|
1 | 1 | # Multiomics Survival Pipeline |
| 2 | +Nikolaos Nikolaou, Domingo Salazar, Harish RaviPrakash, Miguel Gonçalves, Rob Mulla, Nikolay Burlutskiy, Natasha Markuzon & Etai Jacob |
2 | 3 |
|
3 | 4 | ## Table of Contents |
4 | 5 |
|
5 | 6 | - [About](#about) |
6 | 7 | - [Getting Started](#getting_started) |
7 | 8 | - [Usage](#usage) |
8 | | -- [Contributing](../CONTRIBUTING.md) |
| 9 | +- [Reference](#reference) |
9 | 10 |
|
10 | 11 | ## About <a name = "about"></a> |
11 | 12 |
|
12 | 13 | A python library for multimodal feature integration and survival prediction. |
13 | | -This pipeline has been developped by the AstraZeneca Oncology Biometrics R&D ML/AI Team. It can be used to preprocess and reduce the dimensionality of tabular datasets (unimodal or multimodal) and train & evaluate survival models on them. Its functionalities include several preprocessing & imputation options, flexibility regarding when to integrate modalities (in the case of multimodal data), a range of feature reduction approaches and survival modelling methods, rigorous evaluation including reporting the models' feature importance. |
| 14 | +This pipeline has been developped by the AstraZeneca Oncology Data Science team. It can be used to preprocess and reduce the dimensionality of tabular datasets (unimodal or multimodal) and train & evaluate survival models on them. Its functionalities include several preprocessing & imputation options, flexibility regarding when to integrate modalities (in the case of multimodal data), a range of feature reduction approaches and survival modelling methods, rigorous evaluation including reporting the models' feature importance. |
| 15 | +If you use it, please cite our paper - [A machine learning approach for multimodal data fusion for survival prediction in cancer patients.](https://www.nature.com/articles/s41698-025-00917-6?utm_source=rct_congratemailt&utm_medium=email&utm_campaign=oa_20250506&utm_content=10.1038/s41698-025-00917-6) |
14 | 16 |
|
15 | 17 | ## Getting Started <a name = "getting_started"></a> |
16 | 18 |
|
17 | 19 | Fork the current repository to get a copy of the repo that can be modified. |
18 | | -Instructions on how to fork a repo can be found at: https://support.atlassian.com/bitbucket-cloud/docs/fork-a-repository |
| 20 | +Instructions on how to fork a repo can be found at:https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/fork-a-repo |
19 | 21 |
|
20 | 22 | ### Prerequisites |
21 | 23 |
|
@@ -79,7 +81,11 @@ Cross-validation is performed for model selection. Once optimal hyperparameters |
79 | 81 |
|
80 | 82 | 7. Evaluation |
81 | 83 | Option for multiple runs, option for different train/test splits, in each of which test set data are only used during evaluation. |
82 | | -Detailed results can include TODO... |
83 | 84 |
|
84 | 85 | The pipeline currently supports only survival analysis (regression) tasks. |
85 | | -The pipeline currently supports only early & intermediate modality fusion approaches. |
| 86 | + |
| 87 | +## Reference <a name = "reference"></a> |
| 88 | +For more details see our paper - [A machine learning approach for multimodal data fusion for survival prediction in cancer patients.](https://www.nature.com/articles/s41698-025-00917-6?utm_source=rct_congratemailt&utm_medium=email&utm_campaign=oa_20250506&utm_content=10.1038/s41698-025-00917-6). If you use this tool please cite it as follows: |
| 89 | +``` |
| 90 | +Nikolaou, N., Salazar, D., RaviPrakash, H., Gonçalves, M., Mulla, R., Burlutskiy, N., Markuzon, N. and Jacob, E., 2025. A machine learning approach for multimodal data fusion for survival prediction in cancer patients. npj Precision Oncology, 9(1), pp.1-14. |
| 91 | +``` |
0 commit comments