the pressures computed from deep learning were lower than those computed from multi-atlas method. CONCLUSION: To reduce biases in the calculation of aortic wall pressure, accurate segmentation is needed, particularly in regions with high blood flow velocities. Thus, the deep learning segmen-tation method should be preferred.
Assuntos
Aprendizado Profundo , Humanos , Processamento de Imagem Assistida por Computador/métodos , Imageamento por Ressonância Magnética/métodos , Aorta Abdominal/diagnóstico por imagem , Velocidade do Fluxo SanguíneoRESUMO
Recently, deep learning (DL) models have been increasingly adopted for automatic analyses of medical data, including electrocardiograms (ECGs). Large, available ECG datasets, generally of high quality, often lack specific distortions, which could be helpful for enhancing DL-based algorithms. Synthetic ECG datasets could overcome this limitation. A generative adversarial network (GAN) was used to synthesize realistic 3D magnetohydrodynamic (MHD) distortion templates, as observed during magnetic resonance imaging (MRI), and then added to available ECG recordings to produce an augmented dataset. Similarity metrics, as well as the accuracy of a DL-based R-peak detector trained with and without data augmentation, were used to evaluate the effectiveness of the synthesized data. Three-dimensional MHD distortions produced by the proposed GAN were similar to the measured ones used as input. The precision of a DL-based R-peak detector, tested on actual unseen data, was significantly enhanced by data augmentation; its recall was higher when trained with augmented data. Using synthesized MHD-distorted ECGs significantly improves the accuracy of a DL-based R-peak detector, with a good generalization capacity. This provides a simple and effective alternative to collecting new patient data. DL-based algorithms for ECG analyses can suffer from bias or gaps in training datasets. Using a GAN to synthesize new data, as well as metrics to evaluate its performance, can overcome the scarcity issue of data availability.
Assuntos
Eletrocardiografia , Coração , Humanos , Algoritmos , Benchmarking , Imageamento por Ressonância MagnéticaRESUMO
Infection by the SARS-CoV-2 leading to COVID-19 disease is still rising and techniques to either diagnose or evaluate the disease are still thoroughly investigated. The use of CT as a complementary tool to other biological tests is still under scrutiny as the CT scans are prone to many false positives as other lung diseases display similar characteristics on CT scans. However, fully investigating CT images is of tremendous interest to better understand the disease progression and therefore thousands of scans need to be segmented by radiologists to study infected areas. Over the last year, many deep learning models for segmenting CT-lungs were developed. Unfortunately, the lack of large and shared annotated multicentric datasets led to models that were either under-tested (small dataset) or not properly compared (own metrics, none shared dataset), often leading to poor generalization performance. To address, these issues, we developed a model that uses a multiscale and multilevel feature extraction strategy for COVID19 segmentation and extensively validated it on several datasets to assess its generalization capability for other segmentation tasks on similar organs. The proposed model uses a novel encoder and decoder with a proposed kernel-based atrous spatial pyramid pooling module that is used at the bottom of the model to extract small features with a multistage skip connection concatenation approach. The results proved that our proposed model could be applied on a small-scale dataset and still produce generalizable performances on other segmentation tasks. The proposed model produced an efficient Dice score of 90% on a 100 cases dataset, 95% on the NSCLC dataset, 88.49% on the COVID19 dataset, and 97.33 on the StructSeg 2019 dataset as compared to existing state-of-the-art models. The proposed solution could be used for COVID19 segmentation in clinic applications. The source code is publicly available at https://github.com/RespectKnowledge/Mutiscale-based-Covid-_segmentation-usingDeep-Learning-models.
