Identification of functional white matter networks in BOLD fMRI.

White matter signals in resting state blood oxygen level dependent functional magnetic resonance (BOLD-fMRI) have been largely discounted, yet there is growing evidence that these signals are indicative of brain activity. Understanding how these white matter signals capture function can provide insight into brain physiology. Moreover, functional signals could potentially be used as early markers for neurological changes, such as in Alzheimer's Disease. To investigate white matter brain networks, we leveraged the OASIS-3 dataset to extract white matter signals from resting state BOLD-FMRI data on 711 subjects. The imaging was longitudinal with a total of 2,026 images. Hierarchical clustering was performed to investigate clusters of voxel-level correlations on the timeseries data. The stability of clusters was measured with the average Dice coefficients on two different cross fold validations. The first validated the stability between scans, and the second validated the stability between subject populations. Functional clusters at hierarchical levels 4, 9, 13, 18, and 24 had local maximum stability, suggesting better clustered white matter. In comparison with JHU-DTI-SS Type-I Atlas defined regions, clusters at lower hierarchical levels identified well defined anatomical lobes. At higher hierarchical levels, functional clusters mapped motor and memory functional regions, identifying 50.00%, 20.00%, 27.27%, and 35.14% of the frontal, occipital, parietal, and temporal lobe regions respectively.

Lung CT harmonization of paired reconstruction kernel images using generative adversarial networks.

BACKGROUND: The kernel used in CT image reconstruction is an important factor that determines the texture of the CT image. Consistency of reconstruction kernel choice is important for quantitative CT-based assessment as kernel differences can lead to substantial shifts in measurements unrelated to underlying anatomical structures. PURPOSE: In this study, we investigate kernel harmonization in a multi-vendor low-dose CT lung cancer screening cohort and evaluate our approach's validity in quantitative CT-based assessments. METHODS: Using the National Lung Screening Trial, we identified CT scan pairs of the same sessions with one reconstructed from a soft tissue kernel and one from a hard kernel. In total, 1000 pairs of five different paired kernel types (200 each) were identified. We adopt the pix2pix architecture to train models for kernel conversion. Each model was trained on 100 pairs and evaluated on 100 withheld pairs. A total of 10 models were implemented. We evaluated the efficacy of kernel conversion based on image similarity metrics including root mean squared error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM) as well as the capability of the models to reduce measurement shifts in quantitative emphysema and body composition measurements. Additionally, we study the reproducibility of standard radiomic features for all kernel pairs before and after harmonization. RESULTS: Our approach effectively converts CT images from one kernel to another in all paired kernel types, as indicated by the reduction in RMSE (p < 0.05) and an increase in the PSNR (p < 0.05) and SSIM (p < 0.05) for both directions of conversion for all pair types. In addition, there is an increase in the agreement for percent emphysema, skeletal muscle area, and subcutaneous adipose tissue (SAT) area for both directions of conversion. Furthermore, radiomic features were reproducible when compared with the ground truth features. CONCLUSIONS: Kernel conversion using deep learning reduces measurement variation in percent emphysema, muscle area, and SAT area.

Identification of functional white matter networks in BOLD fMRI.

White matter signals in resting state blood oxygen level dependent functional magnetic resonance (BOLD-fMRI) have been largely discounted, yet there is growing evidence that these signals are indicative of brain activity. Understanding how these white matter signals capture function can provide insight into brain physiology. Moreover, functional signals could potentially be used as early markers for neurological changes, such as in Alzheimer's Disease. To investigate white matter brain networks, we leveraged the OASIS-3 dataset to extract white matter signals from resting state BOLD-FMRI data on 711 subjects. The imaging was longitudinal with a total of 2,026 images. Hierarchical clustering was performed to investigate clusters of voxel-level correlations on the timeseries data. The stability of clusters was measured with the average Dice coefficients on two different cross fold validations. The first validated the stability between scans, and the second validated the stability between populations. Functional clusters at hierarchical levels 4, 9, 13, 18, and 24 had local maximum stability, suggesting better clustered white matter. In comparison with JHU-DTI-SS Type-I Atlas defined regions, clusters at lower hierarchical levels identified well-defined anatomical lobes. At higher hierarchical levels, functional clusters mapped motor and memory functional regions, identifying 50.00%, 20.00%, 27.27%, and 35.14% of the frontal, occipital, parietal, and temporal lobe regions respectively.

