A Multi-Atlas Based Method for Automated Anatomical Rat Brain MRI Segmentation and Extraction of PET Activity

Introduction

Preclinical in vivo imaging requires precise and reproducible delineation of brain structures. Manual segmentation is time consuming and operator dependent. Automated segmentation as usually performed via single atlas registration fails to account for anatomo-physiological variability. We present, evaluate, and make available a multi-atlas approach for automatically segmenting rat brain MRI and extracting PET activies.

Methods

High-resolution 7T 2DT2 MR images of 12 Sprague-Dawley rat brains were manually segmented into 27-VOI label volumes using detailed protocols. Automated methods were developed with 7/12 atlas datasets, i.e. the MRIs and their associated label volumes. MRIs were registered to a common space, where an MRI template and a maximum probability atlas were created. Three automated methods were tested: 1/registering individual MRIs to the template, and using a single atlas (SA), 2/using the maximum probability atlas (MP), and 3/registering the MRIs from the multi-atlas dataset to an individual MRI, propagating the label volumes and fusing them in individual MRI space (propagation & fusion, PF). Evaluation was performed on the five remaining rats which additionally underwent [¹⁸F]FDG PET. Automated and manual segmentations were compared for morphometric performance (assessed by comparing volume bias and Dice overlap index) and functional performance (evaluated by comparing extracted PET measures).

Results

Only the SA method showed volume bias. Dice indices were significantly different between methods (PF>MP>SA). PET regional measures were more accurate with multi-atlas methods than with SA method.

Conclusions

Multi-atlas methods outperform SA for automated anatomical brain segmentation and PET measure’s extraction. They perform comparably to manual segmentation for FDG-PET quantification. Multi-atlas methods are suitable for rapid reproducible VOI analyses.     

A high throughput machine-learning driven analysis of Ca2+ spatio-temporal maps

High-resolution Ca²⁺ imaging to study cellular Ca²⁺ behaviors has led to the creation of large datasets with a profound need for standardized and accurate analysis. To analyze these datasets, spatio-temporal maps (STMaps) that allow for 2D visualization of Ca²⁺ signals as a function of time and space are often used. Methods of STMap analysis rely on a highly arduous process of user defined segmentation and event-based data retrieval. These methods are often time consuming, lack accuracy, and are extremely variable between users. We designed a novel automated machine-learning based plugin for the analysis of Ca²⁺ STMaps (STMapAuto). The plugin includes optimized tools for Ca²⁺ signal preprocessing, automated segmentation, and automated extraction of key Ca²⁺ event information such as duration, spatial spread, frequency, propagation angle, and intensity in a variety of cell types including the Interstitial cells of Cajal (ICC). The plugin is fully implemented in Fiji and able to accurately detect and expeditiously quantify Ca²⁺ transient parameters from ICC. The plugin’s speed of analysis of large-datasets was 197-fold faster than the commonly used single pixel-line method of analysis. The automated machine-learning based plugin described dramatically reduces opportunities for user error and provides a consistent method to allow high-throughput analysis of STMap datasets.     

Multi-Modal Glioblastoma Segmentation: Man versus Machine

Background and Purpose

Reproducible segmentation of brain tumors on magnetic resonance images is an important clinical need. This study was designed to evaluate the reliability of a novel fully automated segmentation tool for brain tumor image analysis in comparison to manually defined tumor segmentations.

Methods

We prospectively evaluated preoperative MR Images from 25 glioblastoma patients. Two independent expert raters performed manual segmentations. Automatic segmentations were performed using the Brain Tumor Image Analysis software (BraTumIA). In order to study the different tumor compartments, the complete tumor volume TV (enhancing part plus non-enhancing part plus necrotic core of the tumor), the TV+ (TV plus edema) and the contrast enhancing tumor volume CETV were identified. We quantified the overlap between manual and automated segmentation by calculation of diameter measurements as well as the Dice coefficients, the positive predictive values, sensitivity, relative volume error and absolute volume error.

