Using Oracle Cloud for data science, Prosperdtx saves time and money in building digital healthcare plans that prevent hospitalization.
“Because of the time and effort that we’ve saved by using Oracle Cloud, that’s translated into a 25% cost savings.”
Prosperdtx provides digital healthcare plans for patients with cancer. Using machine learning, the company identifies common causes of hospitalization, then provides patients with care recommendations to prevent these causes.
Prosperdtx’s business model relies on building machine learning models, which require large amounts of training data. However, the company’s previous solutions for data management and analytics didn’t let it take in and analyze the vast amounts of data needed for data science models.
The Oracle Cloud team is our partner in our endeavor; Oracle embraces our mission. We’re moving forward not just because of them, but we’re moving forward together.
Why Prosperdtx Chose Oracle
Prosperdtx was impressed by how quickly Oracle developed the COVID-19 Therapeutic Learning System, which was announced on April 1, 2020. Because that application was capable of consuming large amounts of patient data in a short amount of time, Prosperdtx used it as a model for its own machine learning platform.
The company chose Oracle Cloud Infrastructure Data Science to build machine learning models that would predict patients’ risk factors for hospitalization, then provide recommendations on how to prevent those risks.
Prosperdtx chose Oracle not just for the technology, but for the service. The company views Oracle as a strategic partner that has provided robust technical support and insights from working with other healthcare customers.
Using Oracle Cloud, Prosperdtx reduced its development time for migrating data and developing machine learning models from months to weeks. The company also saved 25% in costs during the first 6 months of developing its product.
With Oracle Cloud’s security-first approach, Prosperdtx helps ensure end-to-end HIPAA compliance, from data ingestion to model deployment, while saving time spent in security engineering.