skip to main content

Title: AxonEM Dataset: 3D Axon Instance Segmentation of Brain Cortical Regions
Electron microscopy (EM) enables the reconstruction of neural circuits at the level of individual synapses, which has been transformative for scientific discoveries. However, due to the complex morphology, an accurate reconstruction of cortical axons has become a major challenge. Worse still, there is no publicly available large-scale EM dataset from the cortex that provides dense ground truth segmentation for axons, making it difficult to develop and evaluate large-scale axon reconstruction methods. To address this, we introduce the AxonEM dataset, which consists of two 30x30x30 cubic mm EM image volumes from the human and mouse cortex, respectively. We thoroughly proofread over 18,000 axon instances to provide dense 3D axon instance segmentation, enabling large- scale evaluation of axon reconstruction methods. In addition, we densely annotate nine ground truth subvolumes for training, per each data volume. With this, we reproduce two published state-of-the-art methods and provide their evaluation results as a baseline. We publicly release our code and data at AxonEM/index.html to foster the development of advanced methods.  more » « less
Award ID(s):
Author(s) / Creator(s):
; ; ; ; ; ; ; ; ; ; ; ; ; ; ;
Date Published:
Journal Name:
International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI)
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Virtual Mental Health Assistants (VMHAs) are utilized in health care to provide patient services such as counseling and suggestive care. They are not used for patient diagnostic assistance because they cannot adhere to safety constraints and specialized clinical process knowledge ( ProKnow ) used to obtain clinical diagnoses. In this work, we define ProKnow as an ordered set of information that maps to evidence-based guidelines or categories of conceptual understanding to experts in a domain. We also introduce a new dataset of diagnostic conversations guided by safety constraints and ProKnow that healthcare professionals use ( ProKnow - data ). We develop a method for natural language question generation (NLG) that collects diagnostic information from the patient interactively ( ProKnow - algo ). We demonstrate the limitations of using state-of-the-art large-scale language models (LMs) on this dataset. ProKnow - algo incorporates the process knowledge through explicitly modeling safety, knowledge capture, and explainability. As computational metrics for evaluation do not directly translate to clinical settings, we involve expert clinicians in designing evaluation metrics that test four properties: safety, logical coherence, and knowledge capture for explainability while minimizing the standard cross entropy loss to preserve distribution semantics-based similarity to the ground truth. LMs with ProKnow - algo generated 89% safer questions in the depression and anxiety domain (tested property: safety ). Further, without ProKnow - algo generations question did not adhere to clinical process knowledge in ProKnow - data (tested property: knowledge capture ). In comparison, ProKnow - algo -based generations yield a 96% reduction in our metrics to measure knowledge capture. The explainability of the generated question is assessed by computing similarity with concepts in depression and anxiety knowledge bases. Overall, irrespective of the type of LMs, ProKnow - algo achieved an averaged 82% improvement over simple pre-trained LMs on safety, explainability, and process-guided question generation. For reproducibility, we will make ProKnow - data and the code repository of ProKnow - algo publicly available upon acceptance. 
    more » « less
  2. As the basis of oceanic food webs and a key component of the biological carbon pump, planktonic organisms play major roles in the oceans. Their study benefited from the development of in situ imaging instruments, which provide higher spatio-temporal resolution than previous tools. But these instruments collect huge quantities of images, the vast majority of which are of marine snow particles or imaging artifacts. Among them, the In Situ Ichthyoplankton Imaging System (ISIIS) samples the largest water volumes (> 100 L s -1 ) and thus produces particularly large datasets. To extract manageable amounts of ecological information from in situ images, we propose to focus on planktonic organisms early in the data processing pipeline: at the segmentation stage. We compared three segmentation methods, particularly for smaller targets, in which plankton represents less than 1% of the objects: (i) a traditional thresholding over the background, (ii) an object detector based on maximally stable extremal regions (MSER), and (iii) a content-aware object detector, based on a Convolutional Neural Network (CNN). These methods were assessed on a subset of ISIIS data collected in the Mediterranean Sea, from which a ground truth dataset of > 3,000 manually delineated organisms is extracted. The naive thresholding method captured 97.3% of those but produced ~340,000 segments, 99.1% of which were therefore not plankton (i.e. recall = 97.3%, precision = 0.9%). Combining thresholding with a CNN missed a few more planktonic organisms (recall = 91.8%) but the number of segments decreased 18-fold (precision increased to 16.3%). The MSER detector produced four times fewer segments than thresholding (precision = 3.5%), missed more organisms (recall = 85.4%), but was considerably faster. Because naive thresholding produces ~525,000 objects from 1 minute of ISIIS deployment, the more advanced segmentation methods significantly improve ISIIS data handling and ease the subsequent taxonomic classification of segmented objects. The cost in terms of recall is limited, particularly for the CNN object detector. These approaches are now standard in computer vision and could be applicable to other plankton imaging devices, the majority of which pose a data management problem. 
