| 1. |
|
|
| 2. |
- Bruzzi, M, et al.
(författare)
-
Radiation-hard semiconductor detectors for SuperLHC
- 2005
-
Ingår i: Nuclear Instruments & Methods in Physics Research. Section A: Accelerators, Spectrometers, Detectors, and Associated Equipment. - : Elsevier BV. - 0167-5087 .- 0168-9002. ; 541:1-2, s. 189-201
-
Tidskriftsartikel (refereegranskat)abstract
- An option of increasing the luminosity of the Large Hadron Collider (LHC) at CERN to 1035 cm-2 s-1 has been envisaged to extend the physics reach of the machine. An efficient tracking down to a few centimetres from the interaction point will be required to exploit the physics potential of the upgraded LHC. As a consequence, the semiconductor detectors close to the interaction region will receive severe doses of fast hadron irradiation and the inner tracker detectors will need to survive fast hadron fluences of up to above 1016cm-2. The CERN-RD50 project "Development of Radiation Hard Semiconductor Devices for Very High Luminosity Colliders" has been established in 2002 to explore detector materials and technologies that will allow to operate devices up to, or beyond, this limit. The strategies followed by RD50 to enhance the radiation tolerance include the development of new or defect engineered detector materials (SiC, GaN, Czochralski and epitaxial silicon, oxygen enriched Float Zone silicon), the improvement of present detector designs and the understanding of the microscopic defects causing the degradation of the irradiated detectors. The latest advancements within the RD50 collaboration on radiation hard semiconductor detectors will be reviewed and discussed in this work.
|
|
| 3. |
- Arganda-Carreras, Ignacio, et al.
(författare)
-
Crowdsourcing the creation of image segmentation algorithms for connectomics
- 2015
-
Ingår i: Frontiers in Neuroanatomy. - : Frontiers Media S.A.. - 1662-5129. ; 9:142
-
Tidskriftsartikel (refereegranskat)abstract
- To stimulate progress in automating the reconstruction of neural circuits, we organized the first international challenge on 2D segmentation of electron microscopic (EM) images of the brain. Participants submitted boundary maps predicted for a test set of images, and were scored based on their agreement with a consensus of human expert annotations. The winning team had no prior experience with EM images, and employed a convolutional network. This "deep learning" approach has since become accepted as a standard for segmentation of FM images. The challenge has continued to accept submissions, and the best so far has resulted from cooperation between two teams. The challenge has probably saturated, as algorithms cannot progress beyond limits set by ambiguities inherent in 2D scoring and the size of the test dataset. Retrospective evaluation of the challenge scoring system reveals that it was not sufficiently robust to variations in the widths of neurite borders. We propose a solution to this problem, which should be useful for a future 3D segmentation challenge.
|
|
