One of the major challenges for EHR vendors like Allscripts is that their strategy and focus has been on reproducing the paper medical record. EHRs have thus been generally incapable of analyzing or interpreting clinical information in a meaningful way or providing advice about sophisticated test ordering. It now appears that at least one of these companies is trying to remedy this problem. Allscripts has launched a subsidiary focusing on optimizing the ordering of genomic tests (see: NIH and Allscripts subsidiary 2bPrecise launch cloud-based genomic technology to boost precision medicine). Below is an excerpt from the article announcing this new company:
2bPrecise, a wholly owned subsidiary of Allscripts Healthcare, has launched an early adopter program at the National Institutes of Health.The NIH, a longtime Allscripts Sunrise EHR client, will deploy a cloud-based genomics and precision medicine platform designed by 2bPrecise. 2bPrecise uses advanced methods in clinical-genomic ontology and data harmonization and returns merged, semantically harmonized knowledge in a machine-readable and structured format.... Medical researchers can combine genomic data with clinical and environmental information in a way that enables statistical analysis and pattern detection. The 2bPrecise EHR-agnostic technology will make it possible for NIH physicians and researchers to use genomic information in their existing Allscripts Sunrise workflow. The goal is to simplify and optimize the complex process of finding, selecting, ordering and receiving genomic tests. Using the 2bPrecise genomics and precision medicine tools, results of genomic sequencing tests will be made available at the point of care through Allscripts Sunrise EHR. Genomic medicine experts from NIH will be the early adopters and include physicians and researchers at the National Human Genome Research Institute, the National Cancer Institute and the NIH Clinical Center. “2bPrecise will soon be able to take a big step toward bringing an easy-to-use, clinical-genomic solution into the mainstream," 2bPrecise General Manager Assaf Halevy said in a statement. He added the offering soon would be available to any provider, integrated delivery network or payer organization to help unlock new insights of molecular diagnostics and new treatments.
Obviously, ordering the most appropriate genomic tests for a patient demands sophisticated computer support. All of this sounds like a good start on a deep learning initiative for one EHR vendor. Just so we are on the same page, here's a good definition for deep learning from Wikipedia (see: Deep learning):
Deep learning is part of a broader family of machine learning methods based on learning representations of data. An observation (e.g., an image) can be represented in many ways such as a vector of intensity values per pixel, or in a more abstract way as a set of edges, regions of particular shape, etc. Some representations are better than others at simplifying the learning task (e.g., face recognition or facial expression recognition). One of the promises of deep learning is replacing handcrafted features with efficient algorithms for unsupervised or semi-supervised feature learning and hierarchical feature extraction.
One aspect of this announcement seems to me to be particularly important. EHR vendors that embrace new technology like deep learning as a component of their overarching strategy would be well advised, like Allscripts, to create new companies to execute this new strategy. Not to do so would jeopardize the strategy. The best deep learning experts would probably not be attracted to an EHR company that has historically worked mainly with older technology. In general, healthcare IT is far behind the national curve. Perhaps these companies like 2bPrecise will help to reverse this trend.