RESUMO
OBJECTIVE: The aim of our study was to evaluate the impact of aortic root replacement by graft on the elastic properties of the descending thoracic aorta using cardiac magnetic resonance imaging (MRI) and automatic post-processing. MATERIALS AND METHODS: Nineteen patients were operated for an aortic root aneurysm. Cardiac MRI was performed before and after surgery to measure aortic compliance. Images were acquired on a 1.5 T MRI with a conventional aortic MRI protocol plus one additional kinetic sequence orientated perpendicularly to the aorta at the level of pulmonary trunk. The contours of the ascending and descending aortas were detected automatically for each phase with homemade software. RESULTS: Mean time between surgical procedure and earliest post-operative MRI was 18.2 ± 7.1 months. There was no significant difference between the pre- and earliest post-operative mean descending aorta areas and no significant modification in descending aortic compliance after aortic root replacement (1.43 ± 0.84 vs 1.37 ± 0.58 mm2/mmHg, p = 0.47). Pre- and post-operative systolic and diastolic blood pressure were similar. There was a significant decrease in ascending aortic compliance after surgery (2.52 ± 1.24 vs 0.91 ± 0.52 mm2/mmHg; p < 0.0001). DISCUSSION: The aortic root replacement by graft was not associated with changes in elastic properties of the descending aorta at short term. CLINICAL REGISTRATION NUMBER: NCT03817008.
Assuntos
Aorta Torácica , Aorta , Pressão Sanguínea , Humanos , Imageamento por Ressonância MagnéticaRESUMO
BACKGROUND: The definition of left ventricular (LV) non-compaction is controversial, and discriminating between normal and excessive LV trabeculation remains challenging. Our goal was to quantify LV trabeculation on cardiovascular magnetic resonance (CMR) images in a genetic mouse model of non-compaction using a dedicated semi-automatic software package and to compare our results to the histology used as a gold standard. METHODS: Adult mice with ventricular non-compaction were generated by conditional trabecular deletion of Nkx2-5. Thirteen mice (5 controls, 8 Nkx2-5 mutants) were included in the study. Cine CMR series were acquired in the mid LV short axis plane (resolution 0.086 × 0.086x1mm3) (11.75 T). In a sub set of 6 mice, 5 to 7 cine CMR were acquired in LV short axis to cover the whole LV with a lower resolution (0.172 × 0.172x1mm3). We used semi-automatic software to quantify the compacted mass (Mc), the trabeculated mass (Mt) and the percentage of trabeculation (Mt/Mc) on all cine acquisitions. After CMR all hearts were sliced along the short axis and stained with eosin, and histological LV contouring was performed manually, blinded from the CMR results, and Mt, Mc and Mt/Mc were quantified. Intra and interobserver reproducibility was evaluated by computing the intra class correlation coefficient (ICC). RESULTS: Whole heart acquisition showed no statistical significant difference between trabeculation measured at the basal, midventricular and apical parts of the LV. On the mid-LV cine CMR slice, the median Mt was 0.92 mg (range 0.07-2.56 mg), Mc was 12.24 mg (9.58-17.51 mg), Mt/Mc was 6.74% (0.66-17.33%). There was a strong correlation between CMR and the histology for Mt, Mc and Mt/ Mc with respectively: r2 = 0.94 (p < 0.001), r2 = 0.91 (p < 0.001), r2 = 0.83 (p < 0.001). Intra- and interobserver reproducibility was 0.97 and 0.8 for Mt; 0.98 and 0.97 for Mc; 0.96 and 0.72 for Mt/Mc, respectively and significantly more trabeculation was observed in the Mc Mutant mice than the controls. CONCLUSION: The proposed semi-automatic quantification software is accurate in comparison to the histology and reproducible in evaluating Mc, Mt and Mt/ Mc on cine CMR.
Assuntos
Ventrículos do Coração/diagnóstico por imagem , Interpretação de Imagem Assistida por Computador/métodos , Miocárdio Ventricular não Compactado Isolado/diagnóstico por imagem , Imagem Cinética por Ressonância Magnética/métodos , Miocárdio/patologia , Animais , Automação , Biópsia , Modelos Animais de Doenças , Ventrículos do Coração/patologia , Proteína Homeobox Nkx-2.5/deficiência , Proteína Homeobox Nkx-2.5/genética , Miocárdio Ventricular não Compactado Isolado/genética , Miocárdio Ventricular não Compactado Isolado/patologia , Camundongos Knockout , Valor Preditivo dos Testes , Reprodutibilidade dos TestesRESUMO
Predicting aortic aneurysm ruptures is a complex problem that has been investigated by many research teams over several decades. Work on this issue is notably complex and involves both the mechanical behavior of the artery and the blood flow. Magnetic resonance imaging (MRI) can provide measurements concerning the shape of an organ and the blood that flows through it. Measuring local distortion of the artery wall is the first essential factor to evaluate in a ruptured artery. This paper aims to demonstrate the feasibility of this measure using MRI on a phantom of an abdominal aortic aneurysm (AAA) with realistic shape. The aortic geometry is obtained from a series of cine-MR images and reconstructed using Mimics software. From 4D flow and MRI measurements, the field of velocity is determined and introduced into a computational fluid dynamic (CFD) model to determine the mechanical boundaries applied on the wall artery (pressure and ultimately wall shear stress (WSS)). These factors are then converted into a solid model that enables wall deformations to be calculated. This approach was applied to a silicone phantom model of an AAA reconstructed from a patient's computed tomography-scan examination. The calculated deformations were then compared to those obtained in identical conditions by stereovision. The results of both methods were found to be close. Deformations of the studied AAA phantom with complex shape were obtained within a gap of 12% by modeling from MR data.
