A medical device-grade T1 and ECV phantom for global T1 mapping quality assurance-the T1 Mapping and ECV Standardization in cardiovascular magnetic resonance (T1MES) program.

BACKGROUND: T1 mapping and extracellular volume (ECV) have the potential to guide patient care and serve as surrogate end-points in clinical trials, but measurements differ between cardiovascular magnetic resonance (CMR) scanners and pulse sequences. To help deliver T1 mapping to global clinical care, we developed a phantom-based quality assurance (QA) system for verification of measurement stability over time at individual sites, with further aims of generalization of results across sites, vendor systems, software versions and imaging sequences. We thus created T1MES: The T1 Mapping and ECV Standardization Program. METHODS: A design collaboration consisting of a specialist MRI small-medium enterprise, clinicians, physicists and national metrology institutes was formed. A phantom was designed covering clinically relevant ranges of T1 and T2 in blood and myocardium, pre and post-contrast, for 1.5 T and 3 T. Reproducible mass manufacture was established. The device received regulatory clearance by the Food and Drug Administration (FDA) and Conformité Européene (CE) marking. RESULTS: The T1MES phantom is an agarose gel-based phantom using nickel chloride as the paramagnetic relaxation modifier. It was reproducibly specified and mass-produced with a rigorously repeatable process. Each phantom contains nine differently-doped agarose gel tubes embedded in a gel/beads matrix. Phantoms were free of air bubbles and susceptibility artifacts at both field strengths and T1 maps were free from off-resonance artifacts. The incorporation of high-density polyethylene beads in the main gel fill was effective at flattening the B 1 field. T1 and T2 values measured in T1MES showed coefficients of variation of 1 % or less between repeat scans indicating good short-term reproducibility. Temperature dependency experiments confirmed that over the range 15-30 °C the short-T1 tubes were more stable with temperature than the long-T1 tubes. A batch of 69 phantoms was mass-produced with random sampling of ten of these showing coefficients of variations for T1 of 0.64 ± 0.45 % and 0.49 ± 0.34 % at 1.5 T and 3 T respectively. CONCLUSION: The T1MES program has developed a T1 mapping phantom to CE/FDA manufacturing standards. An initial 69 phantoms with a multi-vendor user manual are now being scanned fortnightly in centers worldwide. Future results will explore T1 mapping sequences, platform performance, stability and the potential for standardization.

Deep Learning for Detection and Localization of B-Lines in Lung Ultrasound.

Lung ultrasound (LUS) is an important imaging modality used by emergency physicians to assess pulmonary congestion at the patient bedside. B-line artifacts in LUS videos are key findings associated with pulmonary congestion. Not only can the interpretation of LUS be challenging for novice operators, but visual quantification of B-lines remains subject to observer variability. In this work, we investigate the strengths and weaknesses of multiple deep learning approaches for automated B-line detection and localization in LUS videos. We curate and publish, BEDLUS, a new ultrasound dataset comprising 1,419 videos from 113 patients with a total of 15,755 expert-annotated B-lines. Based on this dataset, we present a benchmark of established deep learning methods applied to the task of B-line detection. To pave the way for interpretable quantification of B-lines, we propose a novel "single-point" approach to B-line localization using only the point of origin. Our results show that (a) the area under the receiver operating characteristic curve ranges from 0.864 to 0.955 for the benchmarked detection methods, (b) within this range, the best performance is achieved by models that leverage multiple successive frames as input, and (c) the proposed single-point approach for B-line localization reaches an F 1-score of 0.65, performing on par with the inter-observer agreement. The dataset and developed methods can facilitate further biomedical research on automated interpretation of lung ultrasound with the potential to expand the clinical utility.

Automatic evaluation of graft orientation during Descemet membrane endothelial keratoplasty using intraoperative OCT.

