Högskolan i Skövde

his.sePublications
Planned maintenance
A system upgrade is planned for 10/12-2024, at 12:00-13:00. During this time DiVA will be unavailable.
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Analysis of deep learning techniques for cancer genomics using federated learning framework
University of Skövde, School of Informatics.
2021 (English)Independent thesis Advanced level (degree of Master (One Year)), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

Genomics data is complex, high dimensional and confidential. Analysis of such type of data can lead to better understanding of diseases such as cancer. Single cell RNA sequencing is a recent advancement in technology for analyzing cancer at the single cell level. Deep learning techniques have been effectively used by researchers to analyze genomics data. These techniques include feed forward neural networks (FFNN), convolutional neural network (CNN) and long short term memory (LSTM) networks. Federated learning is a framework that involves sharing and aggregation of the machine learning model instead of the data. The project investigates the performance of deep learning techniques for cancer genomics data when implemented using federated learning framework. The two cancer datasets from the Gene Expression Omnibus (GEO) database are identified for the project that contains cell type information based on the gene expressions data. The three deep learning techniques are applied to solve a classification problem. The performance of the models is measured in terms of the f1-score due to class imbalance as these datasets are from tumor sites; therefore the majority class is that of the tumor cell type. Each of the three deep learning models is implemented using the centralised learning framework as well as the federated learning framework. The results demonstrate that the performance of the deep learning models using federated learning for gene expression data is slightly better as compared to that using the centralised learning framework. This supports the fact for further investigation into building deep learning models for heterogeneous genomics data using federated learning to better understand complex diseases such as cancer.

Place, publisher, year, edition, pages
2021. , p. vi, 42
Keywords [en]
Cancer genomics, single cell gene expression, deep learning, federated learning, classification
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:his:diva-20655OAI: oai:DiVA.org:his-20655DiVA, id: diva2:1603679
Subject / course
Informationsteknologi
Educational program
Data Science - Master’s Programme
Supervisors
Examiners
Available from: 2021-10-17 Created: 2021-10-17 Last updated: 2021-10-17Bibliographically approved

Open Access in DiVA

No full text in DiVA

By organisation
School of Informatics
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 653 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf