Contrastive semi-supervised harmonization of single-shell to multi-shell diffusion MRI

Diffusion weighted MRI (DW-MRI) harmonization is necessary for multi-site or multi-acquisition studies. Current statistical methods address the need to harmonize from one site to another, but do not simultaneously consider the use of multiple datasets which are comprised of multiple sites, acquisitions protocols, and age demographics. This work explores deep learning methods which can generalize across these variations through semi-supervised and unsupervised learning while also learning to estimate multi-shell data from single-shell data using the Multi-shell Diffusion MRI Harmonization Challenge (MUSHAC) and Baltimore Longitudinal Study on Aging (BLSA) datasets. We compare disentanglement harmonization models, which seek to encode anatomy and acquisition in separate latent spaces, and a CycleGAN harmonization model, which uses generative adversarial networks (GAN) to perform style transfer between sites, to the baseline preprocessing and to SHORE interpolation. We find that the disentanglement models achieve superior performance in harmonizing all data while at the same transforming the input data to a single target space across several diffusion metrics (fractional anisotropy, mean diffusivity, mean kurtosis, primary eigenvector).

Learning white matter subject-specific segmentation from structural MRI

PURPOSE

Mapping brain white matter (WM) is essential for building an understanding of brain anatomy and function. Tractography-based methods derived from diffusion-weighted MRI (dMRI) are the principal tools for investigating WM. These procedures rely on time-consuming dMRI acquisitions that may not always be available, especially for legacy or time-constrained studies. To address this problem, we aim to generate WM tracts from structural magnetic resonance imaging (MRI) image by deep learning.

METHODS

Following recently proposed innovations in structural anatomical segmentation, we evaluate the feasibility of training multiply spatial localized convolution neural networks to learn context from fixed spatial patches from structural MRI on standard template. We focus on six widely used dMRI tractography algorithms (TractSeg, RecoBundles, XTRACT, Tracula, automated fiber quantification (AFQ), and AFQclipped) and train 125 U-Net models to learn these techniques from 3870 T1-weighted images from the Baltimore Longitudinal Study of Aging, the Human Connectome Project S1200 release, and scans acquired at Vanderbilt University.

RESULTS

The proposed framework identifies fiber bundles with high agreement against tractography-based pathways with a median Dice coefficient from 0.62 to 0.87 on a test cohort, achieving improved subject-specific accuracy when compared to population atlas-based methods. We demonstrate the generalizability of the proposed framework on three externally available datasets.

CONCLUSIONS

We show that patch-wise convolutional neural network can achieve robust bundle segmentation from T1w. We envision the use of this framework for visualizing the expected course of WM pathways when dMRI is not available.

Efficient Quality Control with Mixed CT and CTA Datasets

Deep learning promises the extraction of valuable information from traumatic brain injury (TBI) datasets and depends on efficient navigation when using large-scale mixed computed tomography (CT) datasets from clinical systems. To ensure a cleaner signal while training deep learning models, removal of computed tomography angiography (CTA) and scans with streaking artifacts is sensible. On massive datasets of heterogeneously sized scans, time-consuming manual quality assurance (QA) by visual inspection is still often necessary, despite the expectation of CTA annotation (artifact annotation is not expected). We propose an automatic QA approach for retrieving CT scans without artifacts by representing 3D scans as 2D axial slice montages and using a multi-headed convolutional neural network to detect CT vs CTA and artifact vs no artifact. We sampled 848 scans from a mixed CT dataset of TBI patients and performed 4-fold stratified cross-validation on 698 montages followed by an ablation experiment-150 stratified montages were withheld for external validation evaluation. Aggregate AUC for our main model was 0.978 for CT detection, 0.675 for artifact detection during cross-validation and 0.965 for CT detection, 0.698 for artifact detection on the external validation set, while the ablated model showed 0.946 for CT detection, 0.735 for artifact detection during cross-validation and 0.937 for CT detection, 0.708 for artifact detection on the external validation set. While our approach is successful for CT detection, artifact detection performance is potentially depressed due to the heterogeneity of present streaking artifacts and a suboptimal number of artifact scans in our training data.

Mapping the Impact of Non-Linear Gradient Fields on Diffusion MRI Tensor Estimation

Non-linear gradients impact diffusion weighted (DW) MRI by corrupting the experimental setup and lead to problems during image encoding including the effects in-plane distortion, in-plane shifts, intensity modulations and phase errors. Recent studies have been shown this may present significant complication in the interpretation of results and conclusion while studying tractography and tissue microstructure in data. To interpret the degree in consequences of gradient non-linearities between the desired and achieved gradients, we introduced empirically derived gradient nonlinear fields at different orientations and different tensor properties. The impact is assessed through diffusion tensor properties including mean diffusivity (MD), fractional anisotropy (FA) and principal eigen vector (PEV). The study shows lower FA are more susceptible to LR fields and LR fields with determinant <1 or >1 corrupt tensor more. The corruption can result in significantly different FA based on true-FA and LR field. Apparent MD decreases for negative determinant, on the other hand positive determinant shows the opposite effect. LR field have a larger impact on PEV when FA value is small. The results are dependent on the underlying orientation, non-linear field corruption can cause both increase and decrease of estimated FA, MD and PEV value. This work provides insight into characterizing the non-linear gradient error and aid in selecting correction techniques to address the inaccuracies in b-values.

TractEM: Evaluation of protocols for deterministic tractography white matter atlas

Reproducible identification of white matter pathways across subjects is essential for the study of structural connectivity of the human brain. One of the key challenges is anatomical differences between subjects and human rater subjectivity in labeling. Labeling white matter regions of interest presents many challenges due to the need to integrate both local and global information. Clearly communicating the manual processes to capture this information is cumbersome, yet essential to lay a solid foundation for comprehensive atlases. Segmentation protocols must be designed so the interpretation of the requested tasks as well as locating structural landmarks is anatomically accurate, intuitive and reproducible. In this work, we quantified the reproducibility of a first iteration of an open/public multi-bundle segmentation protocol. This allowed us to establish a baseline for its reproducibility as well as to identify the limitations for future iterations. The protocol was tested/evaluated on both typical 3 T research acquisition Baltimore Longitudinal Study of Aging (BLSA) and high-acquisition quality Human Connectome Project (HCP) datasets. The results show that a rudimentary protocol can produce acceptable intra-rater and inter-rater reproducibility. However, this work highlights the difficulty in generalizing reproducible results and the importance of reaching consensus on anatomical description of white matter pathways. The protocol has been made available in open source to improve generalizability and reliability in collaboration. The goal is to improve upon the first iteration and initiate a discussion on the anatomical validity (or lack thereof) of some bundle definitions and the importance of reproducibility of tractography segmentation.