Correct Descemet Membrane Endothelial Keratoplasty (DMEK) graft orientation is imperative for success of DMEK surgery, but intraoperative evaluation can be challenging. We present a method for automatic evaluation of the graft orientation in intraoperative optical coherence tomography (iOCT), exploiting the natural rolling behavior of the graft. The method encompasses a deep learning model for graft segmentation, post-processing to obtain a smooth line representation, and curvature calculations to determine graft orientation. For an independent test set of 100 iOCT-frames, the automatic method correctly identified graft orientation in 78 frames and obtained an area under the receiver operating characteristic curve (AUC) of 0.84. When we replaced the automatic segmentation with the manual masks, the AUC increased to 0.92, corresponding to an accuracy of 86%. In comparison, two corneal specialists correctly identified graft orientation in 90% and 91% of the iOCT-frames.

Knowledge distillation with ensembles of convolutional neural networks for medical image segmentation.

Purpose: Ensembles of convolutional neural networks (CNNs) often outperform a single CNN in medical image segmentation tasks, but inference is computationally more expensive and makes ensembles unattractive for some applications. We compared the performance of differently constructed ensembles with the performance of CNNs derived from these ensembles using knowledge distillation, a technique for reducing the footprint of large models such as ensembles. Approach: We investigated two different types of ensembles, namely, diverse ensembles of networks with three different architectures and two different loss-functions, and uniform ensembles of networks with the same architecture but initialized with different random seeds. For each ensemble, additionally, a single student network was trained to mimic the class probabilities predicted by the teacher model, the ensemble. We evaluated the performance of each network, the ensembles, and the corresponding distilled networks across three different publicly available datasets. These included chest computed tomography scans with four annotated organs of interest, brain magnetic resonance imaging (MRI) with six annotated brain structures, and cardiac cine-MRI with three annotated heart structures. Results: Both uniform and diverse ensembles obtained better results than any of the individual networks in the ensemble. Furthermore, applying knowledge distillation resulted in a single network that was smaller and faster without compromising performance compared with the ensemble it learned from. The distilled networks significantly outperformed the same network trained with reference segmentation instead of knowledge distillation. Conclusion: Knowledge distillation can compress segmentation ensembles of uniform or diverse composition into a single CNN while maintaining the performance of the ensemble.

Comparing the signal enhancement of a gadolinium based and an iron-oxide based contrast agent in low-field MRI.

Recently, there has been a renewed interest in low-field MRI. Contrast agents (CA) in MRI have magnetic behavior dependent on magnetic field strength. Therefore, the optimal contrast agent for low-field MRI might be different from what is used at higher fields. Ultra-small superparamagnetic iron-oxides (USPIOs), commonly used as negative CA, might also be used for generating positive contrast in low-field MRI. The purpose of this study was to determine whether an USPIO or a gadolinium based contrast agent is more appropriate at low field strengths. Relaxivity values of ferumoxytol (USPIO) and gadoterate (gadolinium based) were used in this research to simulate normalized signal intensity (SI) curves within a concentration range of 0-15 mM. Simulations were experimentally validated on a 0.25T MRI scanner. Simulations and experiments were performed using spin echo (SE), spoiled gradient echo (SGE), and balanced steady-state free precession (bSSFP) sequences. Maximum achievable SIs were assessed for both CAs in a range of concentrations on all sequences. Simulations at 0.25T showed a peak in SIs at low concentrations ferumoxytol versus a wide top at higher concentrations for gadoterate in SE and SGE. Experiments agreed well with the simulations in SE and SGE, but less in the bSSFP sequence due to overestimated relaxivities in simulations. At low magnetic field strengths, ferumoxytol generates similar signal enhancement at lower concentrations than gadoterate.

Corneal pachymetry by AS-OCT after Descemet's membrane endothelial keratoplasty.

