Retrieving similar substructures on 3D neuron reconstructions

Abstract Since manual tracing is time consuming and the performance of automatic tracing is unstable, it is still a challenging task to generate accurate neuron reconstruction efficiently and effectively. One strategy is generating a reconstruction automatically and then amending its inaccurate parts manually. Aiming at finding inaccurate substructures efficiently, we propose a pipeline to retrieve similar substructures on one or more neuron reconstructions, which are very similar to a marked problematic substructure. The pipeline consists of four steps: getting a marked substructure, constructing a query substructure, generating candidate substructures and retrieving most similar substructures. The retrieval procedure was tested on 163 gold standard reconstructions provided by the BigNeuron project and a reconstruction of a mouse’s large neuron. Experimental results showed that the implementation of the proposed methods is very efficient and all retrieved substructures are very similar to the marked one in numbers of nodes and branches, and degree of curvature.

Download Full-text

Retrieving Similar Substructures on 3D Neuron Reconstructions

10.21203/rs.3.rs-88298/v1 ◽

2020 ◽

Author(s):

Jian Yang ◽

Yishan He ◽

Xuefeng Liu

Keyword(s):

Gold Standard ◽

Experimental Results ◽

Neuron Morphology ◽

Neuron Reconstruction ◽

Large Neuron ◽

And Function

Abstract An accurate neuron reconstruction is very important to understand neuron morphology and function, but it is still a challenging task due to the time consuming of manual tracing and the unsatisfactory accuracy of automatic tracing. One way to address the challenge is generating a reconstruction automatically and then checking and amending the result manually. Aiming at implementing this process efficiently, we propose a pipeline to retrieve substructures on one or more neuron reconstructions, which are very similar to a marked problematic substructure and need to be checked one by one. The pipeline consists of four steps: getting a marked substructure, constructing a query substructure, generating candidate substructures and retrieving most similar substructures. The retrieval procedure was tested on 163 gold standard reconstructions provided by the BigNeuron project and a reconstruction of a mouse’s large neuron. Experimental results showed that the implementation of the proposed methods is very efficient and all retrieved substructures are very similar to the marked one in numbers of nodes and branches, and degree of curvature.

Download Full-text

Better synonyms for enriching biomedical search

Journal of the American Medical Informatics Association ◽

10.1093/jamia/ocaa151 ◽

2020 ◽

Vol 27 (12) ◽

pp. 1894-1902 ◽

Cited By ~ 1

Author(s):

Lana Yeganova ◽

Sun Kim ◽

Qingyu Chen ◽

Grigory Balasanov ◽

W John Wilbur ◽

...

Keyword(s):

Gold Standard ◽

Literature Search ◽

Similarity Score ◽

Isotonic Regression ◽

Biomedical Literature ◽

Cosine Similarity ◽

Experimental Results ◽

Word Embeddings ◽

Embedding Method ◽

Embedding Methods

Abstract Objective In a biomedical literature search, the link between a query and a document is often not established, because they use different terms to refer to the same concept. Distributional word embeddings are frequently used for detecting related words by computing the cosine similarity between them. However, previous research has not established either the best embedding methods for detecting synonyms among related word pairs or how effective such methods may be. Materials and Methods In this study, we first create the BioSearchSyn set, a manually annotated set of synonyms, to assess and compare 3 widely used word-embedding methods (word2vec, fastText, and GloVe) in their ability to detect synonyms among related pairs of words. We demonstrate the shortcomings of the cosine similarity score between word embeddings for this task: the same scores have very different meanings for the different methods. To address the problem, we propose utilizing pool adjacent violators (PAV), an isotonic regression algorithm, to transform a cosine similarity into a probability of 2 words being synonyms. Results Experimental results using the BioSearchSyn set as a gold standard reveal which embedding methods have the best performance in identifying synonym pairs. The BioSearchSyn set also allows converting cosine similarity scores into probabilities, which provides a uniform interpretation of the synonymy score over different methods. Conclusions We introduced the BioSearchSyn corpus of 1000 term pairs, which allowed us to identify the best embedding method for detecting synonymy for biomedical search. Using the proposed method, we created PubTermVariants2.0: a large, automatically extracted set of synonym pairs that have augmented PubMed searches since the spring of 2019.

Download Full-text

Diagnostics of Gold Laser Plasmas

International Astronomical Union Colloquium ◽

10.1017/s0252921100108085 ◽

1988 ◽

Vol 102 ◽

pp. 357-360

Author(s):

J.C. Gauthier ◽

J.P. Geindre ◽

P. Monier ◽

C. Chenais-Popovics ◽

N. Tragin ◽

...