MASiVar: Multisite, multiscanner, and multisubject acquisitions for studying variability in diffusion weighted MRI

PURPOSE

Diffusion-weighted imaging allows investigators to identify structural, microstructural, and connectivity-based differences between subjects, but variability due to session and scanner biases is a challenge.

METHODS

To investigate DWI variability, we present MASiVar, a multisite data set consisting of 319 diffusion scans acquired at 3 T from b = 1000 to 3000 s/mm2 across 14 healthy adults, 83 healthy children (5 to 8 years), three sites, and four scanners as a publicly available, preprocessed, and de-identified data set. With the adult data, we demonstrate the capacity of MASiVar to simultaneously quantify the intrasession, intersession, interscanner, and intersubject variability of four common DWI processing approaches: (1) a tensor signal representation, (2) a multi-compartment neurite orientation dispersion and density model, (3) white-matter bundle segmentation, and (4) structural connectomics. Respectively, we evaluate region-wise fractional anisotropy, mean diffusivity, and principal eigenvector; region-wise CSF volume fraction, intracellular volume fraction, and orientation dispersion index; bundle-wise shape, volume, fractional anisotropy, and length; and whole connectome correlation and maximized modularity, global efficiency, and characteristic path length.

RESULTS

We plot the variability in these measures at each level and find that it consistently increases with intrasession to intersession to interscanner to intersubject effects across all processing approaches and that sometimes interscanner variability can approach intersubject variability.

CONCLUSIONS

This study demonstrates the potential of MASiVar to more globally investigate DWI variability across multiple levels and processing approaches simultaneously and suggests harmonization between scanners for multisite analyses should be considered before inference of group differences on subjects.

P–097 The impact of SARS-CoV–2 on male gonadal function. A longitudinal study

Study question

Can severe acute respiratory syndrome coronavirus 2 (SARS-CoV–2) be detected in the semen of SARS-CoV–2 positive men, and does SARS-CoV–2 infection affect male reproductive function?

Summary answer

No SARS-CoV–2 RNA was detected in semen. An impact of SARS-CoV–2 infection on semen quality and reproductive hormone profile awaits evaluation at 3 + 6 months follow-up.

What is known already

SARS-CoV–2 may use angiotensin-converting enzyme (ACE)2 as an entry point into the cell. As ACE2 is expressed in testicular tissue, it has been speculated that SARS-CoV–2 may affect the male reproductive system. A cohort study including 38 male COVID–19 patients showed that SARS-CoV–2 was present in the semen of six patients (15.8%) [Li et al., 2020]. Later studies including a total of 223 patients have not provided evidence of transmission of SARS-CoV–2 via semen. There are to date no available longitudinal studies on semen quality following SARS-CoV–2 infection.

Study design, size, duration

Longitudinal cohort study including 50 non-hospitalized men from the general population in the Capital Region of Denmark. All participants had a confirmed SARS-CoV–2 infection by reverse-transcription polymerase chain reaction (RT-PCR) on oropharyngeal swab material within the last week. The presence of SARS-CoV–2 in semen samples by RT-PCR, semen parameters and reproductive hormone profile were assessed at inclusion and at 3 + 6 months follow-up. SARS-CoV–2 antibody levels were assessed 3–5 weeks after inclusion.

Participants/materials, setting, methods

SARS-CoV–2-positive males (age 18–60 years) were included. Oropharyngeal and semen samples were tested by RT-PCR applying the E-Sarbeco primers and probe published by Corman et al. 2020 and adapted to TaqMan Fast Virus 1-step master mix and LightCycler 480 as previously reported by Jørgensen et al. 2020. SARS-CoV–2 antibodies were detected using the serological immunoassay from Shenzhen YHLO Biotech on the iFlash 1800. Semen quality parameters were analysed according to World Health Organisation (WHO) standards.

Main results and the role of chance

To date, 25 men with a mean age of 35 years have been included in the study. SARS-CoV–2 RNA could not be detected in the semen samples of any of the 25 men at the time of inclusion. Twenty-one of the 25 men (84,0%) had a same day RT-PCR-confirmed SARS-CoV–2 infection in an oropharyngeal swab. RT-PCR cycle threshold (ct) values were distributed as follows: four (19,0%) were strongly positive (ct < 25), 16 (76,2%) intermediately positive (ct 25–35) and one (4,8%) weakly positive (ct 35–45). The four men without PCR-confirmed SARS-CoV–2 infection all had a positive IgG response to SARS-CoV–2 at the time of inclusion. Longitudinal semen and reproductive hormone profiles analyses will be performed. Further studies are needed to prove whether SARS-CoV–2 can be transmitted to the male reproductive tract and whether SARS-CoV–2 infection may cause alterations of spermatogenesis and endocrine function.

Limitations, reasons for caution

Strengths of this study are the unselected population of men examined within a week after confirmed SARS-CoV–2 infection and the follow-up of semen parameters and endocrine profile. Limitations are the limited sample size and the fact that semen quality was not known before the participants were diagnosed with COVID–19.

Wider implications of the findings: Knowledge of viral detection and semen persistence of SARS-CoV–2 is essential for clinical practice and public health. There is a need for evidence-based counselling on the impact of SARS-CoV–2 infection for patients undergoing assisted reproduction technology and patients who have a need for semen cryopreservation.

Trial registration number

H–20027362

PreQual: An automated pipeline for integrated preprocessing and quality assurance of diffusion weighted MRI images

PURPOSE

Diffusion weighted MRI imaging (DWI) is often subject to low signal-to-noise ratios (SNRs) and artifacts. Recent work has produced software tools that can correct individual problems, but these tools have not been combined with each other and with quality assurance (QA). A single integrated pipeline is proposed to perform DWI preprocessing with a spectrum of tools and produce an intuitive QA document.