Assuntos
Aneurisma da Aorta Abdominal/diagnóstico por imagem , Imageamento por Ressonância Magnética/instrumentação , Imagens de Fantasmas , Estresse Mecânico , Idoso , Aneurisma da Aorta Abdominal/fisiopatologia , Fenômenos Biomecânicos , Simulação por Computador , Humanos , Hidrodinâmica , Imageamento Tridimensional , Masculino , Modelos Cardiovasculares , Tomografia Computadorizada por Raios XRESUMO
BACKGROUND: We investigate the use of different denoising filters on low signal-to-noise ratio cardiac images of the rat heart acquired with a birdcage volume coil at 7T. Accuracy and variability of cardiac function parameters were measured from manual segmentation of rat heart images with and without filtering. METHODS: Ten rats were studied using a 7T Varian system. End-diastolic and end-systolic volumes, ejection fraction and left ventricle mass (LVM) were calculated from manual segmentation by two experts on cine-FLASH short-axis slices covering the left ventricle. Series were denoised with an anisotropic diffusion filter, a whole variation regularization or an optimized Rician non-local means (ORNLM) filtering technique. The effect of the different filters was evaluated by the calculation of signal-to-noise (SNR) and contrast-to-noise (CNR) ratios, followed by a study of intra- and inter-expert variability of the measurement of physiological parameters. The calculated LVM was compared to the LVM obtained by weighing the heart ex vivo. RESULTS: The SNR and the CNR increased after application of the different filters. The performance of the ORNLM filter was superior for all the parameters of the cardiac function, as judged from the inter- and intra-observer variabilities. Moreover, this filtering technique resulted in the lowest variability in the LVM evaluation. CONCLUSIONS: In cardiac MRI of rats, filtering is an interesting alternative that yields better contrast between myocardium and surrounding tissues and the ORNLM filter provided the largest improvements.
Assuntos
Coração/diagnóstico por imagem , Imagem Cinética por Ressonância Magnética/métodos , Intensificação de Imagem Radiográfica/instrumentação , Algoritmos , Animais , Feminino , Imagem Cinética por Ressonância Magnética/instrumentação , Imagem Cinética por Ressonância Magnética/veterinária , Masculino , Intensificação de Imagem Radiográfica/métodos , Ratos , Ratos Endogâmicos F344 , Razão Sinal-RuídoRESUMO
PURPOSE: To propose, assess, and validate a semiautomatic method allowing rapid and reproducible measurement of trabeculated and compacted left ventricular (LV) masses from cardiac magnetic resonance imaging (MRI). MATERIALS AND METHODS: We developed a method to automatically detect noncompacted, endocardial, and epicardial contours. Papillary muscles were segmented using semiautomatic thresholding and were included in the compacted mass. Blood was removed from trabeculae using the same threshold tool. Trabeculated, compacted masses and ratio of noncompacted to compacted (NC:C) masses were computed. Preclinical validation was performed on four transgenic mice with hypertrabeculation of the LV (high-resolution cine imaging, 11.75T). Then analysis was performed on normal cine-MRI examinations (steady-state free precession [SSFP] sequences, 1.5T or 3T) obtained from 60 healthy participants (mean age 49 ± 16 years) with 10 men and 10 women for each of the following age groups: [20,39], [40,59], and [60,79]. Interobserver and interexamination segmentation reproducibility was assessed by using Bland-Altman analysis and by computing the correlation coefficient. RESULTS: In normal participants, noncompacted and compacted masses were 6.29 ± 2.03 g/m(2) and 62.17 ± 11.32 g/m(2) , respectively. The NC:C mass ratio was 10.26 ± 3.27%. Correlation between the two observers was from 0.85 for NC:C ratio to 0.99 for end-diastolic volume (P < 10(-5) ). The bias between the two observers was -1.06 ± 1.02 g/m(2) for trabeculated mass, -1.41 ± 2.78 g/m(2) for compacted mass, and -1.51 ± 1.77% for NC:C ratio. CONCLUSION: We propose a semiautomatic method based on region growing, active contours, and thresholding to calculate the NC:C mass ratio. This method is highly reproducible and might help in the diagnosis of LV noncompaction cardiomyopathy. J. Magn. Reson. Imaging 2016;43:1398-1406.