Cross-scale multi-instance learning for pathological image diagnosis.

Analyzing high resolution whole slide images (WSIs) with regard to information across multiple scales poses a significant challenge in digital pathology. Multi-instance learning (MIL) is a common solution for working with high resolution images by classifying bags of objects (i.e. sets of smaller image patches). However, such processing is typically performed at a single scale (e.g., 20× magnification) of WSIs, disregarding the vital inter-scale information that is key to diagnoses by human pathologists. In this study, we propose a novel cross-scale MIL algorithm to explicitly aggregate inter-scale relationships into a single MIL network for pathological image diagnosis. The contribution of this paper is three-fold: (1) A novel cross-scale MIL (CS-MIL) algorithm that integrates the multi-scale information and the inter-scale relationships is proposed; (2) A toy dataset with scale-specific morphological features is created and released to examine and visualize differential cross-scale attention; (3) Superior performance on both in-house and public datasets is demonstrated by our simple cross-scale MIL strategy. The official implementation is publicly available at https://github.com/hrlblab/CS-MIL.

Inter-vendor harmonization of CT reconstruction kernels using unpaired image translation.

The reconstruction kernel in computed tomography (CT) generation determines the texture of the image. Consistency in reconstruction kernels is important as the underlying CT texture can impact measurements during quantitative image analysis. Harmonization (i.e., kernel conversion) minimizes differences in measurements due to inconsistent reconstruction kernels. Existing methods investigate harmonization of CT scans in single or multiple manufacturers. However, these methods require paired scans of hard and soft reconstruction kernels that are spatially and anatomically aligned. Additionally, a large number of models need to be trained across different kernel pairs within manufacturers. In this study, we adopt an unpaired image translation approach to investigate harmonization between and across reconstruction kernels from different manufacturers by constructing a multipath cycle generative adversarial network (GAN). We use hard and soft reconstruction kernels from the Siemens and GE vendors from the National Lung Screening Trial dataset. We use 50 scans from each reconstruction kernel and train a multipath cycle GAN. To evaluate the effect of harmonization on the reconstruction kernels, we harmonize 50 scans each from Siemens hard kernel, GE soft kernel and GE hard kernel to a reference Siemens soft kernel (B30f) and evaluate percent emphysema. We fit a linear model by considering the age, smoking status, sex and vendor and perform an analysis of variance (ANOVA) on the emphysema scores. Our approach minimizes differences in emphysema measurement and highlights the impact of age, sex, smoking status and vendor on emphysema quantification.

Harmonized connectome resampling for variance in voxel sizes.

To date, there has been no comprehensive study characterizing the effect of diffusion-weighted magnetic resonance imaging voxel resolution on the resulting connectome for high resolution subject data. Similarity in results improved with higher resolution, even after initial down-sampling. To ensure robust tractography and connectomes, resample data to 1 mm isotropic resolution.

Characterizing patterns of DTI variance in aging brains.

Background: As large analyses merge data across sites, a deeper understanding of variance in statistical assessment across the sources of data becomes critical for valid analyses. Diffusion tensor imaging (DTI) exhibits spatially varying and correlated noise, so care must be taken with distributional assumptions. Purpose: We characterize the role of physiology, subject compliance, and the interaction of subject with the scanner in the understanding of DTI variability, as modeled in spatial variance of derived metrics in homogeneous regions. Methods: We analyze DTI data from 1035 subjects in the Baltimore Longitudinal Study of Aging (BLSA), with ages ranging from 22.4 to 103 years old. For each subject, up to 12 longitudinal sessions were conducted. We assess variance of DTI scalars within regions of interest (ROIs) defined by four segmentation methods and investigate the relationships between the variance and covariates, including baseline age, time from the baseline (referred to as "interval"), motion, sex, and whether it is the first scan or the second scan in the session. Results: Covariate effects are heterogeneous and bilaterally symmetric across ROIs. Inter-session interval is positively related (p ≪ 0.001) to FA variance in the cuneus and occipital gyrus, but negatively (p ≪ 0.001) in the caudate nucleus. Males show significantly (p ≪ 0.001) higher FA variance in the right putamen, thalamus, body of the corpus callosum, and cingulate gyrus. In 62 out of 176 ROIs defined by the Eve type-1 atlas, an increase in motion is associated (p < 0.05) with a decrease in FA variance. Head motion increases during the rescan of DTI (Δµ = 0.045 millimeters per volume). Conclusions: The effects of each covariate on DTI variance, and their relationships across ROIs are complex. Ultimately, we encourage researchers to include estimates of variance when sharing data and consider models of heteroscedasticity in analysis. This work provides a foundation for study planning to account for regional variations in metric variance.

