skip to main content

Search for: All records

Creators/Authors contains: "De, S"

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. Nuclei segmentation is a fundamental task in histopathological image analysis. Typically, such segmentation tasks require significant effort to manually generate pixel-wise annotations for fully supervised training. To alleviate the manual effort, in this paper we propose a novel approach using points only annotation. Two types of coarse labels with complementary information are derived from the points annotation, and are then utilized to train a deep neural network. The fully- connected conditional random field loss is utilized to further refine the model without introducing extra computational complexity during inference. Experimental results on two nuclei segmentation datasets reveal that the proposed method is able to achieve competitive performance compared to the fully supervised counterpart and the state-of-the-art methods while requiring significantly less annotation effort. Our code is publicly available.
  2. Nuclei segmentation and classification are two important tasks in the histopathology image analysis, because the mor- phological features of nuclei and spatial distributions of dif- ferent types of nuclei are highly related to cancer diagnosis and prognosis. Existing methods handle the two problems independently, which are not able to obtain the features and spatial heterogeneity of different types of nuclei at the same time. In this paper, we propose a novel deep learning based method which solves both tasks in a unified framework. It can segment individual nuclei and classify them into tumor, lymphocyte and stroma nuclei. Perceptual loss is utilized to enhance the segmentation of details. We also take advantages of transfer learning to promote the training of deep neural net- works on a relatively small lung cancer dataset. Experimental results prove the effectiveness of the proposed method. The code is publicly available
  3. Free, publicly-accessible full text available October 1, 2023
  4. Free, publicly-accessible full text available July 7, 2023
  5. Free, publicly-accessible full text available June 1, 2023
  6. Free, publicly-accessible full text available June 1, 2023
  7. A bstract A measurement of inclusive, prompt, and non-prompt J/ ψ production in p-Pb collisions at a nucleon-nucleon centre-of-mass energy $$ \sqrt{s_{\mathrm{NN}}} $$ s NN = 5 . 02 TeV is presented. The inclusive J/ ψ mesons are reconstructed in the dielectron decay channel at midrapidity down to a transverse momentum p T = 0. The inclusive J/ ψ nuclear modification factor R pPb is calculated by comparing the new results in p-Pb collisions to a recently measured proton-proton reference at the same centre-of-mass energy. Non-prompt J/ ψ mesons, which originate from the decay of beauty hadrons, are separated from promptly produced J/ ψ on a statistical basis for p T larger than 1.0 GeV/ c . These results are based on the data sample collected by the ALICE detector during the 2016 LHC p-Pb run, corresponding to an integrated luminosity $$ \mathcal{L} $$ L int = 292 ± 11 μ b − 1 , which is six times larger than the previous publications. The total uncertainty on the p T -integrated inclusive J/ ψ and non-prompt J/ ψ cross section are reduced by a factor 1.7 and 2.2, respectively. The measured cross sections and R pPb are compared withmore »theoretical models that include various combinations of cold nuclear matter effects. From the non-prompt J/ ψ production cross section, the $$ \mathrm{b}\overline{\mathrm{b}} $$ b b ¯ production cross section at midrapidity, $$ {\mathrm{d}\sigma}_{\mathrm{b}\overline{\mathrm{b}}} $$ d σ b b ¯ / d y , and the total cross section extrapolated over full phase space, $$ {\sigma}_{\mathrm{b}\overline{\mathrm{b}}} $$ σ b b ¯ , are derived.« less
    Free, publicly-accessible full text available June 1, 2023
  8. Abstract In particle collider experiments, elementary particle interactions with large momentum transfer produce quarks and gluons (known as partons) whose evolution is governed by the strong force, as described by the theory of quantum chromodynamics (QCD) 1 . These partons subsequently emit further partons in a process that can be described as a parton shower 2 , which culminates in the formation of detectable hadrons. Studying the pattern of the parton shower is one of the key experimental tools for testing QCD. This pattern is expected to depend on the mass of the initiating parton, through a phenomenon known as the dead-cone effect, which predicts a suppression of the gluon spectrum emitted by a heavy quark of mass m Q and energy E , within a cone of angular size m Q / E around the emitter 3 . Previously, a direct observation of the dead-cone effect in QCD had not been possible, owing to the challenge of reconstructing the cascading quarks and gluons from the experimentally accessible hadrons. We report the direct observation of the QCD dead cone by using new iterative declustering techniques 4,5 to reconstruct the parton shower of charm quarks. This result confirms a fundamental featuremore »of QCD. Furthermore, the measurement of a dead-cone angle constitutes a direct experimental observation of the non-zero mass of the charm quark, which is a fundamental constant in the standard model of particle physics.« less
    Free, publicly-accessible full text available May 19, 2023
  9. Free, publicly-accessible full text available May 1, 2023