The volume of healthcare data has grown by about 48 percent annually since 2013, and 2020 alone is expected to produce 2,314 exabytes1.
This growth is thanks to the industry movement toward digitizing existing medical information, and the easy collection of new information through computers and the growing class of Internet of Things devices — whether it be a step counter or an insulin pump. But the availability of this data would be nowhere near as impactful to the industry without the parallel growth of compute power necessary to process all of this data.
All this data and processing power, together with the data-mining abilities of artificial intelligence, means the healthcare industry has the ability to generate a new magnitude of medical insights. One method of generating these insights is called predictive modeling — a statistical technique that uses past behavior to predict3 future outcomes. Predictive modeling has been around since the mid-1900s, but with more data and processing power, as well as the rise of artificial intelligence, healthcare professionals can get better answers, faster, to more questions than ever before —and it’s impacting all stages of patient care.
This content is produced by WIRED Brand Lab in collaboration with Western Digital Corporation.
- US HIV Statistics
- Prediction Models Using EHR Data Can Identify Potential PrEP Candidates
- Predicting Breast Cancer Response to Neoadjuvant Chemotherapy Based on Tumor Vascular Features in Needle Biopsies
- Towards Precision Oncology in Breast Cancer
- Sepsis is a Common Cause of Hospital Deaths
- An Interpretable Machine Learning Model for Accurate Prediction of Sepsis in the ICU
- New analytics model predicts readmission of congestive heart failure patients