METHODS

The proposed pipeline, built around the FSL, MRTrix3, and ANTs software packages, performs DWI denoising; inter-scan intensity normalization; susceptibility-, eddy current-, and motion-induced artifact correction; and slice-wise signal drop-out imputation. To perform QA on the raw and preprocessed data and each preprocessing operation, the pipeline documents qualitative visualizations, quantitative plots, gradient verifications, and tensor goodness-of-fit and fractional anisotropy analyses.

RESULTS

Raw DWI data were preprocessed and quality checked with the proposed pipeline and demonstrated improved SNRs; physiologic intensity ratios; corrected susceptibility-, eddy current-, and motion-induced artifacts; imputed signal-lost slices; and improved tensor fits. The pipeline identified incorrect gradient configurations and file-type conversion errors and was shown to be effective on externally available datasets.

CONCLUSIONS

The proposed pipeline is a single integrated pipeline that combines established diffusion preprocessing tools from major MRI-focused software packages with intuitive QA.

Pandora: 4-D White Matter Bundle Population-Based Atlases Derived from Diffusion MRI Fiber Tractography

Brain atlases have proven to be valuable neuroscience tools for localizing regions of interest and performing statistical inferences on populations. Although many human brain atlases exist, most do not contain information about white matter structures, often neglecting them completely or labelling all white matter as a single homogenous substrate. While few white matter atlases do exist based on diffusion MRI fiber tractography, they are often limited to descriptions of white matter as spatially separate "regions" rather than as white matter "bundles" or fascicles, which are well-known to overlap throughout the brain. Additional limitations include small sample sizes, few white matter pathways, and the use of outdated diffusion models and techniques. Here, we present a new population-based collection of white matter atlases represented in both volumetric and surface coordinates in a standard space. These atlases are based on 2443 subjects, and include 216 white matter bundles derived from 6 different automated state-of-the-art tractography techniques. This atlas is freely available and will be a useful resource for parcellation and segmentation.

Empirical field mapping for gradient nonlinearity correction of multi-site diffusion weighted MRI

BACKGROUND

Achieving inter-site / inter-scanner reproducibility of diffusion weighted magnetic resonance imaging (DW-MRI) metrics has been challenging given differences in acquisition protocols, analysis models, and hardware factors.

PURPOSE

Magnetic field gradients impart scanner-dependent spatial variations in the applied diffusion weighting that can be corrected if the gradient nonlinearities are known. However, retrieving manufacturer nonlinearity specifications is not well supported and may introduce errors in interpretation of units or coordinate systems. We propose an empirical approach to mapping the gradient nonlinearities with sequences that are supported across the major scanner vendors.

STUDY TYPE

Prospective observational study.

SUBJECTS

A spherical isotropic diffusion phantom, and a single human control volunteer.

FIELD STRENGTH/SEQUENCE

3 T (two scanners). Stejskal-Tanner spin echo sequence with b-values of 1000, 2000 s/mm² with 12, 32, and 384 diffusion gradient directions per shell.

ASSESSMENT

We compare the proposed correction with the prior approach using manufacturer specifications against typical diffusion pre-processing pipelines (i.e., ignoring spatial gradient nonlinearities). In phantom data, we evaluate metrics against the ground truth. In human and phantom data, we evaluate reproducibility across scans, sessions, and hardware.

STATISTICAL TESTS

Wilcoxon rank-sum test between uncorrected and corrected data.

RESULTS

In phantom data, our correction method reduces variation in mean diffusivity across sessions over uncorrected data (p < 0.05). In human data, we show that this method can also reduce variation in mean diffusivity across scanners (p < 0.05).

CONCLUSION

Our method is relatively simple, fast, and can be applied retroactively. We advocate incorporating voxel-specific b-value and b-vector maps should be incorporated in DW-MRI harmonization preprocessing pipelines to improve quantitative accuracy of measured diffusion parameters.

Semi-supervised Machine Learning with MixMatch and Equivalence Classes

Semi-supervised methods have an increasing impact on computer vision tasks to make use of scarce labels on large datasets, yet these approaches have not been well translated to medical imaging. Of particular interest, the MixMatch method achieves significant performance improvement over popular semi-supervised learning methods with scarce labels in the CIFAR-10 dataset. In a complementary approach, Nullspace Tuning on equivalence classes offers the potential to leverage multiple subject scans when the ground truth for the subject is unknown. This work is the first to (1) explore MixMatch with Nullspace Tuning in the context of medical imaging and (2) characterize the impacts of the methods with diminishing labels. We consider two distinct medical imaging domains: skin lesion diagnosis and lung cancer prediction. In both cases we evaluate models trained with diminishing labeled data using supervised, MixMatch, and Nullspace Tuning methods as well as MixMatch with Nullspace Tuning together. MixMatch with Nullspace Tuning together is able to achieve an AUC of 0.755 in lung cancer diagnosis with only 200 labeled subjects on the National Lung Screening Trial and a balanced multi-class accuracy of 77% with only 779 labeled examples on HAM10000. This performance is similar to that of the fully supervised methods when all labels are available. In advancing data driven methods in medical imaging, it is important to consider the use of current state-of-the-art semi-supervised learning methods from the greater machine learning community and their impact on the limitations of data acquisition and annotation.

Enabling Multi-Shell b-Value Generalizability of Data-Driven Diffusion Models with Deep SHORE

Intra-voxel models of the diffusion signal are essential for interpreting organization of the tissue environment at micrometer level with data at millimeter resolution. Recent advances in data driven methods have enabled direct comparison and optimization of methods for in-vivo data with externally validated histological sections with both 2-D and 3-D histology. Yet, all existing methods make limiting assumptions of either (1) model-based linkages between b-values or (2) limited associations with single shell data. We generalize prior deep learning models that used single shell spherical harmonic transforms to integrate the recently developed simple harmonic oscillator reconstruction (SHORE) basis. To enable learning on the SHORE manifold, we present an alternative formulation of the fiber orientation distribution (FOD) object using the SHORE basis while representing the observed diffusion weighted data in the SHORE basis. To ensure consistency of hyper-parameter optimization for SHORE, we present our Deep SHORE approach to learn on a data-optimized manifold. Deep SHORE is evaluated with eight-fold cross-validation of a preclinical MRI-histology data with four b-values. Generalizability of in-vivo human data is evaluated on two separate 3T MRI scanners. Specificity in terms of angular correlation (ACC) with the preclinical data improved on single shell: 0.78 relative to 0.73 and 0.73, multi-shell: 0.80 relative to 0.74 (p < 0.001). In the in-vivo human data, Deep SHORE was more consistent across scanners with 0.63 relative to other multi-shell methods 0.39, 0.52 and 0.57 in terms of ACC. In conclusion, Deep SHORE is a promising method to enable data driven learning with DW-MRI under conditions with varying b-values, number of diffusion shells, and gradient directions per shell.

