Topics

Visual Correspondences for Unsupervised Domain Adaptation on Electron Microscopy Images.

08:00 EDT 9th October 2019 | BioPortfolio

Summary of "Visual Correspondences for Unsupervised Domain Adaptation on Electron Microscopy Images."

We present an Unsupervised Domain Adaptation strategy to compensate for domain shifts on Electron Microscopy volumes. Our method aggregates visual correspondences-motifs that are visually similar across different acquisitions-to infer changes on the parameters of pretrained models, and enable them to operate on new data. In particular, we examine the annotations of an existing acquisition to determine pivot locations that characterize the reference segmentation, and use a patch matching algorithm to find their candidate visual correspondences in a new volume. We aggregate all the candidate correspondences by a voting scheme and we use them to construct a consensus heatmap: a map of how frequently locations on the new volume are matched to relevant locations from the original acquisition. This information allows us to perform model adaptations in two different ways: either by a) optimizing model parameters under a Multiple Instance Learning formulation, so that predictions between reference locations and their sets of correspondences agree, or by b) using high-scoring regions of the heatmap as soft labels to be incorporated in other domain adaptation pipelines, including deep learning ones. We show that these unsupervised techniques allow us to obtain high-quality segmentations on unannotated volumes, qualitatively consistent with results obtained under full supervision, for both mitochondria and synapses, with no need for new annotation effort. 1.

Affiliation

Journal Details

This article was published in the following journal.

Name: IEEE transactions on medical imaging
ISSN: 1558-254X
Pages:

Links

DeepDyve research library

PubMed Articles [14291 Associated PubMed Articles listed on BioPortfolio]

Unsupervised Domain Adaptation for Classification of Histopathology Whole-Slide Images.

Computational image analysis is one means for evaluating digitized histopathology specimens that can increase the reproducibility and reliability with which cancer diagnoses are rendered while simulta...

Unsupervised Domain Adaptation for Depth Prediction from Images.

State-of-the-art methods to infer dense and accurate depth measurements from images rely on deep CNN models trained in an end-to-end fashion on a significant amount of data. However, despite the outst...

Inferring Latent Domains for Unsupervised Deep Domain Adaptation.

Unsupervised Domain Adaptation (UDA) refers to the problem of learning a model in a target domain where labeled data are not available by leveraging information from annotated data in a source domain....

Magnetic domain wall contrast under zero domain contrast conditions in spin polarized low energy electron microscopy.

Important applications of spin polarized low energy electron microscopy (SPLEEM) employ this technique's vector imaging capability to resolve domain wall (DW) spin textures. Studying several thin film...

Learning unsupervised feature representations for single cell microscopy images with paired cell inpainting.

Cellular microscopy images contain rich insights about biology. To extract this information, researchers use features, or measurements of the patterns of interest in the images. Here, we introduce a c...

Clinical Trials [3861 Associated Clinical Trials listed on BioPortfolio]

Comparison of Characteristics of Bronchoalveolar Lavage Dusting and of Specific Questionnary Data in Healthy Volunteers and Patients Suffering From Sarcoidosis

Comparison of dusting rate of bronchoalveolar lavage in healthy volunteers and patients suffering from sarcoidosis. Blind analysis of bronchoalveolar lavage with Optical Microscopy, Trans...

Adaptation Time Required for the Optimisation of Maia Microperimetry Visual Field Testing

Microperimetry is a relatively new and extremely sensitive method of assessing visual function. It projects light stimuli onto a defined area of the retina to map retinal perceptual thresh...

Internal Limiting Membrane and Macular Hole

Observational study of Internal Limiting Membrane peeled in macular Hole surgery and studied by Transmission electronic Microscopy (TEM) and Optical Coherence Tomography (OCT) findings in ...

Evaluation of Patients Who Have Skin Diseases

RATIONALE: Evaluating patients who have skin diseases may help doctors plan more effective treatment and follow-up for skin diseases and systemic diseases that show up in the skin. PURPOS...

Pilot Study In Vivo Confocal Microscopy Probe for Cervical Precancer

The goal of this proposal is to determine whether a hand-held confocal microscopy probe may perform better than standard methods in detecting cancer or pre-cancerous abnormalities of the c...

Medical and Biotech [MESH] Definitions

A tomographic technique for obtaining 3-dimensional images with transmission electron microscopy.

Microscopy using an electron beam, instead of light, to visualize the sample, thereby allowing much greater magnification. The interactions of ELECTRONS with specimens are used to provide information about the fine structure of that specimen. In TRANSMISSION ELECTRON MICROSCOPY the reactions of the electrons that are transmitted through the specimen are imaged. In SCANNING ELECTRON MICROSCOPY an electron beam falls at a non-normal angle on the specimen and the image is derived from the reactions occurring above the plane of the specimen.

The use of instrumentation and techniques for visualizing material and details that cannot be seen by the unaided eye. It is usually done by enlarging images, transmitted by light or electron beams, with optical or magnetic lenses that magnify the entire image field. With scanning microscopy, images are generated by collecting output from the specimen in a point-by-point fashion, on a magnified scale, as it is scanned by a narrow beam of light or electrons, a laser, a conductive probe, or a topographical probe.

A type of TRANSMISSION ELECTRON MICROSCOPY in which the object is examined directly by an extremely narrow electron beam scanning the specimen point-by-point and using the reactions of the electrons that are transmitted through the specimen to create the image. It should not be confused with SCANNING ELECTRON MICROSCOPY.

Microscopy in which the object is examined directly by an electron beam scanning the specimen point-by-point. The image is constructed by detecting the products of specimen interactions that are projected above the plane of the sample, such as backscattered electrons. Although SCANNING TRANSMISSION ELECTRON MICROSCOPY also scans the specimen point by point with the electron beam, the image is constructed by detecting the electrons, or their interaction products that are transmitted through the sample plane, so that is a form of TRANSMISSION ELECTRON MICROSCOPY.

Quick Search


DeepDyve research library

Relevant Topics

Biotechnology Business
Alliances Astrazeneca Bioethics Boehringer Clinical Research Organization Collaborations GSK Johnson & Johnson Lilly Merck Mergers and Acquisitions Nexium Novartis Pfizer Roche Sanofi ...

Mergers & Acquisitions
Commercial and market reports on mergers and acquisitions in the biotechnology, pharmaceutical, medical device and life-science industries. Mergers and acquisitions (abbreviated M&A;) is an aspect of corporate strategy, corporate finance and manageme...


Searches Linking to this Article