Results

Comparison of automated versus manual extraction of 2-dimensional diameter measurements showed no significant difference (p = 0.29). Comparison of automated versus manual segmentation of volumetric segmentations showed significant differences for TV+ and TV (p<0.05) but no significant differences for CETV (p>0.05) with regard to the Dice overlap coefficients. Spearman''s rank correlation coefficients (ρ) of TV+, TV and CETV showed highly significant correlations between automatic and manual segmentations. Tumor localization did not influence the accuracy of segmentation.

Conclusions

In summary, we demonstrated that BraTumIA supports radiologists and clinicians by providing accurate measures of cross-sectional diameter-based tumor extensions. The automated volume measurements were comparable to manual tumor delineation for CETV tumor volumes, and outperformed inter-rater variability for overlap and sensitivity.     

Development and Evaluation of an Algorithm for the Computer-Assisted Segmentation of the Human Hypothalamus on 7-Tesla Magnetic Resonance Images

Post mortem studies have shown volume changes of the hypothalamus in psychiatric patients. With 7T magnetic resonance imaging this effect can now be investigated in vivo in detail. To benefit from the sub-millimeter resolution requires an improved segmentation procedure. The traditional anatomical landmarks of the hypothalamus were refined using 7T T1-weighted magnetic resonance images. A detailed segmentation algorithm (unilateral hypothalamus) was developed for colour-coded, histogram-matched images, and evaluated in a sample of 10 subjects. Test-retest and inter-rater reliabilities were estimated in terms of intraclass-correlation coefficients (ICC) and Dice''s coefficient (DC). The computer-assisted segmentation algorithm ensured test-retest reliabilities of ICC≥.97 (DC≥96.8) and inter-rater reliabilities of ICC≥.94 (DC = 95.2). There were no significant volume differences between the segmentation runs, raters, and hemispheres. The estimated volumes of the hypothalamus lie within the range of previous histological and neuroimaging results. We present a computer-assisted algorithm for the manual segmentation of the human hypothalamus using T1-weighted 7T magnetic resonance imaging. Providing very high test-retest and inter-rater reliabilities, it outperforms former procedures established at 1.5T and 3T magnetic resonance images and thus can serve as a gold standard for future automated procedures.     

RIEMS: a software pipeline for sensitive and comprehensive taxonomic classification of reads from metagenomics datasets

Background

Fuelled by the advent and subsequent development of next generation sequencing technologies, metagenomics became a powerful tool for the analysis of microbial communities both scientifically and diagnostically. The biggest challenge is the extraction of relevant information from the huge sequence datasets generated for metagenomics studies. Although a plethora of tools are available, data analysis is still a bottleneck.

Results

To overcome the bottleneck of data analysis, we developed an automated computational workflow called RIEMS – Reliable Information Extraction from Metagenomic Sequence datasets. RIEMS assigns every individual read sequence within a dataset taxonomically by cascading different sequence analyses with decreasing stringency of the assignments using various software applications. After completion of the analyses, the results are summarised in a clearly structured result protocol organised taxonomically. The high accuracy and performance of RIEMS analyses were proven in comparison with other tools for metagenomics data analysis using simulated sequencing read datasets.

Conclusions

RIEMS has the potential to fill the gap that still exists with regard to data analysis for metagenomics studies. The usefulness and power of RIEMS for the analysis of genuine sequencing datasets was demonstrated with an early version of RIEMS in 2011 when it was used to detect the orthobunyavirus sequences leading to the discovery of Schmallenberg virus.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-015-0503-6) contains supplementary material, which is available to authorized users.     