Assuntos
Cardiopatias Congênitas/diagnóstico por imagem , Cardiopatias Congênitas/patologia , Ventrículos do Coração/diagnóstico por imagem , Ventrículos do Coração/patologia , Interpretação de Imagem Assistida por Computador/métodos , Imagem Cinética por Ressonância Magnética/métodos , Reconhecimento Automatizado de Padrão/métodos , Adulto , Idoso , Algoritmos , Animais , Feminino , Humanos , Aumento da Imagem/métodos , Imageamento Tridimensional/métodos , Aprendizado de Máquina , Masculino , Camundongos , Camundongos Transgênicos , Pessoa de Meia-Idade , Reprodutibilidade dos Testes , Sensibilidade e EspecificidadeRESUMO
OBJECTIVE: To segment and classify the different attenuation regions from MRI at the pelvis level using the T 1 and T 2 relaxation times and anatomical knowledge as a first step towards the creation of PET/MR attenuation maps. MATERIALS AND METHODS: Relaxation times were calculated by fitting the pixel-wise intensities of acquired T 1- and T 2-weighted images from eight men with inversion-recovery and multi-echo multi-slice spin-echo sequences. A decision binary tree based on relaxation times was implemented to segment and classify fat, muscle, prostate, and air (within the body). Connected component analysis and an anatomical knowledge-based procedure were implemented to localize the background and bone. RESULTS: Relaxation times at 3 T are reported for fat (T 1 = 385 ms, T 2 = 121 ms), muscle (T 1 = 1295 ms, T 2 = 40 ms), and prostate (T 1 = 1700 ms, T 2 = 80 ms). The relaxation times allowed the segmentation-classification of fat, prostate, muscle, and air, and combined with anatomical knowledge, they allowed classification of bone. The good segmentation-classification of prostate [mean Dice similarity score (mDSC) = 0.70] suggests a viable implementation in oncology and that of fat (mDSC = 0.99), muscle (mDSC = 0.99), and bone (mDSCs = 0.78) advocates for its implementation in PET/MR attenuation correction. CONCLUSION: Our method allows the segmentation and classification of the attenuation-relevant structures required for the generation of the attenuation map of PET/MR systems in prostate imaging: air, background, bone, fat, muscle, and prostate.
Assuntos
Processamento de Imagem Assistida por Computador/métodos , Imageamento por Ressonância Magnética , Próstata/diagnóstico por imagem , Neoplasias da Próstata/diagnóstico por imagem , Algoritmos , Artefatos , Árvores de Decisões , Humanos , Masculino , Músculos/diagnóstico por imagem , Tomografia por Emissão de Pósitrons , Fatores de TempoRESUMO
We have recently described two kindreds presenting thoracic aortic aneurysm and/or aortic dissection (TAAD) and patent ductus arteriosus (PDA) and mapped the disease locus to 16p12.2-p13.13 (ref. 3). We now demonstrate that the disease is caused by mutations in the MYH11 gene affecting the C-terminal coiled-coil region of the smooth muscle myosin heavy chain, a specific contractile protein of smooth muscle cells (SMC). All individuals bearing the heterozygous mutations, even if asymptomatic, showed marked aortic stiffness. Examination of pathological aortas showed large areas of medial degeneration with very low SMC content. Abnormal immunological recognition of SM-MHC and the colocalization of wild-type and mutant rod proteins in SMC, in conjunction with differences in their coimmunoprecipitation capacities, strongly suggest a dominant-negative effect. Human MYH11 gene mutations provide the first example of a direct change in a specific SMC protein leading to an inherited arterial disease.