Alleviating tiling effect by random walk sliding window in high-resolution histological whole slide image synthesis.

Multiplex immunofluorescence (MxIF) is an advanced molecular imaging technique that can simultaneously provide biologists with multiple (i.e., more than 20) molecular markers on a single histological tissue section. Unfortunately, due to imaging restrictions, the more routinely used hematoxylin and eosin (H&E) stain is typically unavailable with MxIF on the same tissue section. As biological H&E staining is not feasible, previous efforts have been made to obtain H&E whole slide image (WSI) from MxIF via deep learning empowered virtual staining. However, the tiling effect is a long-lasting problem in high-resolution WSI-wise synthesis. The MxIF to H&E synthesis is no exception. Limited by computational resources, the cross-stain image synthesis is typically performed at the patch-level. Thus, discontinuous intensities might be visually identified along with the patch boundaries assembling all individual patches back to a WSI. In this work, we propose a deep learning based unpaired high-resolution image synthesis method to obtain virtual H&E WSIs from MxIF WSIs (each with 27 markers/stains) with reduced tiling effects. Briefly, we first extend the CycleGAN framework by adding simultaneous nuclei and mucin segmentation supervision as spatial constraints. Then, we introduce a random walk sliding window shifting strategy during the optimized inference stage, to alleviate the tiling effects. The validation results show that our spatially constrained synthesis method achieves a 56% performance gain for the downstream cell segmentation task. The proposed inference method reduces the tiling effects by using 50% fewer computation resources without compromising performance. The proposed random sliding window inference method is a plug-and-play module, which can be generalized for other high-resolution WSI image synthesis applications. The source code with our proposed model are available at https://github.com/MASILab/RandomWalkSlidingWindow.git.

Nucleus subtype classification using inter-modality learning.

Understanding the way cells communicate, co-locate, and interrelate is essential to understanding human physiology. Hematoxylin and eosin (H&E) staining is ubiquitously available both for clinical studies and research. The Colon Nucleus Identification and Classification (CoNIC) Challenge has recently innovated on robust artificial intelligence labeling of six cell types on H&E stains of the colon. However, this is a very small fraction of the number of potential cell classification types. Specifically, the CoNIC Challenge is unable to classify epithelial subtypes (progenitor, endocrine, goblet), lymphocyte subtypes (B, helper T, cytotoxic T), or connective subtypes (fibroblasts, stromal). In this paper, we propose to use inter-modality learning to label previously un-labelable cell types on virtual H&E. We leveraged multiplexed immunofluorescence (MxIF) histology imaging to identify 14 subclasses of cell types. We performed style transfer to synthesize virtual H&E from MxIF and transferred the higher density labels from MxIF to these virtual H&E images. We then evaluated the efficacy of learning in this approach. We identified helper T and progenitor nuclei with positive predictive values of 0.34 ± 0.15 (prevalence 0.03 ± 0.01) and 0.47 ± 0.1 (prevalence 0.07 ± 0.02) respectively on virtual H&E. This approach represents a promising step towards automating annotation in digital pathology.

Cell Spatial Analysis in Crohn's Disease: Unveiling Local Cell Arrangement Pattern with Graph-based Signatures.

Crohn's disease (CD) is a chronic and relapsing inflammatory condition that affects segments of the gastrointestinal tract. CD activity is determined by histological findings, particularly the density of neutrophils observed on Hematoxylin and Eosin stains (H&E) imaging. However, understanding the broader morphometry and local cell arrangement beyond cell counting and tissue morphology remains challenging. To address this, we characterize six distinct cell types from H&E images and develop a novel approach for the local spatial signature of each cell. Specifically, we create a 10-cell neighborhood matrix, representing neighboring cell arrangements for each individual cell. Utilizing t-SNE for non-linear spatial projection in scatter-plot and Kernel Density Estimation contour-plot formats, our study examines patterns of differences in the cellular environment associated with the odds ratio of spatial patterns between active CD and control groups. This analysis is based on data collected at the two research institutes. The findings reveal heterogeneous nearest-neighbor patterns, signifying distinct tendencies of cell clustering, with a particular focus on the rectum region. These variations underscore the impact of data heterogeneity on cell spatial arrangements in CD patients. Moreover, the spatial distribution disparities between the two research sites highlight the significance of collaborative efforts among healthcare organizations. All research analysis pipeline tools are available at https://github.com/MASILab/cellNN.