Deep learning reveals untapped information for local white-matter fiber reconstruction in diffusion-weighted MRI

PURPOSE

Diffusion-weighted magnetic resonance imaging (DW-MRI) is of critical importance for characterizing in-vivo white matter. Models relating microarchitecture to observed DW-MRI signals as a function of diffusion sensitization are the lens through which DW-MRI data are interpreted. Numerous modern approaches offer opportunities to assess more complex intra-voxel structures. Nevertheless, there remains a substantial gap between intra-voxel estimated structures and ground truth captured by 3-D histology.

METHODS

Herein, we propose a novel data-driven approach to model the non-linear mapping between observed DW-MRI signals and ground truth structures using a sequential deep neural network regression using residual block deep neural network (ResDNN). Training was performed on two 3-D histology datasets of squirrel monkey brains and validated on a third. A second validation was performed using scan-rescan datasets of 12 subjects from Human Connectome Project. The ResDNN was compared with multiple micro-structure reconstruction methods and super resolved-constrained spherical deconvolution (sCSD) in particular as baseline for both the validations.

RESULTS

Angular correlation coefficient (ACC) is a correlation/similarity measure and can be interpreted as accuracy when compared with a ground truth. The median ACC of ResDNN is 0.82 and median ACC's of different variants of CSD are 0.75, 0.77, 0.79. The mean, median and std. of ResDNN & sCSD ACC across 12 subjects from HCP are 0.74, 0.88, 0.31 and 0.61, 0.71, 0.31 respectively.

CONCLUSION

This work highlights the ability of deep learning to capture linkages between ex-vivo ground truth data with feasible MRI sequences. The data-driven approach is applicable to human in-vivo data and results in intriguingly high reproducibility of orientation structure.

Monitoring of drinking water quality using automated ATP quantification

A microfluidic based system was developed for automated online method for the rapid detection and monitoring of drinking water contamination utilising microbial Adrenosine-5'-Triphosphate (ATP) as a bacterial indicator. The system comprises a polymethyl methacrylate based microfluidic cartridge inserted into an enclosure incorporating the functions of fluid storage and delivery, lysis steps and real-time detection. Design, integration and operation of the resulting automated system are reported, including the lysis method, the design of the mixing circuit, the choices of flow rate, temperature and reagent amount. Calibration curves of both total and free ATP were demonstrated to be highly linear over a range from 2.5-5000 pg/mL with the limit of detection being lower than 2.5 pg/mL of total ATP. The system was trialled in a lab study with different types of water, with lysis efficiency being found to be strongly dependent upon water type. Further development is required before online implementation.

Inter-Scanner Harmonization of High Angular Resolution DW-MRI using Null Space Deep Learning

Diffusion-weighted magnetic resonance imaging (DW-MRI) allows for non-invasive imaging of the local fiber architecture of the human brain at a millimetric scale. Multiple classical approaches have been proposed to detect both single (e.g., tensors) and multiple (e.g., constrained spherical deconvolution, CSD) fiber population orientations per voxel. However, existing techniques generally exhibit low reproducibility across MRI scanners. Herein, we propose a data-driven technique using a neural network design which exploits two categories of data. First, training data were acquired on three squirrel monkey brains using ex-vivo DW-MRI and histology of the brain. Second, repeated scans of human subjects were acquired on two different scanners to augment the learning of the network proposed. To use these data, we propose a new network architecture, the null space deep network (NSDN), to simultaneously learn on traditional observed/truth pairs (e.g., MRI-histology voxels) along with repeated observations without a known truth (e.g., scan-rescan MRI). The NSDN was tested on twenty percent of the histology voxels that were kept completely blind to the network. NSDN significantly improved absolute performance relative to histology by 3.87% over CSD and 1.42% over a recently proposed deep neural network approach. Moreover, it improved reproducibility on the paired data by 21.19% over CSD and 10.09% over a recently proposed deep approach. Finally, NSDN improved generalizability of the model to a third in vivo human scanner (which was not used in training) by 16.08% over CSD and 10.41% over a recently proposed deep learning approach. This work suggests that data-driven approaches for local fiber reconstruction are more reproducible, informative and precise and offers a novel, practical method for determining these models.

Consideration of Cerebrospinal Fluid Intensity Variation in Diffusion Weighted MRI

Diffusion weighted MRI (DW-MRI) depends on accurate quantification signal intensities that reflect directional apparent diffusion coefficients (ADC). Signal drift and fluctuations during imaging can cause systematic non-linearities that manifest as ADC changes if not corrected. Here, we present a case study on a large longitudinal dataset of typical diffusion tensor imaging. We investigate observed variation in the cerebral spinal fluid (CSF) regions of the brain, which should represent compartments with isotropic diffusivity. The study contains 3949 DW-MRI acquisitions of the human brain with 918 subjects and 542 with repeated scan sessions. We provide an analysis of the inter-scan, inter-session, and intra-session variation and an analysis of the associations with the applied diffusion gradient directions. We investigate a hypothesis that CSF models could be used in lieu of an interspersed minimally diffusion-weighted image (b0) correction. Variation in CSF signal is not largely attributable to within-scan dynamic anatomical changes (3.6%), but rather has substantial variation across scan sessions (10.6%) and increased variation across individuals (26.6%). Unfortunately, CSF intensity is not solely explained by a main drift model or a gradient model, but rather has statistically significant associations with both possible explanations. Further exploration is necessary for CSF drift to be used as an effective harmonization technique.

