[ad_1]
Amazon Web Services (AWS), a subsidiary of the technology giant and e-commerce in the cloud, has launched Amazon Comprehend Medical, a cloud-based software that developers of information technology, healthcare providers health, pharmaceutical companies and others can exploit to exploit structured data. According to the notes of the unstructured clinician in electronic health records (EHRs) and other clinical databases, according to a company statement.
The extracted data can be used for clinical decision support, revenue cycle management, clinical trial management, and population health management, Amazon said. The company has already sold similar text analysis software to companies outside the health sector for purposes such as travel booking and customer support, according to the the Wall Street newspaper .
the Newspaper Taha Kass-Hout, MD, MS, former head of health informatics with the US Food and Drug Administration, said the Amazon software was working as well, or better, than the apps. similar studied. According to Kass-Hout, it does not seem that a decisive step has been made.
A number of other companies, including IBM Watson Health, Optum, as well as small developers and EHR providers, are using natural language processing to identify key clinical data in text notes and convert them into structured data that the authors Analytical applications can then interpret. The challenge is that doctors use many different terms and abbreviations for the same concept. As a result, even the most advanced forms of natural language processing have limited accuracy and are likely to make obvious errors, such as the misinterpretation of a test command. eliminate diabetes as proof that a patient at Diabetes.
According to Amazon, Comprehend Medical "allows developers to automatically identify key types of medical information, with great precision and without requiring a large number of custom rules.Comprehend Medical can identify conditions, anatomical terms, drugs, tests , treatments and procedures. "
The data mining application of Amazon is not associated with clinical analysis software. Instead, the results of data mining are made available to analytics solutions that can access data in the Amazon cloud. "Via Comprehend Medical API [application programming interface]these new features can be easily integrated with existing health services and systems, "said Amazon.
Vendors and developers must send unstructured data to Comprehend Medical's cloud server before they can exploit it. Amazon claims that its service protects the privacy of the patient and that the company has no access to patient data, which requires that a viewer has an encrypted key. In addition, Amazon notes that "the service is also covered by the AWA HIPAA and AWS AWA eligibility criteria. [business associate agreement]. "
Skeptical answers
Dean Sittig, PhD, a professor of biomedical informatics at the University of Texas Health Sciences Center in Houston, has a hard time knowing what value the Amazon data mining software could have. Medscape Medical News.
One of the reasons, he said, is that Amazon has not released any data to verify if the software is "extremely accurate". For the most part, he said, natural language processing applications have an accuracy of 75% to 80%, far less than would be required in clinical care, where even a 95% accuracy is not enough. is not considered sufficient. But in areas such as disease screening and associated medications in patients, the software might prove useful.
Greg Kuhnen, senior director and health informatics advisor for The Advisory Board Co, a health consulting firm, echoed this sentiment, saying it would be difficult for a data mining application to Be sufficiently precise and specific to allow clinical decision support.
"However, Amazon seems to have carefully selected the areas they want to target with the new software," he said. Medscape Medical News. "The areas they have named – recruitment in the life sciences and clinical trials – are two of the easiest areas to address because they fuel a human testing process that tolerates relatively well the false positives. "
Build care pathways
Health organizations clearly have a taste for structured data, associated with analytics, to support quality improvement and health management of the population. For example, Flagler Hospital, a community hospital in St. Augustine, Florida, uses an artificial intelligence (AI) application to create optimal clinical pathways for patients with particular conditions and comorbidities, according to Health News.
Flagler's quality improvement team selected and applied the new lanes to the prescription sets on the hospital's EHR. As part of its pneumonia pilot project, Flagler saved $ 1356 per patient in direct costs and reduced the length of stay by 2 days. The hospital is now applying the same deep learning software to other conditions and hopes to save millions of dollars.
Similarly, Penn Medicine in Philadelphia has used in-depth learning to rethink its care pathways. His first big success was to predict which patients might get septicemia in the hospital before showing signs of an often fatal illness. One of the problems Penn faced was the fact that much of his clinical data was unstructured.
Does data mining such as that advocated by Amazon significantly increase the accuracy of artificial intelligence applications designed to improve the quality of care? Sittig was skeptical. While increasing the amount of structured data introduced into the AI algorithms could slightly improve the resulting pathways of care, he said, this would not solve the problem of human variability.
Nevertheless, if the majority of the patients benefited from it and if the doctors could modify the protocol to adapt it to each patient, this approach would be precious, he added.
For more news, join us on Twitter and Facebook
[ad_2]
Source link
