Providers adjusting to greater use of social media in healthcare
A comprehensive collection of articles, videos and more, hand-picked by our editors
The introduction of IBM's Watson is one of the most recognizable instances of big data and its use in healthcare....
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
With its processing power, Watson can quickly "understand" data and assist clinical providers during patient visits, as well as provide specific information based on large amounts of data mined from medical journals.
One critical component has enabled Watson and many other healthcare big data initiatives to deliver some truly amazing results. That area is called "natural language processing" (NLP).
Natural language processing consists of software and algorithms that are capable of mining and analyzing unstructured information in order to understand human language within a specific context. To use a basic example from the clinical perspective, a clinician can ask a computer to extract a patient's chief complaints from a large data set or unstructured visit notes within an EHR system. Even if the chief complaint is not captured in a specific field in the EHR, NLP has the ability to process all the available data and (with a high degree of accuracy) identify and extract the information.
Based on existing cases in healthcare and other markets, NLP is likely to have an increasing effect in the following five areas in coming years:
Clinical data and virtual administrative assistants
Similarly to the way IBM's Watson is being used in clinical environments, NLP can have a significant impact on clinicians if future NLP systems become more interactive and better able to understand commands and requests.
Natural language processing can increase its effectiveness through the use of applications that are able to interact with and collect information from patients and clinical and administrative staff. These could come either in the form of mobile apps that present us with a virtual assistant (similar to Siri or others apps), from which we request items, or possibly in the form of kiosks that patients or users sit in front of. These kiosks would be self-service systems that have a computer-based character that can interact with the patient. It's also likely that future virtual employees will be selected based on user preference. For example, younger patients might be more comfortable with a younger, same-gender assistant, while seniors might prefer an older virtual assistant.
Administratively, NLP has been integrated with many of today's advanced interactive voice response (IVR) systems. These are the automated-response phone systems that we interact with when we contact a hospital. When we try to pay a hospital bill or even when we call our bank to check a balance, we are presented with a set of basic prompts like these: "Caller, please say or press 1 for billing." These traditional systems, however, would be considered outdated and inefficient compared with an IVR system that utilized NLP. By using virtual assistants, these systems can provide more functions to the callers and be more interactive and flexible. They could accommodate such requests as scheduling an office visit or paying any outstanding medical bills.
Data mining and analysis
Researchers and clinicians can leverage clinical information collected from EHR and other systems via NLP and data mining to gain greater insight into disease treatment. They can extract and collect information from the available data sets about the effectiveness of treatment plans, patient outcomes and common factors for specific diseases. From there, NLP can aid in tagging and categorizing the information.
Data collection and extraction
Many health systems that are upgrading or introducing new EHR systems to their facilities and their affiliated practices are facing significant challenges around their legacy systems and the data in them. Some facilities' data is unstructured and in various formats, such as voice files, text files, PDFs, or handwritten and scanned documents. Other facilities have been able to extract the old data and deposit it into their new system, and now have one comprehensive data repository. Using NLP can provide the necessary tools to allow conversion teams to extract and mine information. Such capabilities as named-entity recognition typically are used to assist in this case and others for other search and extraction models.
Big data has been very popular with many marketing executives and specialists. The ability to mine information from social media and other sources enables organizations to connect with their clients and be able to get a better sense of the market's overall feelings about the organization's initiatives and brand. Hospitals are able to do more by performing this analysis using NLP combined with social media. By identifying consumers' responses through online posts, NLP has eliminated the need for traditional survey approaches.
Real-time translation services
Translation is a service that both care providers and patients frequently seek. These services ensure clear and accurate communication, guaranteeing more accurate treatment and a better understanding of a patient's symptoms. Some providers have used translators through teleconferencing to reduce the costs associated with the transportation of translators, but that will likely change with NLP's translation capabilities. Companies like Google, Lucent and Microsoft have provided models for users to use apps and other Web services to translate accurately from a wide range of languages. This breakthrough is not necessarily due to the capabilities of machine translation; rather, it derives from the ability to perform "data-driven" translation instead of explicit word-for-word translation. Having an engine capable of machine learning will give healthcare provider organizations a new, on-demand way of translating.
Healthcare IT executives are likely to see a significant increase in demand for big data and NLP, considering all the benefits that can be gained from their use. Those benefits range from identifying new insights to improving patient outcomes, implementing smarter kiosks and giving providers a better overall understanding of their patients All these are areas of tremendous value for any executive, hospital staff or patient. Natural language processing is moving systems to a new level of interaction and understanding.
Reda Chouffani is vice president of development at Biz Technology Solutions Inc., which provides software design, development and deployment services for the healthcare industry. Let us know what you think about the story; email firstname.lastname@example.org or contact @SearchHealthIT on Twitter.