Heterogeneous Ensemble Combination Search Using Genetic Algorithm for Class Imbalanced Data Classification

Classification of datasets with imbalanced sample distributions has always been a challenge. In general, a popular approach for enhancing classification performance is the construction of an ensemble of classifiers. However, the performance of an ensemble is dependent on the choice of constituent base classifiers. Therefore, we propose a genetic algorithm-based search method for finding the optimum combination from a pool of base classifiers to form a heterogeneous ensemble. The algorithm, called GA-EoC, utilises 10 fold-cross validation on training data for evaluating the quality of each candidate ensembles. In order to combine the base classifiers decision into ensemble’s output, we used the simple and widely used majority voting approach. The proposed algorithm, along with the random sub-sampling approach to balance the class distribution, has been used for classifying class-imbalanced datasets. Additionally, if a feature set was not available, we used the (α, β) − k Feature Set method to select a better subset of features for classification. We have tested GA-EoC with three benchmarking datasets from the UCI-Machine Learning repository, one Alzheimer’s disease dataset and a subset of the PubFig database of Columbia University. In general, the performance of the proposed method on the chosen datasets is robust and better than that of the constituent base classifiers and many other well-known ensembles. Based on our empirical study we claim that a genetic algorithm is a superior and reliable approach to heterogeneous ensemble construction and we expect that the proposed GA-EoC would perform consistently in other cases.     

Predicting Future Morphological Changes of Lesions from Radiotracer Uptake in 18F-FDG-PET Images

We introduce a novel computational framework to enable automated identification of texture and shape features of lesions on ¹⁸F-FDG-PET images through a graph-based image segmentation method. The proposed framework predicts future morphological changes of lesions with high accuracy. The presented methodology has several benefits over conventional qualitative and semi-quantitative methods, due to its fully quantitative nature and high accuracy in each step of (i) detection, (ii) segmentation, and (iii) feature extraction. To evaluate our proposed computational framework, thirty patients received 2 ¹⁸F-FDG-PET scans (60 scans total), at two different time points. Metastatic papillary renal cell carcinoma, cerebellar hemongioblastoma, non-small cell lung cancer, neurofibroma, lymphomatoid granulomatosis, lung neoplasm, neuroendocrine tumor, soft tissue thoracic mass, nonnecrotizing granulomatous inflammation, renal cell carcinoma with papillary and cystic features, diffuse large B-cell lymphoma, metastatic alveolar soft part sarcoma, and small cell lung cancer were included in this analysis. The radiotracer accumulation in patients'' scans was automatically detected and segmented by the proposed segmentation algorithm. Delineated regions were used to extract shape and textural features, with the proposed adaptive feature extraction framework, as well as standardized uptake values (SUV) of uptake regions, to conduct a broad quantitative analysis. Evaluation of segmentation results indicates that our proposed segmentation algorithm has a mean dice similarity coefficient of 85.75±1.75%. We found that 28 of 68 extracted imaging features were correlated well with SUV_max (p<0.05), and some of the textural features (such as entropy and maximum probability) were superior in predicting morphological changes of radiotracer uptake regions longitudinally, compared to single intensity feature such as SUV_max. We also found that integrating textural features with SUV measurements significantly improves the prediction accuracy of morphological changes (Spearman correlation coefficient = 0.8715, p<2e-16).     

COVASIAM: an Image Analysis Method That Allows Detection of Confluent Microbial Colonies and Colonies of Various Sizes for Automated Counting

In this work we introduce the confluent and various sizes image analysis method (COVASIAM), an automated colony count technique that uses digital imaging technology for detection and separation of confluent microbial colonies and colonies of various sizes growing on petri dishes. The proposed method takes advantage of the optical properties of the surfaces of most microbial colonies. Colonies in the petri dish are epi-illuminated in order to direct the reflection of concentrated light coming from a halogen lamp towards an image-sensing device. In conjunction, a multilevel threshold algorithm is proposed for colony separation and counting. These procedures improved the quantification of colonies showing confluence or differences in size. We tested COVASIAM with a sample set of microorganisms that form colonies with contrasting physical properties: Saccharomyces cerevisiae, Aspergillus nidulans, Escherichia coli, Azotobacter vinelandii, Pseudomonas aeruginosa, and Rhizobium etli. These physical properties range from smooth to hairy, from bright to opaque, and from high to low convexities. COVASIAM estimated an average of 95.47% (ς = 8.55%) of the manually counted colonies, while an automated method based on a single-threshold segmentation procedure estimated an average of 76% (ς = 16.27) of the manually counted colonies. This method can be easily transposed to almost every image-processing analyzer since the procedures to compile it are generically standard.     

