I was recently invited to attend the XLDB workshop at my alma mater Stony Brook University. The two day workshop brought together experts from industry and academia to explore the application of advanced large scale data analytics to the healthcare sector. Many of the presentations were focused on big data technologies and innovative solutions for leveraging extreme scale data sets to positively impact patient outcomes and efficiency in the healthcare system.
With advancements in imaging modalities and clinical endpoints, clinical trials are routinely incorporating imaging endpoints to assess the efficacy and safety of new therapeutics. The acquisition of patient images (MRI, CT, PET, etc.), especially from large, multisite clinical trials, contribute to large datasets which require specialized tools for proper management and accurate data analysis.
In this post, I'd like to share highlights from a talk by Dr. Elliot Siegel, Professor and Vice Chair of Research Informatics at the University of Maryland, which focused on big data in the imaging space. Dr. Siegel described several ways in which diagnostic imaging data can be made more easily discoverable to radiologists and other medical professionals through the use of searchable image databases containing large numbers of annotated images. As an example, the Annotation and Image Markup (AIM) standard offers a means of adding information to an image in a clinical environment, facilitating automated searches. Clinicians can use these types of tools to better diagnose and treat patients, further ushering in the era of personalized medicine. He also discussed the benefits of computer algorithms that can correlate images, related metadata and complex healthcare information into electronic medical records (EMRs), consolidating comprehensive patient information into a single entity for improved patient care and data access.
Beyond diagnostic applications, Dr. Siegel brought to light big data applications and solutions for imaging informatics. For example, the application of artificial intelligence, IBM Watson Medical Deep QA software to facilitate oncology treatments. Watson has accumulated a wealth of molecular, genomic and medical data from over 26,000 clinical cases enabling him to provide individualized, confidence-scored recommendations to oncologists. To read a blog post about this project, click here. This exciting project has been one of the most successful big data applications to the medical field.
Overall, this meeting was largely successful in bringing big data into the spotlight and emphasizing the potential for big data applications to revolutionize the healthcare sector. The clinical trial imaging space is one of the many areas in healthcare that stands to benefit from Big Data innovators in the coming years.
Stay tuned to my blog for highlights from the National Science Foundation Industry/University Cooperative Research Center for Dynamic Data Analytics (CDDA) Workshop and Industrial Advisory Board Meeting, where I gave a talk on imaging biomarkers and the application of big data to clinical trials.