11 Electron Tomography
281
11.3.3 Future Perspectives. Clustering Data in EELS-SV
Tomographic Reconstructions
The potential benefits arising from the joint performance of new ET algorithms
and better segmentation procedures are manifold: to overcome the problems arising
from low acquisition times, few projections or reduced pixel time. This will lead
to better EELS-SV reconstructions with a higher degree of complexity (ELNES
analysis included), and in otherwise non-treatable cases (e.g. samples susceptible
to beam damage). In this sense, systematic approaches have been recently reported,
applying wavelet transform [28, 52] to reduce the incidence of Poissonian noise [53,
54] in HAADF images previous to the 3D reconstruction via TMV-algorithm [55,
56]. The same method is likely to be implemented in EELS-SV reconstruction, given
the unavoidable Poissonian noise present in EELS due to the signal nature.
The recently introduced clustering approach for the EELS-SI segmentation is still
to be tested on EELS-SV reconstructions. In principle, given the physical nature of
the components extracted from cluster analysis, they should be a valid signal in most
ET algorithms, whereas no thickness-related artefacts break the projection requirement. A systematic study on the accuracy of the EELS-SV reconstruction, using
different algorithms, under different acquisition conditions and including clustering
segmentation techniques, needs to be addressed shortly. Also, an in-depth study of the
suitability of the wide variety of clustering algorithms available (e.g. the described
HAC, K-means [48] and agglomerative clustering [57], among others) applied to
EELS data treatment must be also undertaken.
References
1. V. Luˇ ci´ c, F. Förster, W. Baumeister, Structural studies by electron tomography: from cells to
molecules. Annu. Rev. Biochem. 74, 833–865 (2005)
2. P.A. Midgley, M. Weyland, 3D electron microscopy in the physical sciences: the development
of Z-contrast and EFTEM tomography. Ultramicroscopy 96, 413–431 (2003)
3. D.B. Williams, C.B. Carter, The transmission electron microscope. in Transmission Electron
Microscopy (Springer US, 1996), pp. 3–17. https://doi.org/10.1007/978-1-4757-2519-3_1
4. P.A. Midgley, R.E. Dunin-Borkowski, Electron tomography and holography in materials
science. Nat. Mater. 8, 271–280 (2009)
5. Z. Saghi et al., Compressed sensing electron tomography of needle-shaped biological specimens—potential for improved reconstruction fidelity with reduced dose. Ultramicroscopy 160,
230–238 (2016)
6. P. Penczek, M. Marko, K. Buttle, J. Frank, Double-tilt electron tomography. Ultramicroscopy
60, 393–410 (1995)
7. K.J. Batenburg et al., 3D imaging of nanomaterials by discrete tomography. Ultramicroscopy
109, 730–740 (2009)
8. T. Roelandts et al., Accurate segmentation of dense nanoparticles by partially discrete electron
tomography. Ultramicroscopy 114, 96–105 (2012)
9. J. Radon, Über die Bestimmung von Funktionen durch ihre Integralwerte längs gewisser
Mannigfaltigkeiten. in books.google.com, pp. 71–86 (1983). https://doi.org/10.1090/psapm/
027/692055
281
11.3.3 Future Perspectives. Clustering Data in EELS-SV
Tomographic Reconstructions
The potential benefits arising from the joint performance of new ET algorithms
and better segmentation procedures are manifold: to overcome the problems arising
from low acquisition times, few projections or reduced pixel time. This will lead
to better EELS-SV reconstructions with a higher degree of complexity (ELNES
analysis included), and in otherwise non-treatable cases (e.g. samples susceptible
to beam damage). In this sense, systematic approaches have been recently reported,
applying wavelet transform [28, 52] to reduce the incidence of Poissonian noise [53,
54] in HAADF images previous to the 3D reconstruction via TMV-algorithm [55,
56]. The same method is likely to be implemented in EELS-SV reconstruction, given
the unavoidable Poissonian noise present in EELS due to the signal nature.
The recently introduced clustering approach for the EELS-SI segmentation is still
to be tested on EELS-SV reconstructions. In principle, given the physical nature of
the components extracted from cluster analysis, they should be a valid signal in most
ET algorithms, whereas no thickness-related artefacts break the projection requirement. A systematic study on the accuracy of the EELS-SV reconstruction, using
different algorithms, under different acquisition conditions and including clustering
segmentation techniques, needs to be addressed shortly. Also, an in-depth study of the
suitability of the wide variety of clustering algorithms available (e.g. the described
HAC, K-means [48] and agglomerative clustering [57], among others) applied to
EELS data treatment must be also undertaken.
References
1. V. Luˇ ci´ c, F. Förster, W. Baumeister, Structural studies by electron tomography: from cells to
molecules. Annu. Rev. Biochem. 74, 833–865 (2005)
2. P.A. Midgley, M. Weyland, 3D electron microscopy in the physical sciences: the development
of Z-contrast and EFTEM tomography. Ultramicroscopy 96, 413–431 (2003)
3. D.B. Williams, C.B. Carter, The transmission electron microscope. in Transmission Electron
Microscopy (Springer US, 1996), pp. 3–17. https://doi.org/10.1007/978-1-4757-2519-3_1
4. P.A. Midgley, R.E. Dunin-Borkowski, Electron tomography and holography in materials
science. Nat. Mater. 8, 271–280 (2009)
5. Z. Saghi et al., Compressed sensing electron tomography of needle-shaped biological specimens—potential for improved reconstruction fidelity with reduced dose. Ultramicroscopy 160,
230–238 (2016)
6. P. Penczek, M. Marko, K. Buttle, J. Frank, Double-tilt electron tomography. Ultramicroscopy
60, 393–410 (1995)
7. K.J. Batenburg et al., 3D imaging of nanomaterials by discrete tomography. Ultramicroscopy
109, 730–740 (2009)
8. T. Roelandts et al., Accurate segmentation of dense nanoparticles by partially discrete electron
tomography. Ultramicroscopy 114, 96–105 (2012)
9. J. Radon, Über die Bestimmung von Funktionen durch ihre Integralwerte längs gewisser
Mannigfaltigkeiten. in books.google.com, pp. 71–86 (1983). https://doi.org/10.1090/psapm/
027/692055