Harmonizing 1.5T/3T Diffusion Weighted MRI through Development of Deep Learning Stabilized Microarchitecture Estimators

Diffusion weighted magnetic resonance imaging (DW-MRI) is interpreted as a quantitative method that is sensitive to tissue microarchitecture at a millimeter scale. However, the sensitization is dependent on acquisition sequences (e.g., diffusion time, gradient strength, etc.) and susceptible to imaging artifacts. Hence, comparison of quantitative DW-MRI biomarkers across field strengths (including different scanners, hardware performance, and sequence design considerations) is a challenging area of research. We propose a novel method to estimate microstructure using DW-MRI that is robust to scanner difference between 1.5T and 3T imaging. We propose to use a null space deep network (NSDN) architecture to model DW-MRI signal as fiber orientation distributions (FOD) to represent tissue microstructure. The NSDN approach is consistent with histologically observed microstructure (on previously acquired ex vivo squirrel monkey dataset) and scan-rescan data. The contribution of this work is that we incorporate identical dual networks (IDN) to minimize the influence of scanner effects via scan-rescan data. Briefly, our estimator is trained on two datasets. First, a histology dataset was acquired on three squirrel monkeys with corresponding DW-MRI and confocal histology (512 independent voxels). Second, 37 control subjects from the Baltimore Longitudinal Study of Aging (67-95 y/o) were identified who had been scanned at 1.5T and 3T scanners (b-value of 700 s/mm2, voxel resolution at 2.2mm, 30-32 gradient volumes) with an average interval of 4 years (standard deviation 1.3 years). After image registration, we used paired white matter (WM) voxels for 17 subjects and 440 histology voxels for training and 20 subjects and 72 histology voxels for testing. We compare the proposed estimator with super-resolved constrained spherical deconvolution (CSD) and a previously presented regression deep neural network (DNN). NSDN outperformed CSD and DNN in angular correlation coefficient (ACC) 0.81 versus 0.28 and 0.46, mean squared error (MSE) 0.001 versus 0.003 and 0.03, and general fractional anisotropy (GFA) 0.05 versus 0.05 and 0.09. Further validation and evaluation with contemporaneous imaging are necessary, but the NSDN is promising avenue for building understanding of microarchitecture in a consistent and device-independent manner.

Characterization and correlation of signal drift in diffusion weighted MRI

Diffusion weighted MRI (DWMRI) and the myriad of analysis approaches (from tensors to spherical harmonics and brain tractography to body multi-compartment models) depend on accurate quantification of the apparent diffusion coefficient (ADC). Signal drift during imaging (e.g., due to b0 drift associated with heating) can cause systematic non-linearities that manifest as ADC changes if not corrected. Herein, we present a case study on two phantoms on one scanner. Different scan protocols exhibit different degrees of drift during similar scans and may be sensitive to the order of scans within an exam. Vos et al. recently reviewed the effects of signal drift in DWMRI acquisitions and proposed a temporal model for correction. We propose a novel spatial-temporal model to correct for higher order aspects of the signal drift and derive a statistically robust variant. We evaluate the Vos model and propose a method using two phantoms that mimic the ADC of the relevant brain tissue (0.36-2.2 × 10-3 mm²/s) on a single 3 T scanner. The phantoms are (1) a spherical isotropic sphere consisting of a single concentration of polyvinylpyrrolidone (PVP) and (2) an ice-water phantom with 13 vials of varying PVP concentrations. To characterize the impact of interspersed minimally weighted volumes ("b0's"), image volumes with b-value equal to 0.1 s/mm² are interspersed every 8, 16, 32, 48, and 96 diffusion weighted volumes in different trials. Signal drift is found to have spatially varying effects that are not accounted for with temporal-only models. The novel model captures drift more accurately (i.e., reduces the overall change per-voxel over the course of a scan) and results in more consistent ADC metrics.

Phantom-based field maps for gradient nonlinearity correction in diffusion imaging

Gradient coils in magnetic resonance imaging do not produce perfectly linear gradient fields. For diffusion imaging, the field nonlinearities cause the amplitude and direction of the applied diffusion gradients to vary over the field of view. This leads to site- and scan-specific systematic errors in estimated diffusion parameters such as diffusivity and anisotropy, reducing reliability especially in studies that take place over multiple sites. These errors can be substantially reduced if the actual scanner-specific gradient coil magnetic fields are known. The nonlinearity of the coil fields is measured by scanner manufacturers and used internally for geometric corrections, but obtaining and using the information for a specific scanner may be impractical for many sites that operate without special-purpose local engineering and research support. We have implemented an empirical field-mapping procedure using a large phantom combined with a solid harmonic approximation to the coil fields that is simple to perform and apply. Here we describe the accuracy and precision of the approach in reproducing manufacturer gold standard field maps and in reducing spatially varying errors in quantitative diffusion imaging for a specific scanner. Before correction, median B value error ranged from 33 - 41 relative to manufacturer specification at 100 mm from isocenter; correction reduced this to 0 - 4. On-axis spatial variation in the estimated mean diffusivity of an isotropic phantom was 2.2% - 4.1% within 60 mm of isocenter before correction, 0.5% - 1.6% after. Expected fractional anisotropy in the phantom was 0; highest estimated fractional anisotropy within 60 mm of isocenter was reduced from 0.024 to 0.012 in the phase encoding direction (48% reduction) and from 0.020 to 0.006 in the frequency encoding direction (72% reduction).

A systematic review of drug-induced subacute cutaneous lupus erythematosus

The initial appearance of subacute cutaneous lupus erythematosus (SCLE) skin lesions in conjunction with Ro/SS-A autoantibodies occurring as an adverse reaction to hydrochlorothiazide [i.e. drug-induced SCLE (DI-SCLE)] was first reported in 1985. Over the past decade an increasing number of drugs in different classes has been implicated as triggers for DI-SCLE. The management of DI-SCLE can be especially challenging in patients taking multiple medications capable of triggering DI-SCLE. Our objectives were to review the published English language literature on DI-SCLE and use the resulting summary data pool to address questions surrounding drug-induced SCLE and to develop guidelines that might be of value to clinicians in the diagnosis and management of DI-SCLE. A systematic review of the Medline/PubMed-cited literature on DI-SCLE up to August 2009 was performed. Our data collection and analysis strategies were prospectively designed to answer a series of questions related to the clinical, prognostic and pathogenetic significance of DI-SCLE. One hundred and seventeen cases of DI-SCLE were identified and reviewed. White women made up the large majority of cases, and the mean overall age was 58·0 years. Triggering drugs fell into a number of different classes, highlighted by antihypertensives and antifungals. Time intervals ('incubation period') between drug exposure and appearance of DI-SCLE varied greatly and were drug class dependent. Most cases of DI-SCLE spontaneously resolved within weeks of drug withdrawal. Ro/SS-A autoantibodies were present in 80% of the cases in which such data were reported and most remained positive after resolution of SCLE skin disease activity. No significant differences in the clinical, histopathological or immunopathological features between DI-SCLE and idiopathic SCLE were detected. There is now adequate published experience to suggest that DI-SCLE does not differ clinically, histopathologically or immunologically from idiopathic SCLE. It should be recognized as a distinct clinical constellation differing clinically and immunologically from the classical form of drug-induced systemic lupus erythematosus.