Evaluation of Mean Shift, ComBat, and CycleGAN for Harmonizing Brain Connectivity Matrices Across Sites.

Connectivity matrices derived from diffusion MRI (dMRI) provide an interpretable and generalizable way of understanding the human brain connectome. However, dMRI suffers from inter-site and between-scanner variation, which impedes analysis across datasets to improve robustness and reproducibility of results. To evaluate different harmonization approaches on connectivity matrices, we compared graph measures derived from these matrices before and after applying three harmonization techniques: mean shift, ComBat, and CycleGAN. The sample comprises 168 age-matched, sex-matched normal subjects from two studies: the Vanderbilt Memory and Aging Project (VMAP) and the Biomarkers of Cognitive Decline Among Normal Individuals (BIOCARD). First, we plotted the graph measures and used coefficient of variation (CoV) and the Mann-Whitney U test to evaluate different methods' effectiveness in removing site effects on the matrices and the derived graph measures. ComBat effectively eliminated site effects for global efficiency and modularity and outperformed the other two methods. However, all methods exhibited poor performance when harmonizing average betweenness centrality. Second, we tested whether our harmonization methods preserved correlations between age and graph measures. All methods except for CycleGAN in one direction improved correlations between age and global efficiency and between age and modularity from insignificant to significant with p-values less than 0.05.

Characterizing patterns of diffusion tensor imaging variance in aging brains.

Purpose: As large analyses merge data across sites, a deeper understanding of variance in statistical assessment across the sources of data becomes critical for valid analyses. Diffusion tensor imaging (DTI) exhibits spatially varying and correlated noise, so care must be taken with distributional assumptions. Here, we characterize the role of physiology, subject compliance, and the interaction of the subject with the scanner in the understanding of DTI variability, as modeled in the spatial variance of derived metrics in homogeneous regions. Approach: We analyze DTI data from 1035 subjects in the Baltimore Longitudinal Study of Aging, with ages ranging from 22.4 to 103 years old. For each subject, up to 12 longitudinal sessions were conducted. We assess the variance of DTI scalars within regions of interest (ROIs) defined by four segmentation methods and investigate the relationships between the variance and covariates, including baseline age, time from the baseline (referred to as "interval"), motion, sex, and whether it is the first scan or the second scan in the session. Results: Covariate effects are heterogeneous and bilaterally symmetric across ROIs. Inter-session interval is positively related ( p ≪ 0.001 ) to FA variance in the cuneus and occipital gyrus, but negatively ( p ≪ 0.001 ) in the caudate nucleus. Males show significantly ( p ≪ 0.001 ) higher FA variance in the right putamen, thalamus, body of the corpus callosum, and cingulate gyrus. In 62 out of 176 ROIs defined by the Eve type-1 atlas, an increase in motion is associated ( p < 0.05 ) with a decrease in FA variance. Head motion increases during the rescan of DTI ( Δ µ = 0.045 mm per volume). Conclusions: The effects of each covariate on DTI variance and their relationships across ROIs are complex. Ultimately, we encourage researchers to include estimates of variance when sharing data and consider models of heteroscedasticity in analysis. This work provides a foundation for study planning to account for regional variations in metric variance.

Evaluation of Mean Shift, ComBat, and CycleGAN for Harmonizing Brain Connectivity Matrices Across Sites.

