DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA

Karin Pröll

Research output: Chapter in Book/Report/Conference proceedingsConference contributionpeer-review

Abstract

In this paper we apply convolutional neuronal networks in different configurations to solve prediction tasks on medical data: Given 27 blood parameters obtained by labor blood examination the classes of tumor markers C153 and PSA should be predicted. Based on former work the results of trained Multi-Layer-Perceptrons (MLP) were moderate. Our major interest was now focused on the question if the prediction quality of CNN models outperforms MLPs. We had to transform the vector of input data into a two-dimensional pseudo image and augment it with different correlation values for increasing spatial structure. Various experiments with CNNs show that the prediction quality slightly increases compared to MLPs.
Original languageEnglish
Title of host publication29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017
EditorsFrancesco Longo, Michael Affenzeller, Miquel Angel Piera, Agostino G. Bruzzone, Emilio Jimenez
Pages176-180
Number of pages5
ISBN (Electronic)9781510847651
Publication statusPublished - 2017
EventThe 29th European Modeling & Simulation Symposium EMSS 2017 - Barcelona, Spain
Duration: 18 Sep 201720 Sep 2017
http://www.msc-les.org/conf/emss2017/

Publication series

Name29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017

Conference

ConferenceThe 29th European Modeling & Simulation Symposium EMSS 2017
Country/TerritorySpain
CityBarcelona
Period18.09.201720.09.2017
Internet address

Keywords

  • Convolutional neural networks
  • Deep learning
  • Multi-Layer-Perceptron
  • Transformation of vector into pseudo image

Fingerprint

Dive into the research topics of 'DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA'. Together they form a unique fingerprint.

Cite this