Keyword(s):

Experimental Results ◽

Pure Gold ◽

Electronic Temperature ◽

X Ray ◽

Laser Plasmas ◽

Collisional Radiative Model ◽

Radiative Model

AbstractIn order to achieve a nickel-like X ray laser scheme we need a tool to determine the parameters which characterise the high-Z plasma. The aim of this work is to study gold laser plasmas and to compare experimental results to a collisional-radiative model which describes nickel-like ions. The electronic temperature and density are measured by the emission of an aluminium tracer. They are compared to the predictions of the nickel-like model for pure gold. The results show that the density and temperature can be estimated in a pure gold plasma.

Download Full-text

Observation of Phase Contrast Images in STEM and CTEM Modes by Using 100 kV Field Emission Electron Gun

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100051967 ◽

1974 ◽

Vol 32 ◽

pp. 390-391

Author(s):

Y. Harada ◽

T. Goto ◽

H. Koike ◽

T. Someya

Keyword(s):

Field Emission ◽

Phase Contrast ◽

Image Formation ◽

Fresnel Diffraction ◽

Experimental Results ◽

Electron Gun ◽

Scanning Microscopy ◽

Emission Electron ◽

Lattice Images

Since phase contrasts of STEM images, that is, Fresnel diffraction fringes or lattice images, manifest themselves in field emission scanning microscopy, the mechanism for image formation in the STEM mode has been investigated and compared with that in CTEM mode, resulting in the theory of reciprocity. It reveals that contrast in STEM images exhibits the same properties as contrast in CTEM images. However, it appears that the validity of the reciprocity theory, especially on the details of phase contrast, has not yet been fully proven by the experiments. In this work, we shall investigate the phase contrast images obtained in both the STEM and CTEM modes of a field emission microscope (100kV), and evaluate the validity of the reciprocity theory by comparing the experimental results.

Download Full-text

A new dissociation mode for dislocations in silicon

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100108878 ◽

1978 ◽

Vol 36 (1) ◽

pp. 346-347

Author(s):

A. Ourmazd ◽

G.R. Booker ◽

C.J. Humphreys

Keyword(s):

Detailed Examination ◽

Experimental Results ◽

Dislocation Network ◽

Weak Beam ◽

Phosphorus Doped

A (111) phosphorus-doped Si specimen, thinned to give a TEM foil of thickness ∼ 150nm, contained a dislocation network lying on the (111) plane. The dislocation lines were along the three <211> directions and their total Burgers vectors,ḇt, were of the type , each dislocation being of edge character. TEM examination under proper weak-beam conditions seemed initially to show the standard contrast behaviour for such dislocations, indicating some dislocation segments were undissociated (contrast A), while other segments were dissociated to give two Shockley partials separated by approximately 6nm (contrast B) . A more detailed examination, however, revealed that some segments exhibited a third and anomalous contrast behaviour (contrast C), interpreted here as being due to a new dissociation not previously reported. Experimental results obtained for a dislocation along [211] with for the six <220> type reflections using (g,5g) weak-beam conditions are summarised in the table below, together with the relevant values.

Download Full-text

Multislice calculations of RHEED patterns from vicinal Si (001)

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100150836 ◽

1993 ◽

Vol 51 ◽

pp. 1000-1001

Author(s):

Scott Lordi

Keyword(s):

Misorientation Angle ◽

Incident Angle ◽

Incident Beam ◽

Experimental Results ◽

Bragg Condition ◽

Vicinal Surfaces ◽

Kinematic Theory ◽

Step Edges

Vicinal Si (001) surfaces are interesting because they are good substrates for the growth of III-V semiconductors. Spots in RHEED patterns from vicinal surfaces are split due to scattering from ordered step arrays and this splitting can be used to determine the misorientation angle, using kinematic arguments. Kinematic theory is generally regarded to be inadequate for the calculation of RHEED intensities; however, only a few dynamical RHEED simulations have been attempted for vicinal surfaces. The multislice formulation of Cowley and Moodie with a recently developed edge patching method was used to calculate RHEED patterns from vicinal Si (001) surfaces. The calculated patterns are qualitatively similar to published experimental results and the positions of the split spots quantitatively agree with kinematic calculations.RHEED patterns were calculated for unreconstructed (bulk terminated) Si (001) surfaces misoriented towards [110] ,with an energy of 15 keV, at an incident angle of 36.63 mrad ([004] bragg condition), and a beam azimuth of [110] (perpendicular to the step edges) and the incident beam pointed down the step staircase.

Download Full-text