MRI lung lobe segmentation in pediatric cystic fibrosis patients using a recurrent neural network trained with publicly accessible CT datasets.

PURPOSE: To introduce a widely applicable workflow for pulmonary lobe segmentation of MR images using a recurrent neural network (RNN) trained with chest CT datasets. The feasibility is demonstrated for 2D coronal ultrafast balanced SSFP (ufSSFP) MRI. METHODS: Lung lobes of 250 publicly accessible CT datasets of adults were segmented with an open-source CT-specific algorithm. To match 2D ufSSFP MRI data of pediatric patients, both CT data and segmentations were translated into pseudo-MR images that were masked to suppress anatomy outside the lung. Network-1 was trained with pseudo-MR images and lobe segmentations and then applied to 1000 masked ufSSFP images to predict lobe segmentations. These outputs were directly used as targets to train Network-2 and Network-3 with non-masked ufSSFP data as inputs, as well as an additional whole-lung mask as input for Network-2. Network predictions were compared to reference manual lobe segmentations of ufSSFP data in 20 pediatric cystic fibrosis patients. Manual lobe segmentations were performed by splitting available whole-lung segmentations into lobes. RESULTS: Network-1 was able to segment the lobes of ufSSFP images, and Network-2 and Network-3 further increased segmentation accuracy and robustness. The average all-lobe Dice similarity coefficients were 95.0 ± 2.8 (mean ± pooled SD [%]) and 96.4 ± 2.5, 93.0 ± 2.0; and the average median Hausdorff distances were 6.1 ± 0.9 (mean ± SD [mm]), 5.3 ± 1.1, 7.1 ± 1.3 for Network-1, Network-2, and Network-3, respectively. CONCLUSION: Recurrent neural network lung lobe segmentation of 2D ufSSFP imaging is feasible, in good agreement with manual segmentations. The proposed workflow might provide access to automated lobe segmentations for various lung MRI examinations and quantitative analyses.

The impact of segmentation on whole-lung functional MRI quantification: Repeatability and reproducibility from multiple human observers and an artificial neural network.

PURPOSE: To investigate the repeatability and reproducibility of lung segmentation and their impact on the quantitative outcomes from functional pulmonary MRI. Additionally, to validate an artificial neural network (ANN) to accelerate whole-lung quantification. METHOD: Ten healthy children and 25 children with cystic fibrosis underwent matrix pencil decomposition MRI (MP-MRI). Impaired relative fractional ventilation (RFV ) and relative perfusion (RQ ) from MP-MRI were compared using whole-lung segmentation performed by a physician at two time-points (At1 and At2 ), by an MRI technician (B), and by an ANN (C). Repeatability and reproducibility were assess with Dice similarity coefficient (DSC), paired t-test and Intraclass-correlation coefficient (ICC). RESULTS: The repeatability within an observer (At1 vs At2 ) resulted in a DSC of 0.94 ± 0.01 (mean ± SD) and an unsystematic difference of -0.01% for RFV (P = .92) and +0.1% for RQ (P = .21). The reproducibility between human observers (At1 vs B) resulted in a DSC of 0.88 ± 0.02, and a systematic absolute difference of -0.81% (P < .001) for RFV and -0.38% (P = .037) for RQ . The reproducibility between human and the ANN (At1 vs C) resulted in a DSC of 0.89 ± 0.03 and a systematic absolute difference of -0.36% for RFV (P = .017) and -0.35% for RQ (P = .002). The ICC was >0.98 for all variables and comparisons. CONCLUSIONS: Despite high overall agreement, there were systematic differences in lung segmentation between observers. This needs to be considered for longitudinal studies and could be overcome by using an ANN, which performs as good as human observers and fully automatizes MP-MRI post-processing.

Defect distribution index: A novel metric for functional lung MRI in cystic fibrosis.

