Medical innovation and realistic challenges in the age of big data

In the age of big data, the data environment and analytical techniques that medical research relies on have changed a lot. New predictive analysis and prescriptive analysis, represented by machine learning and deep learning, have overcome the weaknesses of traditional analytical methods. They have been applied in the disease and adverse event risk prediction, clinical auxiliary diagnosis, clinical adjuvant therapy, and precision medical research. Such new application models also bring data-based medical innovation into a broader field. Responding to these trends, Chinese PLA General Hospital has launched a systematic research program in the field of big data, and a centralized data repository has been constructed and more than 20 clinical data analysis tasks are conducted. Most of these applications have achieved good results. However, because the medical big data application is still in its infancy, its development faces some challenges. Problems such as the lack of big data thinking among clinical staff, imperfect data quality, insufficient data managing capability, and inadequate medical data processing and analysis techniques need to be addressed in the future.