Assuntos
Aneurisma da Aorta Torácica/genética , Dissecção Aórtica/genética , Permeabilidade do Canal Arterial/genética , Mutação , Cadeias Pesadas de Miosina/genética , Adulto , Sequência de Aminoácidos , Sequência de Bases , Feminino , Humanos , Masculino , Dados de Sequência Molecular , Linhagem , Estrutura Secundária de ProteínaRESUMO
OBJECTIVE: In clinical practice, thyroid nodules are typically visually evaluated by expert physicians using 2D ultrasound images. Based on their assessment, a fine needle aspiration (FNA) may be recommended. However, visually classifying thyroid nodules from ultrasound images may lead to unnecessary fine needle aspirations for patients. The aim of this study is to develop an automatic thyroid ultrasound image classification system to prevent unnecessary FNAs. METHODS: An automatic computer-aided artificial intelligence system is proposed for classifying thyroid nodules using a fine-tuned deep learning model based on the DenseNet architecture, which incorporates an attention module. The dataset comprises 591 thyroid nodule images categorized based on the Bethesda score. Thyroid nodules are classified as either requiring FNA or not. The challenges encountered in this task include managing variability in image quality, addressing the presence of artifacts in ultrasound image datasets, tackling class imbalance, and ensuring model interpretability. We employed techniques such as data augmentation, class weighting, and gradient-weighted class activation maps (Grad-CAM) to enhance model performance and provide insights into decision making. RESULTS: Our approach achieved excellent results with an average accuracy of 0.94, F1-score of 0.93, and sensitivity of 0.96. The use of Grad-CAM gives insights on the decision making and then reinforce the reliability of the binary classification for the end-user perspective. CONCLUSIONS: We propose a deep learning architecture that effectively classifies thyroid nodules as requiring FNA or not from ultrasound images. Despite challenges related to image variability, class imbalance, and interpretability, our method demonstrated a high classification accuracy with minimal false negatives, showing its potential to reduce unnecessary FNAs in clinical settings.
RESUMO
Generative adversarial networks (GANs) have become increasingly powerful, generating mind-blowing photorealistic images that mimic the content of datasets they have been trained to replicate. One recurrent theme in medical imaging, is whether GANs can also be as effective at generating workable medical data, as they are for generating realistic RGB images. In this paper, we perform a multi-GAN and multi-application study, to gauge the benefits of GANs in medical imaging. We tested various GAN architectures, from basic DCGAN to more sophisticated style-based GANs, on three medical imaging modalities and organs, namely: cardiac cine-MRI, liver CT, and RGB retina images. GANs were trained on well-known and widely utilized datasets, from which their FID scores were computed, to measure the visual acuity of their generated images. We further tested their usefulness by measuring the segmentation accuracy of a U-Net trained on these generated images and the original data. The results reveal that GANs are far from being equal, as some are ill-suited for medical imaging applications, while others performed much better. The top-performing GANs are capable of generating realistic-looking medical images by FID standards, that can fool trained experts in a visual Turing test and comply to some metrics. However, segmentation results suggest that no GAN is capable of reproducing the full richness of medical datasets.
RESUMO
Today, image-guided systems play a significant role in improving the outcome of diagnostic and therapeutic interventions. They provide crucial anatomical information during the procedure to decrease the size and the extent of the approach, to reduce intraoperative complications, and to increase accuracy, repeatability, and safety. Image-to-patient registration is the first step in image-guided procedures. It establishes a correspondence between the patient's preoperative imaging and the intraoperative data. When it comes to the head-and-neck region, the presence of many sensitive structures such as the central nervous system or the neurosensory organs requires a millimetric precision. This review allows evaluating the characteristics and the performances of different registration methods in the head-and-neck region used in the operation room from the perspectives of accuracy, invasiveness, and processing times. Our work led to the conclusion that invasive marker-based methods are still considered as the gold standard of image-to-patient registration. The surface-based methods are recommended for faster procedures and applied on the surface tissues especially around the eyes. In the near future, computer vision technology is expected to enhance these systems by reducing human errors and cognitive load in the operating room.