A growth factor antagonist as a targeting agent for sterically stabilized liposomes in human small cell lung cancer

The ability of a growth factor antagonist, [D-Arg(6),D-Trp(7,9)-N(me)Phe(8)]-substance P(6-11), named antagonist G, to selectively target polyethylene glycol-grafted liposomes (known as sterically stabilized liposomes) to a human classical small cell lung cancer (SCLC) cell line, H69, was examined. Our results showed that radiolabeled antagonist G-targeted sterically stabilized liposomes (SLG) bound to H69 cells with higher avidity than free antagonist G and were internalized (reaching a maximum of 13000 SLG/cell), mainly through a receptor-mediated process, likely involving clathrin-coated pits. This interaction was confirmed by confocal microscopy to be peptide- and cell-specific. Moreover, it was shown that SLG significantly improved the nuclear delivery of encapsulated doxorubicin to the target cells, increasing the cytotoxic activity of the drug over non-targeted liposomes. In mice, [(125)I]tyraminylinulin-containing SLG were long circulating, with a half-life of 13 h. Use of peptides like antagonist G to promote binding and internalization of sterically stabilized liposomes, with their accompanying drug loads, i.e., anticancer drugs, genes or antisense oligonucleotides, into target cells has the potential to improve therapy of SCLC.

Tissue factor-mediated endocytosis, recycling, and degradation of factor VIIa by a clathrin-independent mechanism not requiring the cytoplasmic domain of tissue factor

Endocytosis and recycling of coagulation factor VIIa (VIIa) bound to tissue factor (TF) was investigated in baby hamster kidney (BHK) cells stably transfected with TF or TF derivatives. Cell surface expression of TF on BHK cells was required for VIIa internalization and degradation. Approximately 50% of cell surface-bound VIIa was internalized in one hour, and a majority of the internalized VIIa was degraded soon thereafter. Similar rates of VIIa internalization and degradation were obtained with BHK cells transfected with a cytoplasmic domain-deleted TF variant or with a substitution of serine for cysteine at amino acid residue 245 (C245S). Endocytosis of VIIa bound to TF was an active process. Acidification of the cytosol, known to inhibit the internalization via clathrin-coated pits, did not affect the internalization of VIIa. Furthermore, receptor-associated protein, known to block binding of all established ligands to members of the low-density lipoprotein receptor family, was without an effect on the internalization of VIIa. Addition of tissue factor pathway inhibitor/factor Xa complex did not affect the internalization rate significantly. A substantial portion (20% to 25%) of internalized VIIa was recycled back to the cell surface as an intact and functional protein. Although the recycled VIIa constitutes to only approximately 10% of available cell surface TF/VIIa sites, it accounts for 65% of the maximal activation of factor X by the cell surface TF/VIIa. In summary, the present data provide evidence that TF-dependent internalization of VIIa in kidney cells occurs through a clathrin-independent mechanism and does not require the cytoplasmic domain of TF.

Impermeability of the GIRK2 weaver channel to divalent cations

A single amino acid mutation (G156S) in the putative pore-forming region of the G protein-sensitive, inwardly rectifying K(+) channel subunit, GIRK2, renders the conductance constitutively active and nonselective for monovalent cations. The mutant channel subunit (GIRK2wv) causes the pleiotropic weaver disease in mice, which is characterized by the selective vulnerability of cerebellar granule cells and Purkinje cells, as well as dopaminergic neurons in the mesencephalon, to cell death. It has been proposed that divalent cation permeability through constitutively active GIRK2wv channels contributes to a rise in internal calcium in the GIRK2wv-expressing neurons, eventually leading to cell death. We carried out comparative studies of recombinant GIRK2wv channels expressed in Xenopus oocytes and COS-7 cells to determine the magnitude and relative permeability of the GIRK2wv conductance to Ca(2+). Data from these studies demonstrate that the properties of the expressed current differ in the two systems and that when recombinant GIRK2wv is expressed in mammalian cells it is impermeable to Ca(2+).

Discordant expression of tissue factor and its activity in polarized epithelial cells. Asymmetry in anionic phospholipid availability as a possible explanation

Recent studies have shown a discrepancy between the level of tissue factor (TF) expression and the level of TF procoagulant activity on the apical and basolateral surface domains of polarized epithelial cells. The present investigation was performed to elucidate possible reasons for the discordant expression of TF and its activity on the surface of polarized epithelial cells using a human intestinal epithelial cell line, Caco-2 and Madin-Darby canine kidney epithelial cells, type II (MDCK-II). Functional activity of coagulation factor VIIa (VIIa) in complex with TF was 6- to 7-fold higher on the apical than the basolateral surface in polarized Caco-2 cells. In contrast, no significant difference was found in the formation of TF/VIIa complexes between the apical and basolateral surface. Confocal microscopy of Caco-2 cells showed TF expression on both the apical and the basolateral surface domains. Studies with MDCK-II cells showed that the specific functional activity of TF expressed on the apical cell surface was 5-fold higher than on the basolateral surface. To test whether differential expression of TF pathway inhibitor (TFPI) on the apical and basolateral surface could account for differences in TF/VIIa functional activity, we measured cell-surface-bound TFPI activity in Caco-2 cells. Small but similar amounts of TFPI were found on both surfaces. Further, addition of inhibitory anti-TFPI antibodies induced a similar enhancement of TF/VIIa activity on both surface domains. Because the availability of anionic phospholipids on the outer leaflet of the cell membrane could regulate TF/VIIa functional activity, we measured the distribution of anionic phospholipids on the apical and basolateral surface by annexin V binding and thrombin generation. The results showed that the anionic phospholipid content on the basolateral surface, compared with the apical surface, was 3- to 4-fold lower. Mild acid treatment of polarized Caco-2 cells, which markedly increased the anionic phospholipid content on the basolateral surface membrane, increased the TF/VIIa activity on the basolateral surface without affecting the number of TF/VIIa complexes formed on the surface. Overall, our data suggest that an uneven expression of TF/VIIa activity between the apical and basolateral surface of polarized epithelial cells is caused by differences in anionic phospholipid content between the two surface domains and not from a polar distribution of TFPI.

