DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA

Karin Pröll

DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA

Karin Pröll

Research output: Chapter in Book/Report/Conference proceedings › Conference contribution › peer-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 language	English
Title of host publication	29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017
Editors	Francesco Longo, Michael Affenzeller, Miquel Angel Piera, Agostino G. Bruzzone, Emilio Jimenez
Pages	176-180
Number of pages	5
ISBN (Electronic)	9781510847651
Publication status	Published - 2017
Event	The 29th European Modeling & Simulation Symposium EMSS 2017 - Barcelona, Spain Duration: 18 Sept 2017 → 20 Sept 2017 http://www.msc-les.org/conf/emss2017/

Publication series

Name	29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017

Conference

Conference	The 29th European Modeling & Simulation Symposium EMSS 2017
Country/Territory	Spain
City	Barcelona
Period	18.09.2017 → 20.09.2017
Internet address	http://www.msc-les.org/conf/emss2017/

Keywords

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

Cite this

Pröll, K. (2017). DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA. In F. Longo, M. Affenzeller, M. A. Piera, A. G. Bruzzone, & E. Jimenez (Eds.), 29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017 (pp. 176-180). (29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017).

Pröll, Karin. / DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA. 29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017. editor / Francesco Longo ; Michael Affenzeller ; Miquel Angel Piera ; Agostino G. Bruzzone ; Emilio Jimenez. 2017. pp. 176-180 (29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017).

@inproceedings{92272a28a1574d67b9d1cb927716c99c,

title = "DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA",

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.",

keywords = "Convolutional neural networks, Deep learning, Multi-Layer-Perceptron, Transformation of vector into pseudo image",

author = "Karin Pr{\"o}ll",

year = "2017",

language = "English",

series = "29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017",

pages = "176--180",

editor = "Francesco Longo and Michael Affenzeller and Piera, {Miquel Angel} and Bruzzone, {Agostino G.} and Emilio Jimenez",

booktitle = "29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017",

note = "The 29th European Modeling & Simulation Symposium EMSS 2017 ; Conference date: 18-09-2017 Through 20-09-2017",

url = "http://www.msc-les.org/conf/emss2017/",

}

Pröll, K 2017, DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA. in F Longo, M Affenzeller, MA Piera, AG Bruzzone & E Jimenez (eds), 29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017. 29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017, pp. 176-180, The 29th European Modeling & Simulation Symposium EMSS 2017, Barcelona, Spain, 18.09.2017.

DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA. / Pröll, Karin.
29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017. ed. / Francesco Longo; Michael Affenzeller; Miquel Angel Piera; Agostino G. Bruzzone; Emilio Jimenez. 2017. p. 176-180 (29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017).

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

TY - GEN

T1 - DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA

AU - Pröll, Karin

PY - 2017

Y1 - 2017

N2 - 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.

AB - 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.

KW - Convolutional neural networks

KW - Deep learning

KW - Multi-Layer-Perceptron

KW - Transformation of vector into pseudo image

UR - http://www.scopus.com/inward/record.url?scp=85035113903&partnerID=8YFLogxK

M3 - Conference contribution

T3 - 29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017

SP - 176

EP - 180

BT - 29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017

A2 - Longo, Francesco

A2 - Affenzeller, Michael

A2 - Piera, Miquel Angel

A2 - Bruzzone, Agostino G.

A2 - Jimenez, Emilio

T2 - The 29th European Modeling & Simulation Symposium EMSS 2017

Y2 - 18 September 2017 through 20 September 2017

ER -

Pröll K. DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA. In Longo F, Affenzeller M, Piera MA, Bruzzone AG, Jimenez E, editors, 29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017. 2017. p. 176-180. (29th European Modeling and Simulation Symposium, EMSS 2017, Held at the International Multidisciplinary Modeling and Simulation Multiconference, I3M 2017).

DEEP LEARNING APPROACHES FOR SMALL DIMENSIONAL BIOMEDICAL DATA

Abstract

Publication series

Conference

Keywords

Other files and links

Fingerprint

Cite this