Connectivity matrices derived from diffusion MRI (dMRI) provide an interpretable and generalizable way of understanding the human brain connectome. However, dMRI suffers from inter-site and between-scanner variation, which impedes analysis across datasets to improve robustness and reproducibility of results. To evaluate different harmonization approaches on connectivity matrices, we compared graph measures derived from these matrices before and after applying three harmonization techniques: mean shift, ComBat, and CycleGAN. The sample comprises 168 age-matched, sex-matched normal subjects from two studies: the Vanderbilt Memory and Aging Project (VMAP) and the Biomarkers of Cognitive Decline Among Normal Individuals (BIOCARD). First, we plotted the graph measures and used coefficient of variation (CoV) and the Mann-Whitney U test to evaluate different methods' effectiveness in removing site effects on the matrices and the derived graph measures. ComBat effectively eliminated site effects for global efficiency and modularity and outperformed the other two methods. However, all methods exhibited poor performance when harmonizing average betweenness centrality. Second, we tested whether our harmonization methods preserved correlations between age and graph measures. All methods except for CycleGAN in one direction improved correlations between age and global efficiency and between age and modularity from insignificant to significant with p-values less than 0.05.

Semantic-Aware Contrastive Learning for Multi-Object Medical Image Segmentation.

Medical image segmentation, or computing voxel-wise semantic masks, is a fundamental yet challenging task in medical imaging domain. To increase the ability of encoder-decoder neural networks to perform this task across large clinical cohorts, contrastive learning provides an opportunity to stabilize model initialization and enhances downstream tasks performance without ground-truth voxel-wise labels. However, multiple target objects with different semantic meanings and contrast level may exist in a single image, which poses a problem for adapting traditional contrastive learning methods from prevalent "image-level classification" to "pixel-level segmentation". In this article, we propose a simple semantic-aware contrastive learning approach leveraging attention masks and image-wise labels to advance multi-object semantic segmentation. Briefly, we embed different semantic objects to different clusters rather than the traditional image-level embeddings. We evaluate our proposed method on a multi-organ medical image segmentation task with both in-house data and MICCAI Challenge 2015 BTCV datasets. Compared with current state-of-the-art training strategies, our proposed pipeline yields a substantial improvement of 5.53% and 6.09% on Dice score for both medical image segmentation cohorts respectively (p-value 0.01). The performance of the proposed method is further assessed on external medical image cohort via MICCAI Challenge FLARE 2021 dataset, and achieves a substantial improvement from Dice 0.922 to 0.933 (p-value 0.01).

Influence of Cell-type Ratio on Spatially Resolved Single-cell Transcriptomes using the Tangram Algorithm: Based on Implementation on Single-Cell and MxIF Data.

The Tangram algorithm is a benchmarking method of aligning single-cell (sc/snRNA-seq) data to various forms of spatial data collected from the same region. With this data alignment, the annotation of the single-cell data can be projected to spatial data. However, the cell composition (cell-type ratio) of the single-cell data and spatial data might be different because of heterogeneous cell distribution. Whether the Tangram algorithm can be adapted when the two data have different cell-type ratios has not been discussed in previous works. In our practical application that maps the cell-type classification results of single-cell data to the Multiplex immunofluorescence (MxIF) spatial data, cell-type ratios were different, though they were sampled from adjacent areas. In this work, both simulation and empirical validation were conducted to quantitatively explore the impact of the mismatched cell-type ratio on the Tangram mapping in different situations. Results show that the cell-type difference has a negative influence on classification accuracy.

Feasibility of Universal Anomaly Detection without Knowing the Abnormality in Medical Images.

Many anomaly detection approaches, especially deep learning methods, have been recently developed to identify abnormal image morphology by only employing normal images during training. Unfortunately, many prior anomaly detection methods were optimized for a specific "known" abnormality (e.g., brain tumor, bone fraction, cell types). Moreover, even though only the normal images were used in the training process, the abnormal images were often employed during the validation process (e.g., epoch selection, hyper-parameter tuning), which might leak the supposed "unknown" abnormality unintentionally. In this study, we investigated these two essential aspects regarding universal anomaly detection in medical images by (1) comparing various anomaly detection methods across four medical datasets, (2) investigating the inevitable but often neglected issues on how to unbiasedly select the optimal anomaly detection model during the validation phase using only normal images, and (3) proposing a simple decision-level ensemble method to leverage the advantage of different kinds of anomaly detection without knowing the abnormality. The results of our experiments indicate that none of the evaluated methods consistently achieved the best performance across all datasets. Our proposed method enhanced the robustness of performance in general (average AUC 0.956).

Democratizing Pathological Image Segmentation with Lay Annotators via Molecular-empowered Learning.