Effects of endotoxin lung injury on NMR T2 relaxation

The effects of endotoxin injury on lung NMR relaxation times (T1, CPMG T2, and Hahn decay constant (Hahn T2)) were studied in excised unperfused rat lungs. Blinded histologic examination showed no clear-cut separation between endotoxin and control lungs. Morphometric lung tissue volume density and gravimetric lung water content did not differ significantly between the two groups. In contrast, the values of the fast, intermediate, and slow T2 components, obtained by multiexponential analysis of the CPMG decay curve, increased markedly after endotoxin administration, with minimal overlap between endotoxin and control values. The response of Hahn T2 was, in general, in the same direction as that of CPMG T2; however, Hahn T2 may be more affected by measurement errors and may be less sensitive to the presence of lung injury. T1 showed minimal changes after injury. The present data suggest that CPMG T2 measurements can consistently detect the presence of lung injury even when conventional histologic, morphometric, and gravimetric studies provide negative or equivocal results, and that the CMPG T2 method is superior, in this respect, to the Hahn decay method. T1 does not appear to be sensitive to lung injury in the absence of significant lung water accumulation.

Evaluation of blood clearance rates and biodistribution of poly(2-oxazoline)-grafted liposomes

Two amphipatic polymers of the poly(2-oxazoline) family, poly(2-methyl-2-oxazoline) (PMOZ) and poly(2-ethyl-2-oxazoline) (PEOZ), were synthesized with the carboxylic group positioned at either the initiation or termination ends of the polymer chains. Distearoylphosphatidylethanolamine was covalently linked to the carboxyl groups of the polymers, resulting in conjugates which incorporate readily into liposomes. Systematic evaluation of plasma clearance kinetics and biodistribution of liposomes containing hydrogenated soy phosphatidylcholine, cholesterol, and 5 mol % the polymer-lipid conjugates in mice revealed the following. Both polymers, PMOZ and PEOZ, exhibited long plasma lifetimes and low hepatosplenic uptake. PMOZ was more effective at decreasing blood clearance rates than PEOZ. The best results, which were quantitatively comparable to the results obtained with the optimized preparations of methoxypolyethylene glycol(PEG)-2000-grafted liposomes, were obtained with formulations containing PMOZ of molecular weight 3260.

Attachment of antibodies to sterically stabilized liposomes: evaluation, comparison and optimization of coupling procedures

Several coupling methods for binding antibodies (Ab) to liposomes have previously been developed. We were interested in examining if some of these methods would be suitable for attaching Ab to long-circulating formulations of liposomes (SL), sterically stabilized with poly(ethylene glycol) (PEG). We studied three 'classical' coupling methods in which Ab was attached at the bilayer surface of SL, and two new coupling methods in which Ab was attached at the PEG terminus. Parameters examined including binding efficiency, antibody surface density, the ability of the immunoliposomes to remote-load the anticancer drug doxorubicin, and the specific binding of the resulting immunoliposomes to target cells. The non-covalent biotin-avidin coupling method resulted in low Ab densities at the cell surface, as did a coupling in method in which maleimide-derivatized Ab was attached to the liposome surface through a thiolated phospholipid incorporated into the liposomes. The low levels of Ab achieved in these method was likely due to interference by PEG with the access of the Ab to the liposome surface. However, when a maleimide-derivatized Ab was coupled to thiolated PEG, moving the coupling reaction away from the liposome surface, very high coupling efficiencies were achieved, and these immunoliposomes achieved good specific binding to their target cells. Oxidizing the Fc region of the Ab and coupling it to the PEG terminus through a hydrazone bond was a less efficient coupling method, but had the advantage of retaining Ab orientation. Efficient remote-loading of doxorubicin was found for immunoliposomes in which Ab was attached at the PEG terminus.

A new strategy for attachment of antibodies to sterically stabilized liposomes resulting in efficient targeting to cancer cells

The development of long-circulating formulations of liposomes (S-liposomes), sterically stabilized with lipid derivatives of poly(ethylene glycol) (PEG), has increased the likelihood that these liposomes, coupled to targeting ligands such as antibodies, could be used as drug carriers to deliver therapeutic drugs to specific target cell populations in vivo. We have developed a new methodology for attaching monoclonal antibodies to the terminus of PEG on S-liposomes. A new end-group functionalized PEG-lipid derivative pyridylthiopropionoylamino-PEG- distearoylphosphatidylethanolamine (PDP-PEG-DSPE) was synthesized for this purpose. Incorporation of PDP-PEG-DSPE into S-liposomes followed by mild thiolysis of the PDP groups resulted in formation of reactive thiol groups at the periphery of the lipid vesicles. Efficient attachment of maleimide-derivatized antibodies took place under mild conditions even when the content of the functionalized PEG-lipid in S-liposomes was below 1% of total lipid. The resulting S-immunoliposomes showed efficient drug remote loading, slow drug release rates and increased survival times in circulation compared to liposomes lacking PEG. When antibodies recognizing several different tumor-associated antigens were coupled to the PEG terminus of S-liposomes a significant increase in the in vitro binding of liposomes to the target cells was observed. The binding of S-immunoliposomes containing entrapped doxorubicin to their target cell population resulted in increased cytotoxicity compared to liposomes lacking the targeting antibody.

Subcutaneous administration of liposomes: a comparison with the intravenous and intraperitoneal routes of injection

The development of long-circulating liposomes containing lipid derivatives of poly(ethylene glycol) (PEG), termed Stealth liposomes, has considerably improved the prospects for therapeutic applications of liposomal drug delivery systems. We have examined the pharmacokinetics and biodistribution of long-circulating, as compared to conventional, liposomes after subcutaneous (sc) administration in mice. Results obtained after subcutaneous administration were compared to those obtained after intravenous (iv) and intraperitoneal (ip) administration. Liposomes, following sc administration, appeared intact in the circulation subsequent to moving down the lymph node chains that drain the site of injection. Liposomes containing PEG-distearoylphosphatidylethanolamine (PEG-DSPE) resulted in the highest levels of small (80-90 nm) liposomes in the blood, with up to 30% of vivo label appearing in the blood at 12 to 24 h post-injection. In the absence PEG-DSPE approx. 4-fold lower levels of liposomes were found in the blood. Small size of the liposomes was critical to their ability to move into the circulation, with liposomes above 110-120 nm not appearing in blood to any significant extent. The presence of PEG-DSPE and cholesterol was important for the in vivo stability of the liposome after sc administration. Although liposome levels were significantly higher in the draining lymph nodes after sc administration, levels associated with other tissues were proportionately reduced relative to the iv and ip routes of administration. Liposomes appeared in blood after ip and sc administration with half-lives of approx. 0.6 and 9 h, respectively, and subsequent to appearing in blood had similar biodistribution, pharmacokinetics and half-lives (20.4 h) to liposomes given by the iv route.

