Overview
In this competition, you’ll develop models to improve the prediction of transplant survival rates for patients undergoing allogeneic Hematopoietic Cell Transplantation (HCT) — an important step in ensuring that every patient has a fair chance at a successful outcome, regardless of their background.

Start

Dec 4, 2024
Close
Mar 6, 2025
Merger & Entry
Description
Improving survival predictions for allogeneic HCT patients is a vital healthcare challenge. Current predictive models often fall short in addressing disparities related to socioeconomic status, race, and geography. Addressing these gaps is crucial for enhancing patient care, optimizing resource utilization, and rebuilding trust in the healthcare system.

This competition aims to encourage participants to advance predictive modeling by ensuring that survival predictions are both precise and fair for patients across diverse groups. By using synthetic data—which mirrors real-world situations while protecting patient privacy—participants can build and improve models that more effectively consider diverse backgrounds and conditions.

You’re challenged to develop advanced predictive models for allogeneic HCT that enhance both accuracy and fairness in survival predictions. The goal is to address disparities by bridging diverse data sources, refining algorithms, and reducing biases to ensure equitable outcomes for patients across diverse race groups. Your work will help create a more just and effective healthcare environment, ensuring every patient receives the care they deserve.

Evaluation
Evaluation Criteria
The evaluation of prediction accuracy in the competition will involve a specialized metric known as the Stratified Concordance Index (C-index), adapted to consider different racial groups independently. This method allows us to gauge the predictive performance of models in a way that emphasizes equitability across diverse patient populations, particularly focusing on racial disparities in transplant outcomes.

Concordance index
It represents the global assessment of the model discrimination power: this is the model’s ability to correctly provide a reliable ranking of the survival times based on the individual risk scores. It can be computed with the following formula:



The concordance index is a value between 0 and 1 where:

0.5 is the expected result from random predictions,
1.0 is a perfect concordance and,
0.0 is perfect anti-concordance (multiply predictions with -1 to get 1.0)
Stratified Concordance Index
For this competition, we adjust the standard C-index to account for racial stratification, thus ensuring that each racial group's outcomes are weighed equally in the model evaluation. The stratified c-index is calculated as the mean minus the standard deviation of the c-index scores calculated within the recipient race categories, i.e., the score will be better if the mean c-index over the different race categories is large and the standard deviation of the c-indices over the race categories is small. This value will range from 0 to 1, 1 is the theoretical perfect score, but this value will practically be lower due to censored outcomes.

The submitted risk scores will be evaluated using the score function. This evaluation process involves comparing the submitted risk scores against actual observed values (i.e., survival times and event occurrences) from a test dataset. The function specifically calculates the stratified concordance index across different racial groups, ensuring that the predictions are not only accurate overall but also equitable across diverse patient demographics.

The implementation of the metric is wound in this notebook.
Submission File
Participants must submit their predictions for the test dataset as real-valued risk scores. These scores represent the model's assessment of each patient's risk following transplantation. A higher risk score typically indicates a higher likelihood of the target event occurrence.

The submission file must include a header and follow this format:

ID,prediction
28800,0.5
28801,1.2
28802,0.8
etc.
where:

ID refers to the identifier for each patient in the test dataset.
prediction is the corresponding risk score generated by your model.


Prizes
1st Place - $12,000
2nd Place - $10,000
3rd Place - $10,000
4th Place - $10,000
5th Place - $8,000
Code Requirements
This is a Code Competition
Submissions to this competition must be made through Notebooks. In order for the "Submit" button to be active after a commit, the following conditions must be met:

CPU Notebook <= 9 hours run-time
GPU Notebook <= 9 hours run-time
Internet access disabled
Freely & publicly available external data is allowed, including pre-trained models
Submission file must be named submission.csv
Submission runtimes are slightly obfuscated.
Please see the Code Competition FAQ for more information on how to submit. And review the code debugging doc if you are encountering submission errors.

Background Information
What is an allogeneic HCT?
The human immune system comprises cells that develop from hematopoietic stem cells, a special type of cells that reside in the bone marrow. These stem cells are responsible for generating all blood cells, including red blood cells, platelet-producing cells, and immune system cells such as T cells, B cells, neutrophils, and natural killer (NK) cells. Allogeneic hematopoietic cell transplantation (HCT) can be used to replace an individual's faulty hematopoietic stem cells with stem cells that can produce normal immune system cells. In other words, a successful HCT can help fix a person's immune system by introducing healthy stem cells into their body. When hematopoietic stem cells are transferred from one person to another, the recipient is referred to as the HCT recipient. The term "allogeneic" indicates that the stem cells being used come from someone else, the hematopoietic stem cell donor. If the HCT is successful, the donor's hematopoietic stem cells will replace the recipient's cells, producing blood and immune system cells that work correctly.

The source of hematopoietic stem cells can be bone marrow, peripheral blood, or umbilical cord blood. Depending on the source of the stem cells, HCT procedures may be called bone marrow transplants (BMT), peripheral blood stem cell transplants, or cord blood transplants.

More information on how blood stem cell transplants work.

The competition hosts CIBMTR and NMDP have saved over 130,000 lives through cell therapy.

Citation
Tushar Deshpande, Deniz Akdemir, Walter Reade, Ashley Chow, Maggie Demkin, and Yung-Tsi Bolon. CIBMTR - Equity in post-HCT Survival Predictions. https://kaggle.com/competitions/equity-post-HCT-survival-predictions, 2024. Kaggle.


Cite
Competition Host
CIBMTR

Prizes & Awards
$50,000

Dataset Description
The dataset consists of 59 variables related to hematopoietic stem cell transplantation (HSCT), encompassing a range of demographic and medical characteristics of both recipients and donors, such as age, sex, ethnicity, disease status, and treatment details. The primary outcome of interest is event-free survival, represented by the variable efs, while the time to event-free survival is captured by the variable efs_time. These two variables together encode the target for a censored time-to-event analysis. The data, which features equal representation across recipient racial categories including White, Asian, African-American, Native American, Pacific Islander, and More than One Race, was synthetically generated using the data generator from synthcity, trained on a large cohort of real CIBMTR data.

We have used the SurvivalGAN method, introduced in the paper "SurvivalGAN: Generating Time-to-Event Data for Survival Analysis" which addresses the generation of synthetic survival data with special considerations for censoring. SurvivalGAN is adept at capturing the intricate relationships and interactions among variables within survival data and their influence on time-to-event outcomes. This generative model utilizes a conditional Generative Adversarial Network (GAN) framework, which is specifically tailored to address the complexities of survival analysis, including the critical task of managing censored data. By conditioning on additional information such as censoring status and actual survival times, SurvivalGAN effectively learns the underlying distribution of the data, ensuring that the generated synthetic dataset retains the essential interactions among variables that are predictive of survival outcomes.

Files
train.csv - the training set, with target efs (Event-free survival)
test.csv - the test set; your task is to predict the value of efs for this data
sample_submission.csv - a sample submission file in the correct format with all predictions set to 0.50
data_dictionary.csv - a list of all features and targets used in dataset and their descriptions
Note: The rerun test data contains approximately the same number of observations as the training data.