Extra-Hippocampal Subcortical Limbic Involvement Predicts Episodic Recall Performance in Multiple Sclerosis

Background

Episodic memory impairment is a common but poorly-understood phenomenon in multiple sclerosis (MS). We aim to establish the relative contributions of reduced integrity of components of the extended hippocampal-diencephalic system to memory performance in MS patients using quantitative neuroimaging.

Methodology/Principal Findings

34 patients with relapsing-remitting MS and 24 healthy age-matched controls underwent 3 T MRI including diffusion tensor imaging and 3-D T1-weighted volume acquisition. Manual fornix regions-of-interest were used to derive fornix fractional anisotropy (FA). Normalized hippocampal, mammillary body and thalamic volumes were derived by manual segmentation. MS subjects underwent visual recall, verbal recall, verbal recognition and verbal fluency assessment. Significant differences between MS patients and controls were found for fornix FA (0.38 vs. 0.46, means adjusted for age and fornix volume, P<.0005) and mammillary body volumes (age-adjusted means 0.114 ml vs. 0.126 ml, P<.023). Multivariate regression analysis identified fornix FA and mammillary bodies as predictor of visual recall (R² = .31, P = .003, P = .006), and thalamic volume as predictive of verbal recall (R² = .37, P<.0005). No limbic measures predicted verbal recognition or verbal fluency.

Conclusions/Significance

These findings indicate that structural and ultrastructural alterations in subcortical limbic components beyond the hippocampus predict performance of episodic recall in MS patients with mild memory dysfunction.     

Continuous detection of weak sensory signals in afferent spike trains: the role of anti-correlated interspike intervals in detection performance

An important problem in sensory processing is deciding whether fluctuating neural activity encodes a stimulus or is due to variability in baseline activity. Neurons that subserve detection must examine incoming spike trains continuously, and quickly and reliably differentiate signals from baseline activity. Here we demonstrate that a neural integrator can perform continuous signal detection, with performance exceeding that of trial-based procedures, where spike counts in signal- and baseline windows are compared. The procedure was applied to data from electrosensory afferents of weakly electric fish (Apteronotus leptorhynchus), where weak perturbations generated by small prey add ~1 spike to a baseline of ~300 spikes s^–1. The hypothetical postsynaptic neuron, modeling an electrosensory lateral line lobe cell, could detect an added spike within 10–15 ms, achieving near ideal detection performance (80–95%) at false alarm rates of 1–2 Hz, while trial-based testing resulted in only 30–35% correct detections at that false alarm rate. The performance improvement was due to anti-correlations in the afferent spike train, which reduced both the amplitude and duration of fluctuations in postsynaptic membrane activity, and so decreased the number of false alarms. Anti-correlations can be exploited to improve detection performance only if there is memory of prior decisions.Abbreviations B binomial - CV coefficient of variation - EOD electric organ discharge - ELL electrosensory lateral line lobe - EPSP excitatory postsynaptic potential - ISI interspike interval - M₀ Markov order zero - M₁ Markov order one - N noise - OC operating characteristic - PDF probability density function - ROC receiver operating characteristic - S signal - SNR signal-to-noise ratio - S+N signal in noise     

Three-dimensional segmentation and depth-encoded visualization of choroidal vasculature using swept-source optical coherence tomography