PURPOSE: Lung impairment from functional MRI is frequently assessed as defect percentage. The defect distribution, however, is currently not quantified. The purpose of this work was to develop a novel measure that quantifies how clustered or scattered defects in functional lung MRI appear, and to evaluate it in pediatric cystic fibrosis. THEORY: The defect distribution index (DDI) calculates a score for each lung voxel categorized as defected. The index increases according to how densely and how far an expanding circle around a defect voxel contains more than 50% defect voxels. METHODS: Fractional ventilation and perfusion maps of 53 children with cystic fibrosis were previously acquired with matrix pencil decomposition MRI. In this work, the DDI is compared to a visual score of 3 raters who evaluated how clustered the lung defects appear. Further, spearman correlations between DDI and lung function parameters were determined. RESULTS: The DDI strongly correlates with the visual scoring (r = 0.90 for ventilation; r = 0.88 for perfusion; P < .0001). Although correlations between DDI and defect percentage are moderate to strong (r = 0.61 for ventilation; r = 0.75 for perfusion; P < .0001), the DDI distinguishes between patients with comparable defect percentage. CONCLUSION: The DDI is a novel measure for functional lung MRI. It provides complementary information to the defect percentage because the DDI assesses defect distribution rather than defect size. The DDI is applicable to matrix pencil MRI data of cystic fibrosis patients and shows very good agreement with human perception of defect distributions.

Augmented reality for sentinel lymph node biopsy.

INTRODUCTION: Sentinel lymph node biopsy for oral and oropharyngeal squamous cell carcinoma is a well-established staging method. One variation is to inject a radioactive tracer near the primary tumor of the patient. After a few minutes, audio feedback from an external hand-held [Formula: see text]-detection probe can monitor the uptake into the lymphatic system. Such probes place a high cognitive load on the surgeon during the biopsy, as they require the simultaneous use of both hands and the skills necessary to correlate the audio signal with the location of tracer accumulation in the lymph nodes. Therefore, an augmented reality (AR) approach to directly visualize and thus discriminate nearby lymph nodes would greatly reduce the surgeons' cognitive load. MATERIALS AND METHODS: We present a proof of concept of an AR approach for sentinel lymph node biopsy by ex vivo experiments. The 3D position of the radioactive [Formula: see text]-sources is reconstructed from a single [Formula: see text]-image, acquired by a stationary table-attached multi-pinhole [Formula: see text]-detector. The position of the sources is then visualized using Microsoft's HoloLens. We further investigate the performance of our SLNF algorithm for a single source, two sources, and two sources with a hot background. RESULTS: In our ex vivo experiments, a single [Formula: see text]-source and its AR representation show good correlation with known locations, with a maximum error of 4.47 mm. The SLNF algorithm performs well when only one source is reconstructed, with a maximum error of 7.77 mm. For the more challenging case to reconstruct two sources, the errors vary between 2.23 mm and 75.92 mm. CONCLUSION: This proof of concept shows promising results in reconstructing and displaying one [Formula: see text]-source. Two simultaneously recorded sources are more challenging and require further algorithmic optimization.

Time-Resolved Dynamic Optical Coherence Tomography for Retinal Blood Flow Analysis.

Purpose: Optical coherence tomography (OCT) representations in clinical practice are static and do not allow for a dynamic visualization and quantification of blood flow. This study aims to present a method to analyze retinal blood flow dynamics using time-resolved structural OCT. Methods: We developed novel imaging protocols to acquire video-rate time-resolved OCT B-scans (1024 × 496 pixels, 10 degrees field of view) at four different sensor integration times (integration time of 44.8 µs at a nominal A-scan rate of 20 kHz, 22.4 µs at 40 kHz, 11.2 µs at 85 kHz, and 7.24 µs at 125 kHz). The vessel centers were manually annotated for each B-scan and surrounding subvolumes were extracted. We used a velocity model based on signal-to-noise ratio (SNR) drops due to fringe washout to calculate blood flow velocity profiles in vessels within five optic disc diameters of the optic disc rim. Results: Time-resolved dynamic structural OCT revealed pulsatile SNR changes in the analyzed vessels and allowed the calculation of potential blood flow velocities at all integration times. Fringe washout was stronger in acquisitions with longer integration times; however, the ratio of the average SNR to the peak SNR inside the vessel was similar across all integration times. Conclusions: We demonstrated the feasibility of estimating blood flow profiles based on fringe washout analysis, showing pulsatile dynamics in vessels close to the optic nerve head using structural OCT. Time-resolved dynamic OCT has the potential to uncover valuable blood flow information in clinical settings.