Is chronic lung disease in low birth weight infants preventable? A survey of eight centers

Chronic lung disease in prematurely born infants, defined as the need for increased inspired oxygen at 28 days of age, was thought to be more common in some institutions than in others. To test this hypothesis, we surveyed the experience in the intensive care nurseries at Columbia and Vanderbilt Universities, the Universities of Texas at Dallas, Washington at Seattle, and California at San Francisco, the Brigham and Women's Hospital in Boston, Texas Children's Hospital in Houston, and Mt Sinai Hospital in Toronto. The survey included 1,625 infants with birth weights of 700 to 1,500 g. We confirmed the relationship of risk to low birth weight, white race, and male sex. Significant differences in the incidence of chronic lung disease were found between institutions even when birth weight, race, and sex were taken into consideration through a multivariate logistic regression analysis. Columbia had one of the best outcomes for low birth weight infants and the lowest incidence of chronic lung disease.

Resurgence of killing and in vivo protection mediated by lymphocytes cultured from lymph nodes draining Moloney sarcomas

We have previously documented the development and subsequent disappearance of cytolytic activity mediated by lymphocytes from lymph nodes draining Moloney sarcomas destined either to regress or grow progressively. We now report that these tumour-draining lymphnode cells (LNC) that were no longer cytotoxic, spontaneously regenerated peak levels of killing after culture in vitro for 4 days in the absence of exogenous tumour antigen. Cytolytic activity, which was antigenically specific, was mediated by T lymphocytes. Resurgence of cytolytic activity in vitro was accompanied by proliferative changes (DNA synthesis, blast transformation, cell division) which peaked on the 3rd day of culture. Although normal, nonimmune LNC underwent quantitatively similar proliferative changes in culture, the killing that developed was weak and antigenically nonspecific. Transfer of cultured, tumour-draining LNC to immunologically compromised, syngeneic mice conferred complete protection from Moloney sarcoma progression. Adoptive transfer could be delayed for 6 days after tumour induction without loss of protection. These results suggest that there exists in Moloney sarcoma-bearing mice a mechanism that limits the differentiation of pre-killer cells into cytolytically active T lymphocytes, and that such inhibition is eliminated when LNC are explanted into culture.

Inflammatory cells in solid murine neoplasms. IV. Cytolytic T lymphocytes isolated from regressing or progressing Moloney sarcomas

Highly purified suspensions of intratumoral T lymphocytes, recovered 11 and 13 days after induction of regressing or progressing Moloney sarcomas, were compared in their ability to lyse specifically the MSC cells used for tumor induction. Cytolytic activity, expressed in terms of lytic units/10(6) T cells, was similar for intratumoral T cell suspensions obtained 11 days after induction of either regressing (3.1 +/- 1.3 LU/10(6) T cells) or progressing (4.3 +/- 1.8) neoplasms. By 13 days post-induction, regressing tumors contained T lymphocytes with an increased cytolytic activity (11.1 +/- 4.5) whereas those from progressing tumors were strikingly less able to kill MSC cells (less than or equal to 0.2). This dramatic loss in cytotoxicity could not be attributed to errors associated with the enzymatic disaggregation method, inhibition by copurified endogenous tumor cells, or immunosuppression induced by viral infection. The changes in functional activity of intratumoral T lymphocytes from the two types of sarcoma appeared to be correlated with the stage of neoplasia. In this model system, cytolytic activity of T lymphocytes increased during spontaneous tumor regression whereas losses in cytotoxicity occurred coincident with the onset of inexorable progression.

Isolation of T-lymphocytes from disaggregated tumors, with high purity and good percentage recovery

A combination of two cell separation methods was utilized for the isolation of thymus-derived lymphocytes (TL) from enzymatically disaggregated tumors. Passage through Sephadex G-10-glass bead columns to remove adherent cell types followed by exposure to IgG-coated sheep red blood cell monolayers for removal of Fc receptor-bearing inflammatory cells provided functional TL suspensions of high purity with good percentage recovery.

Inflammatory cells in solid murine neoplasms. II. Cell types found throughout the course of Moloney sarcoma regression or progression

Regressing and progressing Moloney sarcomas, induced in BALB/c mice by the injection of cultured sarcoma cells (MSC)1, were sampled for histologic analysis and then disaggregated using mixtures of trypsin, collagenase and DNAse or collagenase and DNAse alone. The types of inflammatory cells (IC) found in resultant cell suspensions were determined 6, 11, 14 and 18 days post inoculation. Inflammatory infiltrates were composed almost exclusively of three cell types; neutrophils, T lymphocytes and macrophages. The extent to which each was found in tumors was related to the time post inoculation. Neutrophils were part of an early acute inflammatory response seen in both developing regressing and progressing sarcomas. The onset of regression was associated histologically with the appearance within tumors of a mononuclear inflammatory infiltrate. T lymphocytes and macrophages were the principal constituents. A higher percentage of T lymphocytes was recovered at all sampling times from regressing, compared to progressing, sarcomas. During development of the mononuclear inflammatory infiltrate there were relatively more large T cells in regressing, than in progressing tumors, and the percentage of macrophages was higher. Thereafter, the proportion of macrophages in the recovered cell population was approximately the same for both types of tumor. Such equality was more apparent than real, however, since IC were restricted to the peripheries of progressing sarcomas after the acute inflammatory phase, but continued to be found throughout regressing neoplasms. The effective ratio of macrophages and T lymphocytes to tumor cells therefore was much lower in progressing sarcomas than was suggested by percentage figures. The data presented support the concept that T lymphocytes are instrumental in causing the regression of Moloney sarcomas, possibly through interactions with macrophages.