RESUMO
Deep learning-based methods for cardiac MR segmentation have achieved state-of-the-art results. However, these methods can generate incorrect segmentation results which can lead to wrong clinical decisions in the downstream tasks. Automatic and accurate analysis of downstream tasks, such as myocardial tissue characterization, is highly dependent on the quality of the segmentation results. Therefore, it is of paramount importance to use quality control methods to detect the failed segmentations before further analysis. In this work, we propose a fully automatic uncertainty-based quality control framework for T1 mapping and extracellular volume (ECV) analysis. The framework consists of three parts. The first one focuses on segmentation of cardiac structures from a native and post-contrast T1 mapping dataset (n=295) using a Bayesian Swin transformer-based U-Net. In the second part, we propose a novel uncertainty-based quality control (QC) to detect inaccurate segmentation results. The QC method utilizes image-level uncertainty features as input to a random forest-based classifier/regressor to determine the quality of the segmentation outputs. The experimental results from four different types of segmentation results show that the proposed QC method achieves a mean area under the ROC curve (AUC) of 0.927 on binary classification and a mean absolute error (MAE) of 0.021 on Dice score regression, significantly outperforming other state-of-the-art uncertainty based QC methods. The performance gap is notably higher in predicting the segmentation quality from poor-performing models which shows the robustness of our method in detecting failed segmentations. After the inaccurate segmentation results are detected and rejected by the QC method, in the third part, T1 mapping and ECV values are computed automatically to characterize the myocardial tissues of healthy and cardiac pathological cases. The native myocardial T1 and ECV values computed from automatic and manual segmentations show an excellent agreement yielding Pearson coefficients of 0.990 and 0.975 (on the combined validation and test sets), respectively. From the results, we observe that the automatically computed myocardial T1 and ECV values have the ability to characterize myocardial tissues of healthy and cardiac diseases like myocardial infarction, amyloidosis, Tako-Tsubo syndrome, dilated cardiomyopathy, and hypertrophic cardiomyopathy.
Assuntos
Cardiomiopatia Hipertrófica , Miocárdio , Humanos , Incerteza , Teorema de Bayes , Miocárdio/patologia , Coração/diagnóstico por imagem , Cardiomiopatia Hipertrófica/patologia , Imageamento por Ressonância Magnética/métodos , Valor Preditivo dos Testes , Meios de ContrasteRESUMO
This paper presents a novel approach to track objects from 4D-flow MRI data. A salient feature of the proposed method is that it fully exploits the geometrical and dynamical nature of the information provided by this imaging modality. The underlying idea consists in formulating the tracking problem as a data assimilation problem, in which both position and velocity observations are extracted from the 4D-flow MRI data series. Optimal state estimation is then performed in a sequential fashion via Kalman filtering. The capabilities of the method are extensively assessed in a numerical study involving synthetic and clinical data.