Corneal thickness (pachymetry) maps can be used to monitor restoration of corneal endothelial function, for example after Descemet's membrane endothelial keratoplasty (DMEK). Automated delineation of the corneal interfaces in anterior segment optical coherence tomography (AS-OCT) can be challenging for corneas that are irregularly shaped due to pathology, or as a consequence of surgery, leading to incorrect thickness measurements. In this research, deep learning is used to automatically delineate the corneal interfaces and measure corneal thickness with high accuracy in post-DMEK AS-OCT B-scans. Three different deep learning strategies were developed based on 960 B-scans from 50 patients. On an independent test set of 320 B-scans, corneal thickness could be measured with an error of 13.98 to 15.50 µm for the central 9 mm range, which is less than 3% of the average corneal thickness. The accurate thickness measurements were used to construct detailed pachymetry maps. Moreover, follow-up scans could be registered based on anatomical landmarks to obtain differential pachymetry maps. These maps may enable a more comprehensive understanding of the restoration of the endothelial function after DMEK, where thickness often varies throughout different regions of the cornea, and subsequently contribute to a standardized postoperative regime.

Quantifying Graft Detachment after Descemet's Membrane Endothelial Keratoplasty with Deep Convolutional Neural Networks.

Purpose: We developed a method to automatically locate and quantify graft detachment after Descemet's membrane endothelial keratoplasty (DMEK) in anterior segment optical coherence tomography (AS-OCT) scans. Methods: A total of 1280 AS-OCT B-scans were annotated by a DMEK expert. Using the annotations, a deep learning pipeline was developed to localize scleral spur, center the AS-OCT B-scans and segment the detached graft sections. Detachment segmentation model performance was evaluated per B-scan by comparing (1) length of detachment and (2) horizontal projection of the detached sections with the expert annotations. Horizontal projections were used to construct graft detachment maps. All final evaluations were done on a test set that was set apart during training of the models. A second DMEK expert annotated the test set to determine interrater performance. Results: Mean scleral spur localization error was 0.155 mm, whereas the interrater difference was 0.090 mm. The estimated graft detachment lengths were in 69% of the cases within a 10-pixel (∼150 µm) difference from the ground truth (77% for the second DMEK expert). Dice scores for the horizontal projections of all B-scans with detachments were 0.896 and 0.880 for our model and the second DMEK expert, respectively. Conclusions: Our deep learning model can be used to automatically and instantly localize graft detachment in AS-OCT B-scans. Horizontal detachment projections can be determined with the same accuracy as a human DMEK expert, allowing for the construction of accurate graft detachment maps. Translational Relevance: Automated localization and quantification of graft detachment can support DMEK research and standardize clinical decision-making.

Automatic Needle Segmentation and Localization in MRI With 3-D Convolutional Neural Networks: Application to MRI-Targeted Prostate Biopsy.

Image guidance improves tissue sampling during biopsy by allowing the physician to visualize the tip and trajectory of the biopsy needle relative to the target in MRI, CT, ultrasound, or other relevant imagery. This paper reports a system for fast automatic needle tip and trajectory localization and visualization in MRI that has been developed and tested in the context of an active clinical research program in prostate biopsy. To the best of our knowledge, this is the first reported system for this clinical application and also the first reported system that leverages deep neural networks for segmentation and localization of needles in MRI across biomedical applications. Needle tip and trajectory were annotated on 583 T2-weighted intra-procedural MRI scans acquired after needle insertion for 71 patients who underwent transperineal MRI-targeted biopsy procedure at our institution. The images were divided into two independent training-validation and test sets at the patient level. A deep 3-D fully convolutional neural network model was developed, trained, and deployed on these samples. The accuracy of the proposed method, as tested on previously unseen data, was 2.80-mm average in needle tip detection and 0.98° in needle trajectory angle. An observer study was designed in which independent annotations by a second observer, blinded to the original observer, were compared with the output of the proposed method. The resultant error was comparable to the measured inter-observer concordance, reinforcing the clinical acceptability of the proposed method. The proposed system has the potential for deployment in clinical routine.