An augmented reality-based method to assess precordial electrocardiogram leads: a feasibility trial.

Aims: It has been demonstrated that several cardiac pathologies, including myocardial ischaemia, can be detected using smartwatch electrocardiograms (ECGs). Correct placement of bipolar chest leads remains a major challenge in the outpatient population. Methods and results: In this feasibility trial, we propose an augmented reality-based smartphone app that guides the user to place the smartwatch in predefined positions on the chest using the front camera of a smartphone. A machine-learning model using MobileNet_v2 as the backbone was trained to detect the bipolar lead positions V1-V6 and visually project them onto the user's chest. Following the smartwatch recordings, a conventional 10 s, 12-lead ECG was recorded for comparison purposes. All 50 patients participating in the study were able to conduct a 9-lead smartwatch ECG using the app and assistance from the study team. Twelve patients were able to record all the limb and chest leads using the app without additional support. Bipolar chest leads recorded with smartwatch ECGs were assigned to standard unipolar Wilson leads by blinded cardiologists based on visual characteristics. In every lead, at least 86% of the ECGs were assigned correctly, indicating the remarkable similarity of the smartwatch to standard ECG recordings. Conclusion: We have introduced an augmented reality-based method to independently record multichannel smartwatch ECGs in an outpatient setting.

Feasibility of Automated Segmentation of Pigmented Choroidal Lesions in OCT Data With Deep Learning.

Purpose: To evaluate the feasibility of automated segmentation of pigmented choroidal lesions (PCLs) in optical coherence tomography (OCT) data and compare the performance of different deep neural networks. Methods: Swept-source OCT image volumes were annotated pixel-wise for PCLs and background. Three deep neural network architectures were applied to the data: the multi-dimensional gated recurrent units (MD-GRU), the V-Net, and the nnU-Net. The nnU-Net was used to compare the performance of two-dimensional (2D) versus three-dimensional (3D) predictions. Results: A total of 121 OCT volumes were analyzed (100 normal and 21 PCLs). Automated PCL segmentations were successful with all neural networks. The 3D nnU-Net predictions showed the highest recall with a mean of 0.77 ± 0.22 (MD-GRU, 0.60 ± 0.31; V-Net, 0.61 ± 0.25). The 3D nnU-Net predicted PCLs with a Dice coefficient of 0.78 ± 0.13, outperforming MD-GRU (0.62 ± 0.23) and V-Net (0.59 ± 0.24). The smallest distance to the manual annotation was found using 3D nnU-Net with a mean maximum Hausdorff distance of 315 ± 172 µm (MD-GRU, 1542 ± 1169 µm; V-Net, 2408 ± 1060 µm). The 3D nnU-Net showed a superior performance compared with stacked 2D predictions. Conclusions: The feasibility of automated deep learning segmentation of PCLs was demonstrated in OCT data. The neural network architecture had a relevant impact on PCL predictions. Translational Relevance: This work serves as proof of concept for segmentations of choroidal pathologies in volumetric OCT data; improvements are conceivable to meet clinical demands for the diagnosis, monitoring, and treatment evaluation of PCLs.

GAMER-MRI in Multiple Sclerosis Identifies the Diffusion-Based Microstructural Measures That Are Most Sensitive to Focal Damage: A Deep-Learning-Based Analysis and Clinico-Biological Validation.