Multi-class cell segmentation in high-resolution Giga-pixel whole slide images (WSI) is critical for various clinical applications. Training such an AI model typically requires labor-intensive pixel-wise manual annotation from experienced domain experts (e.g., pathologists). Moreover, such annotation is error-prone when differentiating fine-grained cell types (e.g., podocyte and mesangial cells) via the naked human eye. In this study, we assess the feasibility of democratizing pathological AI deployment by only using lay annotators (annotators without medical domain knowledge). The contribution of this paper is threefold: (1) We proposed a molecular-empowered learning scheme for multi-class cell segmentation using partial labels from lay annotators; (2) The proposed method integrated Giga-pixel level molecular-morphology cross-modality registration, molecular-informed annotation, and molecular-oriented segmentation model, so as to achieve significantly superior performance via 3 lay annotators as compared with 2 experienced pathologists; (3) A deep corrective learning (learning with imperfect label) method is proposed to further improve the segmentation performance using partially annotated noisy data. From the experimental results, our learning method achieved F1 = 0.8496 using molecular-informed annotations from lay annotators, which is better than conventional morphology-based annotations (F1 = 0.7015) from experienced pathologists. Our method democratizes the development of a pathological segmentation deep model to the lay annotator level, which consequently scales up the learning process similar to a non-medical computer vision task. The official implementation and cell annotations are publicly available at https://github.com/hrlblab/MolecularEL.

Topological-Preserving Membrane Skeleton Segmentation in Multiplex Immunofluorescence Imaging.

Multiplex immunofluorescence (MxIF) is an emerging imaging technology whose downstream molecular analytics highly rely upon the effectiveness of cell segmentation. In practice, multiple membrane markers (e.g., NaKATPase, PanCK and ß-catenin) are employed to stain membranes for different cell types, so as to achieve a more comprehensive cell segmentation since no single marker fits all cell types. However, prevalent watershed-based image processing might yield inferior capability for modeling complicated relationships between markers. For example, some markers can be misleading due to questionable stain quality. In this paper, we propose a deep learning based membrane segmentation method to aggregate complementary information that is uniquely provided by large scale MxIF markers. We aim to segment tubular membrane structure in MxIF data using global (membrane markers z-stack projection image) and local (separate individual markers) information to maximize topology preservation with deep learning. Specifically, we investigate the feasibility of four SOTA 2D deep networks and four volumetric-based loss functions. We conducted a comprehensive ablation study to assess the sensitivity of the proposed method with various combinations of input channels. Beyond using adjusted rand index (ARI) as the evaluation metric, which was inspired by the clDice, we propose a novel volumetric metric that is specific for skeletal structure, denoted as clDiceSKEL. In total, 80 membrane MxIF images were manually traced for 5-fold cross-validation. Our model outperforms the baseline with a 20.2% and 41.3% increase in clDiceSKEL and ARI performance, which is significant (p<0.05) using the Wilcoxon signed rank test. Our work explores a promising direction for advancing MxIF imaging cell segmentation with deep learning membrane segmentation. Tools are available at https://github.com/MASILab/MxIF_Membrane_Segmentation.

Predicting Crohn's disease severity in the colon using mixed cell nucleus density from pseudo labels.

Crohn's disease (CD) is a debilitating inflammatory bowel disease with no known cure. Computational analysis of hematoxylin and eosin (H&E) stained colon biopsy whole slide images (WSIs) from CD patients provides the opportunity to discover unknown and complex relationships between tissue cellular features and disease severity. While there have been works using cell nuclei-derived features for predicting slide-level traits, this has not been performed on CD H&E WSIs for classifying normal tissue from CD patients vs active CD and assessing slide label-predictive performance while using both separate and combined information from pseudo-segmentation labels of nuclei from neutrophils, eosinophils, epithelial cells, lymphocytes, plasma cells, and connective cells. We used 413 WSIs of CD patient biopsies and calculated normalized histograms of nucleus density for the six cell classes for each WSI. We used a support vector machine to classify the truncated singular value decomposition representations of the normalized histograms as normal or active CD with four-fold cross-validation in rounds where nucleus types were first compared individually, the best was selected, and further types were added each round. We found that neutrophils were the most predictive individual nucleus type, with an AUC of 0.92 ± 0.0003 on the withheld test set. Adding information improved cross-validation performance for the first two rounds and on the withheld test set for the first three rounds, though performance metrics did not increase substantially beyond when neutrophils were used alone.

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.