Deep integrative analysis for survival prediction

Chenglong Huang, Albert Zhang, Guanghua Xiao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Survival prediction is very important in medical treatment. However, recent leading research is challenged by two factors: 1) the datasets usually come with multi-modality; and 2) sample sizes are relatively small. To solve the above challenges, we developed a deep survival learning model to predict patients’ survival outcomes by integrating multi-view data. The proposed network contains two sub-networks, one view-specific and one common sub-network. We designated one CNN-based and one FCN-based sub-network to efficiently handle pathological images and molecular profiles, respectively. Our model first explicitly maximizes the correlation among the views and then transfers feature hierarchies from view commonality and specifically fine-tunes on the survival prediction task. We evaluate our method on real lung and brain tumor data sets to demonstrate the effectiveness of the proposed model using data with multiple modalities across different tumor types.

Original languageEnglish (US)
Title of host publicationPACIFIC SYMPOSIUM ON BIOCOMPUTING 2018
PublisherWorld Scientific Publishing Co. Pte Ltd
Pages343-352
Number of pages10
Edition212669
ISBN (Print)9789813235533
DOIs
StatePublished - Jan 1 2018
Event23rd Pacific Symposium on Biocomputing, PSB 2018 - Kohala Coast, United States
Duration: Jan 3 2018Jan 7 2018

Other

Other23rd Pacific Symposium on Biocomputing, PSB 2018
CountryUnited States
CityKohala Coast
Period1/3/181/7/18

Fingerprint

Tumors
Brain
Deep learning

Keywords

  • Deep learning
  • Integrative analysis
  • Survival prediction

ASJC Scopus subject areas

  • Biomedical Engineering
  • Computational Theory and Mathematics

Cite this

Huang, C., Zhang, A., & Xiao, G. (2018). Deep integrative analysis for survival prediction. In PACIFIC SYMPOSIUM ON BIOCOMPUTING 2018 (212669 ed., pp. 343-352). World Scientific Publishing Co. Pte Ltd. https://doi.org/10.1142/9789813235533_0032

Deep integrative analysis for survival prediction. / Huang, Chenglong; Zhang, Albert; Xiao, Guanghua.

PACIFIC SYMPOSIUM ON BIOCOMPUTING 2018. 212669. ed. World Scientific Publishing Co. Pte Ltd, 2018. p. 343-352.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Huang, C, Zhang, A & Xiao, G 2018, Deep integrative analysis for survival prediction. in PACIFIC SYMPOSIUM ON BIOCOMPUTING 2018. 212669 edn, World Scientific Publishing Co. Pte Ltd, pp. 343-352, 23rd Pacific Symposium on Biocomputing, PSB 2018, Kohala Coast, United States, 1/3/18. https://doi.org/10.1142/9789813235533_0032
Huang C, Zhang A, Xiao G. Deep integrative analysis for survival prediction. In PACIFIC SYMPOSIUM ON BIOCOMPUTING 2018. 212669 ed. World Scientific Publishing Co. Pte Ltd. 2018. p. 343-352 https://doi.org/10.1142/9789813235533_0032
Huang, Chenglong ; Zhang, Albert ; Xiao, Guanghua. / Deep integrative analysis for survival prediction. PACIFIC SYMPOSIUM ON BIOCOMPUTING 2018. 212669. ed. World Scientific Publishing Co. Pte Ltd, 2018. pp. 343-352
@inproceedings{82749d10baec44c4b702204ef51cab8d,
title = "Deep integrative analysis for survival prediction",
abstract = "Survival prediction is very important in medical treatment. However, recent leading research is challenged by two factors: 1) the datasets usually come with multi-modality; and 2) sample sizes are relatively small. To solve the above challenges, we developed a deep survival learning model to predict patients’ survival outcomes by integrating multi-view data. The proposed network contains two sub-networks, one view-specific and one common sub-network. We designated one CNN-based and one FCN-based sub-network to efficiently handle pathological images and molecular profiles, respectively. Our model first explicitly maximizes the correlation among the views and then transfers feature hierarchies from view commonality and specifically fine-tunes on the survival prediction task. We evaluate our method on real lung and brain tumor data sets to demonstrate the effectiveness of the proposed model using data with multiple modalities across different tumor types.",
keywords = "Deep learning, Integrative analysis, Survival prediction",
author = "Chenglong Huang and Albert Zhang and Guanghua Xiao",
year = "2018",
month = "1",
day = "1",
doi = "10.1142/9789813235533_0032",
language = "English (US)",
isbn = "9789813235533",
pages = "343--352",
booktitle = "PACIFIC SYMPOSIUM ON BIOCOMPUTING 2018",
publisher = "World Scientific Publishing Co. Pte Ltd",
address = "Singapore",
edition = "212669",

}

TY - GEN

T1 - Deep integrative analysis for survival prediction

AU - Huang, Chenglong

AU - Zhang, Albert

AU - Xiao, Guanghua

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Survival prediction is very important in medical treatment. However, recent leading research is challenged by two factors: 1) the datasets usually come with multi-modality; and 2) sample sizes are relatively small. To solve the above challenges, we developed a deep survival learning model to predict patients’ survival outcomes by integrating multi-view data. The proposed network contains two sub-networks, one view-specific and one common sub-network. We designated one CNN-based and one FCN-based sub-network to efficiently handle pathological images and molecular profiles, respectively. Our model first explicitly maximizes the correlation among the views and then transfers feature hierarchies from view commonality and specifically fine-tunes on the survival prediction task. We evaluate our method on real lung and brain tumor data sets to demonstrate the effectiveness of the proposed model using data with multiple modalities across different tumor types.

AB - Survival prediction is very important in medical treatment. However, recent leading research is challenged by two factors: 1) the datasets usually come with multi-modality; and 2) sample sizes are relatively small. To solve the above challenges, we developed a deep survival learning model to predict patients’ survival outcomes by integrating multi-view data. The proposed network contains two sub-networks, one view-specific and one common sub-network. We designated one CNN-based and one FCN-based sub-network to efficiently handle pathological images and molecular profiles, respectively. Our model first explicitly maximizes the correlation among the views and then transfers feature hierarchies from view commonality and specifically fine-tunes on the survival prediction task. We evaluate our method on real lung and brain tumor data sets to demonstrate the effectiveness of the proposed model using data with multiple modalities across different tumor types.

KW - Deep learning

KW - Integrative analysis

KW - Survival prediction

UR - http://www.scopus.com/inward/record.url?scp=85048463747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048463747&partnerID=8YFLogxK

U2 - 10.1142/9789813235533_0032

DO - 10.1142/9789813235533_0032

M3 - Conference contribution

C2 - 29218895

AN - SCOPUS:85048463747

SN - 9789813235533

SP - 343

EP - 352

BT - PACIFIC SYMPOSIUM ON BIOCOMPUTING 2018

PB - World Scientific Publishing Co. Pte Ltd

ER -