Conventional magnetic resonance imaging (cMRI) in multiple sclerosis (MS) patients provides measures of focal brain damage and activity, which are fundamental for disease diagnosis, prognosis, and the evaluation of response to therapy. However, cMRI is insensitive to the damage to the microenvironment of the brain tissue and the heterogeneity of MS lesions. In contrast, the damaged tissue can be characterized by mathematical models on multishell diffusion imaging data, which measure different compartmental water diffusion. In this work, we obtained 12 diffusion measures from eight diffusion models, and we applied a deep-learning attention-based convolutional neural network (CNN) (GAMER-MRI) to select the most discriminating measures in the classification of MS lesions and the perilesional tissue by attention weights. Furthermore, we provided clinical and biological validation of the chosen metrics-and of their most discriminative combinations-by correlating their respective mean values in MS patients with the corresponding Expanded Disability Status Scale (EDSS) and the serum level of neurofilament light chain (sNfL), which are measures of disability and neuroaxonal damage. Our results show that the neurite density index from neurite orientation and dispersion density imaging (NODDI), the measures of the intra-axonal and isotropic compartments from microstructural Bayesian approach, and the measure of the intra-axonal compartment from the spherical mean technique NODDI were the most discriminating (respective attention weights were 0.12, 0.12, 0.15, and 0.13). In addition, the combination of the neurite density index from NODDI and the measures for the intra-axonal and isotropic compartments from the microstructural Bayesian approach exhibited a stronger correlation with EDSS and sNfL than the individual measures. This work demonstrates that the proposed method might be useful to select the microstructural measures that are most discriminative of focal tissue damage and that may also be combined to a unique contrast to achieve stronger correlations to clinical disability and neuroaxonal damage.

GAMER MRI: Gated-attention mechanism ranking of multi-contrast MRI in brain pathology.

INTRODUCTION: During the last decade, a multitude of novel quantitative and semiquantitative MRI techniques have provided new information about the pathophysiology of neurological diseases. Yet, selection of the most relevant contrasts for a given pathology remains challenging. In this work, we developed and validated a method, Gated-Attention MEchanism Ranking of multi-contrast MRI in brain pathology (GAMER MRI), to rank the relative importance of MR measures in the classification of well understood ischemic stroke lesions. Subsequently, we applied this method to the classification of multiple sclerosis (MS) lesions, where the relative importance of MR measures is less understood. METHODS: GAMER MRI was developed based on the gated attention mechanism, which computes attention weights (AWs) as proxies of importance of hidden features in the classification. In the first two experiments, we used Trace-weighted (Trace), apparent diffusion coefficient (ADC), Fluid-Attenuated Inversion Recovery (FLAIR), and T1-weighted (T1w) images acquired in 904 acute/subacute ischemic stroke patients and in 6,230 healthy controls and patients with other brain pathologies to assess if GAMER MRI could produce clinically meaningful importance orders in two different classification scenarios. In the first experiment, GAMER MRI with a pretrained convolutional neural network (CNN) was used in conjunction with Trace, ADC, and FLAIR to distinguish patients with ischemic stroke from those with other pathologies and healthy controls. In the second experiment, GAMER MRI with a patch-based CNN used Trace, ADC and T1w to differentiate acute ischemic stroke lesions from healthy tissue. The last experiment explored the performance of patch-based CNN with GAMER MRI in ranking the importance of quantitative MRI measures to distinguish two groups of lesions with different pathological characteristics and unknown quantitative MR features. Specifically, GAMER MRI was applied to assess the relative importance of the myelin water fraction (MWF), quantitative susceptibility mapping (QSM), T1 relaxometry map (qT1), and neurite density index (NDI) in distinguishing 750 juxtacortical lesions from 242 periventricular lesions in 47 MS patients. Pair-wise permutation t-tests were used to evaluate the differences between the AWs obtained for each quantitative measure. RESULTS: In the first experiment, we achieved a mean test AUC of 0.881 and the obtained AWs of FLAIR and the sum of AWs of Trace and ADC were 0.11 and 0.89, respectively, as expected based on previous knowledge. In the second experiment, we achieved a mean test F1 score of 0.895 and a mean AW of Trace = 0.49, of ADC = 0.28, and of T1w = 0.23, thereby confirming the findings of the first experiment. In the third experiment, MS lesion classification achieved test balanced accuracy = 0.777, sensitivity = 0.739, and specificity = 0.814. The mean AWs of T1map, MWF, NDI, and QSM were 0.29, 0.26, 0.24, and 0.22 (p < 0.001), respectively. CONCLUSIONS: This work demonstrates that the proposed GAMER MRI might be a useful method to assess the relative importance of MRI measures in neurological diseases with focal pathology. Moreover, the obtained AWs may in fact help to choose the best combination of MR contrasts for a specific classification problem.