Assuntos
Imageamento por Ressonância Magnética , Modelos Cardiovasculares , Velocidade do Fluxo Sanguíneo , Imageamento por Ressonância Magnética/métodos , Imageamento Tridimensional/métodos , Movimento (Física)RESUMO
Introduction: Linear accelerator (linac) incorporating a magnetic resonance (MR) imaging device providing enhanced soft tissue contrast is particularly suited for abdominal radiation therapy. In particular, accurate segmentation for abdominal tumors and organs at risk (OARs) required for the treatment planning is becoming possible. Currently, this segmentation is performed manually by radiation oncologists. This process is very time consuming and subject to inter and intra operator variabilities. In this work, deep learning based automatic segmentation solutions were investigated for abdominal OARs on 0.35 T MR-images. Methods: One hundred and twenty one sets of abdominal MR images and their corresponding ground truth segmentations were collected and used for this work. The OARs of interest included the liver, the kidneys, the spinal cord, the stomach and the duodenum. Several UNet based models have been trained in 2D (the Classical UNet, the ResAttention UNet, the EfficientNet UNet, and the nnUNet). The best model was then trained with a 3D strategy in order to investigate possible improvements. Geometrical metrics such as Dice Similarity Coefficient (DSC), Intersection over Union (IoU), Hausdorff Distance (HD) and analysis of the calculated volumes (thanks to Bland-Altman plot) were performed to evaluate the results. Results: The nnUNet trained in 3D mode achieved the best performance, with DSC scores for the liver, the kidneys, the spinal cord, the stomach, and the duodenum of 0.96 ± 0.01, 0.91 ± 0.02, 0.91 ± 0.01, 0.83 ± 0.10, and 0.69 ± 0.15, respectively. The matching IoU scores were 0.92 ± 0.01, 0.84 ± 0.04, 0.84 ± 0.02, 0.54 ± 0.16 and 0.72 ± 0.13. The corresponding HD scores were 13.0 ± 6.0 mm, 16.0 ± 6.6 mm, 3.3 ± 0.7 mm, 35.0 ± 33.0 mm, and 42.0 ± 24.0 mm. The analysis of the calculated volumes followed the same behavior. Discussion: Although the segmentation results for the duodenum were not optimal, these findings imply a potential clinical application of the 3D nnUNet model for the segmentation of abdominal OARs for images from 0.35 T MR-Linac.
RESUMO
BACKGROUND: 4D flow MRI allows the analysis of hemodynamic changes in the aorta caused by pathologies such as thoracic aortic aneurysms (TAA). For personalized management of TAA, new biomarkers are required to analyze the effect of fluid structure iteration which can be obtained from 4D flow MRI. However, the generation of these biomarkers requires prior 4D segmentation of the aorta. OBJECTIVE: To develop an automatic deep learning model to segment the aorta in 4D from 4D flow MRI. METHODS: Segmentation is addressed with a U-Net based segmentation model that treats each 4D flow MRI frame as an independent sample. Performance is measured with respect to Dice score (DS) and Hausdorff distance (HD). In addition, the maximum and minimum surface areas at the level of the ascending aorta are measured and compared with those obtained from cine-MRI. RESULTS: The segmentation performance was 0.90 ± 0.02 for the DS and the mean HD was 9.58 ± 4.36 mm. A correlation coefficient of r = 0.85 was obtained for the maximum surface and r = 0.86 for the minimum surface between the 4D flow MRI and cine-MRI. CONCLUSION: The proposed automatic approach of 4D aortic segmentation from 4D flow MRI seems to be accurate enough to contribute to the wider use of this imaging technique in the analysis of pathologies such as TAA.
Assuntos
Aneurisma da Aorta Torácica , Aprendizado Profundo , Humanos , Aorta Torácica , Imageamento por Ressonância Magnética/métodos , Aorta , Imagem Cinética por Ressonância Magnética/métodos , Velocidade do Fluxo SanguíneoRESUMO
OBJECTIVE: We propose a procedure for calibrating 4 parameters governing the mechanical boundary conditions (BCs) of a thoracic aorta (TA) model derived from one patient with ascending aortic aneurysm. The BCs reproduce the visco-elastic structural support provided by the soft tissue and the spine and allow for the inclusion of the heart motion effect. METHODS: We first segment the TA from magnetic resonance imaging (MRI) angiography and derive the heart motion by tracking the aortic annulus from cine-MRI. A rigid-wall fluid-dynamic simulation is performed to derive the time-varying wall pressure field. We build the finite element model considering patient-specific material properties and imposing the derived pressure field and the motion at the annulus boundary. The calibration, which involves the zero-pressure state computation, is based on purely structural simulations. After obtaining the vessel boundaries from the cine-MRI sequences, an iterative procedure is performed to minimize the distance between them and the corresponding boundaries derived from the deformed structural model. A strongly-coupled fluid-structure interaction (FSI) analysis is finally performed with the tuned parameters and compared to the purely structural simulation. RESULTS AND CONCLUSION: The calibration with structural simulations allows to reduce maximum and mean distances between image-derived and simulation-derived boundaries from 8.64 mm to 6.37 mm and from 2.24 mm to 1.83 mm, respectively. The maximum root mean square error between the deformed structural and FSI surface meshes is 0.19 mm. This procedure may prove crucial for increasing the model fidelity in replicating the real aortic root kinematics.