The choroid provides nutritional support for the retinal pigment epithelium and photoreceptors. Choroidal dysfunction plays a major role in several of the most important causes of vision loss including age-related macular degeneration, myopic degeneration, and pachychoroid diseases such as central serous chorioretinopathy and polypoidal choroidal vasculopathy. We describe an imaging technique using depth-resolved swept-source optical coherence tomography (SS-OCT) that provides full-thickness three-dimensional (3D) visualization of choroidal anatomy including topographical features of individual vessels. Enrolled subjects with different clinical manifestations within the pachychoroid disease spectrum underwent 15 mm × 9 mm volume scans centered on the fovea. A fully automated method segmented the choroidal vessels using their hyporeflective lumens. Binarized choroidal vessels were rendered in a 3D viewer as a vascular network within a choroidal slab. The network of choroidal vessels was color depth-encoded with a reference to the Bruch’s membrane segmentation. Topographical features of the choroidal vasculature were characterized and compared with choroidal imaging obtained with indocyanine green angiography (ICGA) from the same subject. The en face SS-OCT projections of the larger choroid vessels closely resembled to that obtained with ICGA, with the automated SS-OCT approach proving additional depth-encoded 3D information. In 16 eyes with pachychoroid disease, the SS-OCT approach added clinically relevant structural details, including choroidal thickness and vessel depth, which the ICGA studies could not provide. Our technique appears to advance the in vivo visualization of the full-thickness choroid, successfully reveals the topographical features of choroidal vasculature, and shows potential for further quantitative analysis when compared with other choroidal imaging techniques. This improved visualization of choroidal vasculature and its 3D structure should provide an insight into choroid-related disease mechanisms as well as their responses to treatment.     

An Automated Three-Dimensional Detection and Segmentation Method for Touching Cells by Integrating Concave Points Clustering and Random Walker Algorithm

Characterizing cytoarchitecture is crucial for understanding brain functions and neural diseases. In neuroanatomy, it is an important task to accurately extract cell populations'' centroids and contours. Recent advances have permitted imaging at single cell resolution for an entire mouse brain using the Nissl staining method. However, it is difficult to precisely segment numerous cells, especially those cells touching each other. As presented herein, we have developed an automated three-dimensional detection and segmentation method applied to the Nissl staining data, with the following two key steps: 1) concave points clustering to determine the seed points of touching cells; and 2) random walker segmentation to obtain cell contours. Also, we have evaluated the performance of our proposed method with several mouse brain datasets, which were captured with the micro-optical sectioning tomography imaging system, and the datasets include closely touching cells. Comparing with traditional detection and segmentation methods, our approach shows promising detection accuracy and high robustness.     

Automated Visual Cognitive Tasks for Recording Neural Activity Using a Floor Projection Maze

Neuropsychological tasks used in primates to investigate mechanisms of learning and memory are typically visually guided cognitive tasks. We have developed visual cognitive tasks for rats using the Floor Projection Maze^1,2 that are optimized for visual abilities of rats permitting stronger comparisons of experimental findings with other species.In order to investigate neural correlates of learning and memory, we have integrated electrophysiological recordings into fully automated cognitive tasks on the Floor Projection Maze^1,2. Behavioral software interfaced with an animal tracking system allows monitoring of the animal''s behavior with precise control of image presentation and reward contingencies for better trained animals. Integration with an in vivo electrophysiological recording system enables examination of behavioral correlates of neural activity at selected epochs of a given cognitive task.We describe protocols for a model system that combines automated visual presentation of information to rodents and intracranial reward with electrophysiological approaches. Our model system offers a sophisticated set of tools as a framework for other cognitive tasks to better isolate and identify specific mechanisms contributing to particular cognitive processes.     

Multi-Environment Model Estimation for Motility Analysis of Caenorhabditis elegans

The nematode Caenorhabditis elegans is a well-known model organism used to investigate fundamental questions in biology. Motility assays of this small roundworm are designed to study the relationships between genes and behavior. Commonly, motility analysis is used to classify nematode movements and characterize them quantitatively. Over the past years, C. elegans'' motility has been studied across a wide range of environments, including crawling on substrates, swimming in fluids, and locomoting through microfluidic substrates. However, each environment often requires customized image processing tools relying on heuristic parameter tuning. In the present study, we propose a novel Multi-Environment Model Estimation (MEME) framework for automated image segmentation that is versatile across various environments. The MEME platform is constructed around the concept of Mixture of Gaussian (MOG) models, where statistical models for both the background environment and the nematode appearance are explicitly learned and used to accurately segment a target nematode. Our method is designed to simplify the burden often imposed on users; here, only a single image which includes a nematode in its environment must be provided for model learning. In addition, our platform enables the extraction of nematode ‘skeletons’ for straightforward motility quantification. We test our algorithm on various locomotive environments and compare performances with an intensity-based thresholding method. Overall, MEME outperforms the threshold-based approach for the overwhelming majority of cases examined. Ultimately, MEME provides researchers with an attractive platform for C. elegans'' segmentation and ‘skeletonizing’ across a wide range of motility assays.     

