### Abstract

Language | English |
---|---|

Title of host publication | Interactive Knowledge Discovery and Data Mining in Biomedical Informatics: State-of-the-Art and Future Challenges. Lecture Notes in Computer Science LNCS 8401 |

Place of Publication | Heidelberg, Berlin |

Publisher | Springer |

Pages | 57-80 |

Volume | 8401 |

Edition | 1 |

ISBN (Electronic) | 978-3-662-43968-5 |

ISBN (Print) | 978-3-662-43967-8 |

DOIs | |

Status | Published - 2014 |

### Fingerprint

### Keywords

- data preprocessing
- point cloud data sets
- dermoscopy
- skin cancer
- Machine Learning
- Health Informatics
- graph cuts

### ASJC Scopus subject areas

- Artificial Intelligence

### Fields of Expertise

- Information, Communication & Computing

### Treatment code (Nähere Zuordnung)

- Basic - Fundamental (Grundlagenforschung)

### Cite this

*Interactive Knowledge Discovery and Data Mining in Biomedical Informatics: State-of-the-Art and Future Challenges. Lecture Notes in Computer Science LNCS 8401*(1 ed., Vol. 8401, pp. 57-80). Heidelberg, Berlin: Springer. https://doi.org/10.1007/978-3-662-43968-5_4

**On the Generation of Point Cloud Data Sets: Step One in the Knowledge Discovery Process.** / Holzinger, Andreas; Malle, Bernd; Bloice, Marcus Daniel; Wiltgen, Marco; Ferri, Massimo; Stanganelli, Ignazio; Hoffmann-Wellenhof, Rainer.

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Research › peer-review

*Interactive Knowledge Discovery and Data Mining in Biomedical Informatics: State-of-the-Art and Future Challenges. Lecture Notes in Computer Science LNCS 8401.*1 edn, vol. 8401, Springer, Heidelberg, Berlin, pp. 57-80. https://doi.org/10.1007/978-3-662-43968-5_4

}

TY - CHAP

T1 - On the Generation of Point Cloud Data Sets: Step One in the Knowledge Discovery Process

AU - Holzinger, Andreas

AU - Malle, Bernd

AU - Bloice, Marcus Daniel

AU - Wiltgen, Marco

AU - Ferri, Massimo

AU - Stanganelli, Ignazio

AU - Hoffmann-Wellenhof, Rainer

PY - 2014

Y1 - 2014

N2 - Computational geometry and topology are areas which have much potential for the analysis of arbitrarily high-dimensional data sets. In order to apply geometric or topological methods one must first generate a representative point cloud data set from the original data source, or at least a metric or distance function, which defines a distance between the elements of a given data set. Consequently, the first question is: How to get point cloud data sets? Or more precise: What is the optimal way of generating such data sets? The solution to these questions is not trivial. If a natural image is taken as an example, we are concerned more with the content, with the shape of the relevant data represented by this image than its mere matrix of pixels. Once a point cloud has been generated from a data source, it can be used as input for the application of graph theory and computational topology. In this paper we first describe the case for natural point clouds, i.e. where the data already are represented by points; we then provide some fundamentals of medical images, particularly dermoscopy, confocal laser scanning microscopy, and total-body photography; we describe the use of graph theoretic concepts for image analysis, give some medical background on skin cancer and concentrate on the challenges when dealing with lesion images. We discuss some relevant algorithms, including the watershed algorithm, region splitting (graph cuts), region merging (minimum spanning tree) and finally describe some open problems and future challenges.

AB - Computational geometry and topology are areas which have much potential for the analysis of arbitrarily high-dimensional data sets. In order to apply geometric or topological methods one must first generate a representative point cloud data set from the original data source, or at least a metric or distance function, which defines a distance between the elements of a given data set. Consequently, the first question is: How to get point cloud data sets? Or more precise: What is the optimal way of generating such data sets? The solution to these questions is not trivial. If a natural image is taken as an example, we are concerned more with the content, with the shape of the relevant data represented by this image than its mere matrix of pixels. Once a point cloud has been generated from a data source, it can be used as input for the application of graph theory and computational topology. In this paper we first describe the case for natural point clouds, i.e. where the data already are represented by points; we then provide some fundamentals of medical images, particularly dermoscopy, confocal laser scanning microscopy, and total-body photography; we describe the use of graph theoretic concepts for image analysis, give some medical background on skin cancer and concentrate on the challenges when dealing with lesion images. We discuss some relevant algorithms, including the watershed algorithm, region splitting (graph cuts), region merging (minimum spanning tree) and finally describe some open problems and future challenges.

KW - data preprocessing

KW - point cloud data sets

KW - dermoscopy

KW - skin cancer

KW - Machine Learning

KW - Health Informatics

KW - graph cuts

UR - http://link.springer.com/chapter/10.1007%2F978-3-662-43968-5_4

U2 - 10.1007/978-3-662-43968-5_4

DO - 10.1007/978-3-662-43968-5_4

M3 - Chapter

SN - 978-3-662-43967-8

VL - 8401

SP - 57

EP - 80

BT - Interactive Knowledge Discovery and Data Mining in Biomedical Informatics: State-of-the-Art and Future Challenges. Lecture Notes in Computer Science LNCS 8401

PB - Springer

CY - Heidelberg, Berlin

ER -