Syntho wins the Global SAS Hackathon in the Category Health Care and Life Sciences
The SAS Hackathon was an extraordinary event that brought together 104 teams from 75 countries, in a truly global showcase of talent. In this extremely competitive environment, we are proud to announced that after months of hard working, Syntho rose to prominence, securing a resounding victory in the healthcare and life sciences category. Surpassing 18 other formidable companies, our outstanding achievement established our position as leaders in this specialized field.
The future of data analytics is poised to be revolutionized by synthetic data, particularly in sectors where privacy-sensitive data, such as healthcare data, is paramount. However, accessing this valuable information is often hindered by cumbersome processes, including time-consuming, fraught with extensive paperwork and numerous restrictions. Recognizing this potential, Syntho joined forces with SAS for the SAS Hackathon to undertake a collaborative project aimed at improving patient care in healthcare institutions. By unlocking privacy-sensitive data through synthetic data and leveraging SAS analytics capabilities, Syntho strives to provide valuable insights that have the potential to shape the future of healthcare.
Unlocking Privacy-Sensitive Healthcare Data with Synthetic Data as part of cancer research for a leading hospital
Patient data is a goldmine of information that can revolutionize healthcare, but its privacy-sensitive nature often poses significant challenges in accessing and utilizing it. Syntho understood this dilemma and sought to overcome it by collaborating with SAS during the SAS Hackathon. The objective was to unlock privacy-sensitive patient data using synthetic data and make it readily available for analytics through SAS Viya. This collaborative effort not only promises to drive improvements in healthcare, specifically in the field of cancer research, making the process of unlocking and analyzing data seamless and efficient, but also ensures the utmost protection of patient privacy.
Integration of Syntho Engine and SAS Viya
Within the hackathon, we successfully incorporated the Syntho Engine API into SAS Viya as a crucial step in our project. This integration not only facilitated the incorporation of synthetic data but also provided an ideal environment to validate its fidelity within SAS Viya. Prior to embarking on our cancer research, extensive testing was conducted using an open dataset to assess the effectiveness of this integrated approach. Through various validation methods available in SAS Viya, we ensured that the synthetic data demonstrated a level of quality and similarity to real data that deemed it truly comparable, affirming its “as-good-as-real” nature.
Does synthetic data match the accuracy of real data?
The correlations and relationships between variables were accurately preserved in synthetic data.
The Area Under the Curve (AUC), a metric for measuring model performance, remained consistent.
Furthermore, the variable importance, which indicated the predictive power of variables in a model, remained intact when comparing synthetic data to the original dataset.
Based on these observations, we can confidently conclude that synthetic data generated by the Syntho Engine in SAS Viya is indeed on par with real data in terms of quality. This validates the use of synthetic data for model development, paving the way for cancer research focused on predicting deterioration and mortality.
Impactful Results with synthetic data in the field of Cancer Research:
The utilization of the integrated Syntho Engine within SAS Viya has yielded impactful results in cancer research for a prominent hospital. By leveraging synthetic data, privacy-sensitive healthcare information was successfully unlocked, enabling analysis with reduced risk, increased data availability, and accelerated access.
Notably, the application of synthetic data led to the development of a model capable of predicting deterioration and mortality, achieving an impressive Area Under the Curve (AUC) of 0.74. Additionally, the combination of synthetic data from multiple hospitals resulted in a remarkable boost in predictive power, as evidenced by the increased AUC. These outcomes underscore the transformative potential of synthetic data in generating data-driven insights and advancements within the field of healthcare.
The result for one leading hospital, an AUC of 0.74 and a model that is able to predict deterioration and mortality
The result for multiple hospitals, an AUC of 0.78, showing that more data results in better predictive power of those models
Results, Future Steps and Implications
During this hackathon, remarkable outcomes were achieved.
1. Syntho, a cutting-edge synthetic data generation tool, was seamlessly integrated into SAS Viya as a crucial step.
2. The successful generation of synthetic data within SAS Viya using Syntho was a significant accomplishment.
3. Notably, the accuracy of the synthetic data was thoroughly validated, as models trained on this data exhibited comparable scores to those trained on the original data.
4. This milestone furthered cancer research by enabling predictions of deterioration and mortality using synthetic data.
5. Remarkably, by combining synthetic data from multiple hospitals, a demonstration revealed an increase in the area under the curve (AUC).
As we celebrate our triumph, we look towards the future with ambitious goals. The next steps involve expanding collaborations with more hospitals, exploring diverse use cases, and extending the application of synthetic data across various sectors. With techniques that are sector-agnostic, we aim to unlock data and realize data-driven insights in healthcare and beyond. The impact of synthetic data in healthcare analytics is just the beginning, as the SAS Hackathon demonstrated the immense interest and participation from data scientists and technology enthusiasts worldwide.
Winning the global SAS hackathon is just the first step for Syntho!
Syntho’s groundbreaking victory in the SAS Hackathon’s Health Care & Life Sciences category signifies a significant milestone in the use of synthetic data for healthcare analytics. The integration of the Syntho Engine within SAS Viya showcased the power and accuracy of synthetic data for predictive modeling and analysis. By collaborating with SAS and unlocking privacy-sensitive data, Syntho has demonstrated the potential of synthetic data to revolutionize patient care, improve research outcomes, and drive data-driven insights in the healthcare industry.