Optimization and comparison of two methods for spike train estimation in an unfused tetanic contraction of low threshold motor units

BackgroundRecent findings have shown that imaging voluntarily activated motor units (MUs) by decomposing ultrasound-based displacement images provides estimates of unfused tetanic signals evoked by spinal motoneurons’ neural discharges (spikes). Two methods have been suggested to estimate its spike trains: band-pass filter (BPM) and Haar wavelet transform (HWM). However, the methods’ optimal parameters and which method performs the best are unknown. This study will answer these questions.MethodHWM and BPM were optimized using simulations. Their performance was evaluated based on simulations and 21 experimental datasets, considering their rate of agreement, spike offset, and spike offset variability to the simulated or experimental spikes.ResultsA range of parameter sets that resulted in the highest possible agreement with simulated spikes was provided. Both methods highly agreed with simulated and experimental spikes, but HWM was a better spike estimation method than BPM because it had a higher agreement, less bias, and less variation (p < 0.001).ConclusionsThe optimized HWM will be an important contributor to further developing the identification and analysis of MUs using imaging, providing indirect access to the neural drive of the spinal cord to the muscle by the unfused tetanic signals.     

Quantitative Analysis of PiB-PET with FreeSurfer ROIs

In vivo quantification of β-amyloid deposition using positron emission tomography is emerging as an important procedure for the early diagnosis of the Alzheimer''s disease and is likely to play an important role in upcoming clinical trials of disease modifying agents. However, many groups use manually defined regions, which are non-standard across imaging centers. Analyses often are limited to a handful of regions because of the labor-intensive nature of manual region drawing. In this study, we developed an automatic image quantification protocol based on FreeSurfer, an automated whole brain segmentation tool, for quantitative analysis of amyloid images. Standard manual tracing and FreeSurfer-based analyses were performed in 77 participants including 67 cognitively normal individuals and 10 individuals with early Alzheimer''s disease. The manual and FreeSurfer approaches yielded nearly identical estimates of amyloid burden (intraclass correlation = 0.98) as assessed by the mean cortical binding potential. An MRI test-retest study demonstrated excellent reliability of FreeSurfer based regional amyloid burden measurements. The FreeSurfer-based analysis also revealed that the majority of cerebral cortical regions accumulate amyloid in parallel, with slope of accumulation being the primary difference between regions.     

Improving accuracy and efficiency in seagrass detection using state-of-the-art AI techniques

Seagrasses provide a wide range of ecosystem services in coastal marine environments. Despite their ecological and economic importance, these species are declining because of human impact. This decline has driven the need for monitoring and mapping to estimate the overall health and dynamics of seagrasses in coastal environments, often based on underwater images. However, seagrass detection from underwater digital images is not a trivial task; it requires taxonomic expertise and is time-consuming and expensive. Recently automatic approaches based on deep learning have revolutionised object detection performance in many computer vision applications, and there has been interest in applying this to automated seagrass detection from imagery. Deep learning–based techniques reduce the need for hardcore feature extraction by domain experts which is required in machine learning-based techniques. This study presents a YOLOv5-based one-stage detector and an EfficientDetD7–based two-stage detector for detecting seagrass, in this case, Halophila ovalis, one of the most widely distributed seagrass species. The EfficientDet-D7–based seagrass detector achieves the highest mAP of 0.484 on the ECUHO-2 dataset and mAP of 0.354 on the ECUHO-1 dataset, which are about 7% and 5% better than the state-of-the-art Halophila ovalis detection performance on those datasets, respectively. The proposed YOLOv5-based detector achieves an average inference time of 0.077 s and 0.043 s respectively which are much lower than the state-of-the-art approach on the same datasets.