Mathematical clustering based on cross-sections in medicine: application to the pancreatic neck
In the context of current surgical techniques, the classification of 3D organs based on two-dimensional cross-sections is a decisive and still challenging task. The goal of this paper is to explore an approach to address this problem. By this means, the expectation is to go further in the directio...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Book/Monograph |
| Language: | English |
| Published: |
Heidelberg
Univ.-Bibliothek
April 22, 2020
|
| Series: | Preprint series of the Engineering Mathematics and Computing Lab (EMCL)
Preprint no. 2020-01 |
| In: |
Preprint series of the Engineering Mathematics and Computing Lab (EMCL) (Preprint no. 2020-01)
|
| DOI: | 10.11588/emclpp.2020.01.72573 |
| Online Access: | Verlag, kostenfrei, Volltext: https://doi.org/10.11588/emclpp.2020.01.72573 Verlag, kostenfrei, Volltext: http://nbn-resolving.de/urn:nbn:de:bsz:16-emclpp-725730 
|
| Author Notes: | Saskia Haupt, Nassim Fard-Rutherford, Philipp D. Lösel, Lars Grenacher, Arianeb Mehrabi, Vincent Heuveline |
| Summary: | In the context of current surgical techniques, the classification of 3D organs based on two-dimensional cross-sections is a decisive and still challenging task. The goal of this paper is to explore an approach to address this problem. By this means, the expectation is to go further in the direction of patient-specific surgery. Based on two-dimensional image data, we analyze different clustering results assuming specific evaluation criteria. By doing so, a determination of the most appropriate number of clusters is possible. As an example, we use this method to classify the shape of the neck of the pancreas of humans, which is relevant for different types of distal pancreatectomy. Hereby, scaling issues of the available data are a key point. Therefore, an overall protocol needs to care for comparable data. |
|---|---|
| Item Description: | Gesehen am 10.09.2020 |
| Physical Description: | Online Resource |
| DOI: | 10.11588/emclpp.2020.01.72573 |