utathya / breastcancer

Stacked Based Ensemble Technique

The deep learning ensemble technique combined with CNN for human breast cancer prognosis prediction.

References

Our manuscipt titled with "Multi-modal classification for human breast cancer prognosis prediction: Proposal of deep-learning based stacked ensemble model" has been accepted at IEEE/ACM Transactions on Computational Biology and Bioinformatics.

Requirements

python 3.6

TensorFilow 1.12

keras 2.2.4

scikit-learn 0.20.0

matplotlib 3.0.1

Usage

cnn_clinical.py

cnn_cnv.py

cnn_exp.py

STACKED_RF_HIDDEN.model

ttest.py

Process to execute the Stacked-based ensemble model.

=> Run cnn_clinical.py, cnn_cnv.py, cnn_exp.py for training individual CNNs for clinical, CNA and gene-expression data.

=> After successfull run you will get the hidden features in three different csv files : clinical_metadata.csv, cnv_metadata.csv and exp_metadata.csv

=> Combine all the hidden features of different modalities to form stacked features : stacked_metadata.csv

=> run RF.py and pass the stacked feature(stacked_metadata.csv) as input to get the final prediction output.

=> Once final prediction has been made use ttest.py to perform statistical significance test.