    more » « less
  3. Speaker diarization determines who spoke and when? in an audio stream. In this study, we propose a model-based approach for robust speaker clustering using i-vectors. The i-vectors extracted from different segments of same speaker are correlated. We model this correlation with a Markov Random Field (MRF) network. Leveraging the advancements in MRF modeling, we used Toeplitz Inverse Covariance (TIC) matrix to represent the MRF correlation network for each speaker. This approaches captures the sequential structure of i-vectors (or equivalent speaker turns) belonging to same speaker in an audio stream. A variant of standard Expectation Maximization (EM) algorithm is adopted for deriving closed-form solution using dynamic programming (DP) and the alternating direction method of multiplier (ADMM). Our diarization system has four steps: (1) ground-truth segmentation; (2) i-vector extraction; (3) post-processing (mean subtraction, principal component analysis, and length-normalization) ; and (4) proposed speaker clustering. We employ cosine K-means and movMF speaker clustering as baseline approaches. Our evaluation data is derived from: (i) CRSS-PLTL corpus, and (ii) two meetings subset of the AMI corpus. Relative reduction in diarization error rate (DER) for CRSS-PLTL corpus is 43.22% using the proposed advancements as compared to baseline. For AMI meetings IS1000a and IS1003b, relative DER reduction is 29.37% and 9.21%, respectively. 
    more » « less
  4. Existing research on learning with noisy labels mainly focuses on synthetic label noise. Synthetic label noise, though has clean structures which greatly enable statistical analyses, often fails to model the real-world noise patterns. The recent literature has observed several efforts to offer real-world noisy datasets, e.g., Food-101N, WebVision, and Clothing1M. Yet the existing efforts suffer from two caveats: firstly, the lack of ground-truth verification makes it hard to theoretically study the property and treatment of real-world label noise. Secondly, these efforts are often of large scales, which may result in unfair comparisons of robust methods within reasonable and accessible computation power. To better understand real-world label noise, it is important to establish controllable and moderate-sized real-world noisy datasets with both ground-truth and noisy labels. This work presents two new benchmark datasets, which we name as CIFAR-10N, CIFAR-100N, equipping the training datasets of CIFAR-10 and CIFAR-100 with human-annotated real-world noisy labels that we collect from Amazon Mechanical Turk. We quantitatively and qualitatively show that real-world noisy labels follow an instance-dependent pattern rather than the classically assumed and adopted ones (e.g., class-dependent label noise). We then initiate an effort to benchmark a subset of the existing solutions using CIFAR-10N and CIFAR-100N. We further proceed to study the memorization of correct and wrong predictions, which further illustrates the difference between human noise and class-dependent synthetic noise. We show indeed the real-world noise patterns impose new and outstanding challenges as compared to synthetic label noise. These observations require us to rethink the treatment of noisy labels, and we hope the availability of these two datasets would facilitate the development and evaluation of future learning with noisy label solutions. The corresponding datasets and the leaderboard are publicly available at 
    more » « less
  5. null (Ed.)
    Denoising is a fundamental challenge in scientific imaging. Deep convolutional neural networks (CNNs) provide the current state of the art in denoising natural images, where they produce impressive results. However, their potential has been inadequately explored in the context of scientific imaging. Denoising CNNs are typically trained on real natural images artificially corrupted with simulated noise. In contrast, in scientific applications, noiseless ground-truth images are usually not available. To address this issue, we propose a simulation-based denoising (SBD) framework, in which CNNs are trained on simulated images. We test the framework on data obtained from transmission electron microscopy (TEM), an imaging technique with widespread applications in material science, biology, and medicine. SBD outperforms existing techniques by a wide margin on a simulated benchmark dataset, as well as on real data. We analyze the generalization capability of SBD, demonstrating that the trained networks are robust to variations of imaging parameters and of the underlying signal structure. Our results reveal that state-of-the-art architectures for denoising photographic images may not be well adapted to scientific-imaging data. For instance, substantially increasing their field-of-view dramatically improves their performance on TEM images acquired at low signal-to-noise ratios. We also demonstrate that standard performance metrics for photographs (such as PSNR and SSIM) may fail to produce scientifically meaningful evaluation. We propose several metrics to remedy this issue for the case of atomic resolution electron microscope images. In addition, we propose a technique, based on likelihood computations, to visualize the agreement between the structure of the denoised images and the observed data. Finally, we release a publicly available benchmark dataset of TEM images, containing 18,000 examples. 
    more » « less