Question

Natural Language Processing (NLP), is not a new technology, but it is one that is not yet fully developed. Many healthcare professionals and organizations are working diligently on how to combine NLP with use in EMRs.

• What is NLP?
• What possible disadvantages will occur due to using NLP technology?
• What are the barriers to using NLP?
• What possible benefits will NLP provide to medical professionals and organizations?
• How can NLP benefit the patient experience?

Answer 1

What is NLP?

Natural language processing (NLP) is a subfield of linguistics, computer science, information engineering, and artificial intelligence concerned with the interactions between computers and human (natural) languages, in particular how to program computers to process and analyze large amounts of natural language data.

                                                     OR

Natural language processing (NLP) is a branch of Artificial Intelligence (AI) that helps computers understand, interpret and manipulate human language. NLP draws from many disciplines, including computer science and computational linguistics, in its pursuit to fill the gap between human communication and computer understanding.

Disadvantages – hard to construct rules, brittle, often fails with variant input, may still require substantial pattern matching even after parsing.

Barriers

Historically, there have been substantial barriers to NLP development in the clinical domain. These barriers are not unique to the clinical domain: they also occur in the fields of software engineering and general NLP.

Lack of access to shared data

Because of concerns regarding patient privacy and worry about revealing unfavorable institutional practices, hospitals and clinics have been extremely reluctant to allow access to clinical data for researchers from outside the associated institutions. The lack of reliable and inexpensive de-identification techniques for narrative reports has compounded the reluctance to share. Such restricted access to shared datasets has hindered collaboration and inhibited the ability to assess and adapt NLP technologies across institutions and among research groups.

Lack of annotated datasets for training and benchmarking

Closely related but not completely conditional on lack of shared datasets is the deficiency of annotated clinical data for training NLP applications and benchmarking performance. The sublanguage of clinical reports often necessitates domain-specific development and training, and, as a consequence, NLP modules developed for general text typically do not perform as well on clinical narratives. We need increased coordination to create annotation sets that can be merged to produce larger training and evaluation sets.

Insufficient common conventions and standards for annotations

Without the ability to share data, the community has lacked incentives for developing common data models for manual and automatic annotations. The result is that annotated datasets are usually unique to the laboratory that generated them and thus remain small and that NLP modules that perform the same tasks cannot be substituted and compared without considerable translational effort. At present, the clinical NLP community is leveraging existing standards and conventions and working together to develop shared data models and to map annotations across information extraction applications.

The formidability of reproducibility

Adopting an existing NLP application or module is complicated—source code and documentation may be unavailable, and published descriptions may lack sufficient detail for reproducibility. Open source releases of clinical information extraction and retrieval systems have improved the opportunity to reproduce performance.12–15 Even with open source release, a tool may work less well in others' hands than in the hands of the original developers. Compounding the problem of reproducibility is the fact that proof-of-concept tools created in academic/research environments may not meet the highest software engineering quality, maintainability, scalability, or usability standards. And sometimes a tool may be over-fitted to a particular application, and modification to solve a similar problem may require wholesale changes. As Pedersen asserted,16 the NLP community needs to invest more in assisting others in applying and reproducing our results.

Limited collaboration

In part due to previously listed barriers, collaboration within the clinical NLP community has been nominal. Development of NLP systems within the academic environment has centered around single institutions and single laboratories, and rather than building upon the foundations of previous work, the majority of clinical NLP systems developed over the last four decades have been reinvented as silos that are neither expanded nor applied outside of the individual laboratory. Other factors limiting collaboration include insufficient infrastructure for facilitating cooperation and the reality that collaboration is inherently inefficient. Nevertheless, as with the biomedical research community at large, a surge in progression beyond the last half century of research can only come through enhanced teamwork.

Lack of user-centered development and scalability

Although we are improving incrementally the predictive performance of clinical NLP tools, clinical NLP applications are seldom deployed in clinical, public health, or health services research settings. Currently, the perceived cost of applying NLP outweighs the perceived benefit. Deploying an NLP system typically requires a substantial amount of time from an expert NLP developer—normally, applications do not generalize and must be rebuilt, retrained, enhanced, and re-evaluated for each new task; the output of an NLP system typically requires extensive mapping to the specific problem being addressed; and the ability to aid a user in customizing the application is generally inadequate.20 We need a shift of focus from accuracy in one task to generalizabiliy across many and from the production of papers as the sole output to production of usable software for medically relevant applications. We also need to understand where NLP tools fit into an overall user workflow so that the tools can be integrated into end-to-end applications for clinical, public health, and clinical research users.

BENEFITS

NLP, a branch of AI, aims at primarily reducing the distance between the capabilities of a human and a machine. As it beginning to get more and more traction in the healthcare space, providers are focusing on developing solutions that can understand, analyze, and generate languages can humans can understand.

There is a further need for voice recognition systems that can automatically respond to queries from patients and healthcare users. There are many more drivers of NLP in Healthcare as elucidated below –

Handle the Surge in Clinical Data

The increased use of patient health record systems and the digital transformation of medicine has led to a spike in the volume of data available with healthcare organizations. The need to make sense out of this data and draw credible insights happens to be a major driver.

Support Value-Based Care and Population Health Management

The shift in business models and outcome expectations is driving the need for better use of unstructured data. Traditional health information systems have been focusing on deriving value from the 20 percent of healthcare data that comes in structured formats through clinical channels.

For advanced patient health record systems, managed care, PHM applications, and analytics and reporting, there is an urgent need to tap into the reservoir of unstructured information that is only getting piled up with healthcare organizations.

NLP in Healthcare could solve these challenges through a number of use cases. Let’s explore a couple of them:

Improving Clinical Documentation – Electronic Health Record solutions often have a complex structure, so that documenting data in them is a hassle. With speech-to-text dictation, data can be automatically captured at the point of care, freeing up physicians from the tedious task of documenting care delivery.

Making CAC more Efficient – Computer-assisted coding can be improved in so many ways with NLP. CAC extracts information about procedures to capture codes and maximize claims. This can truly help hcos make the shift from fee-for-service to a value-based model, thereby improving the patient experience significantly.

Improve Patient-Provider Interactions with EHR

Patients in this day and age need undivided attention from their healthcare providers. This leaves doctors feeling overwhelmed and burned out as they have to offer personalized services while also managing burdensome documentation including billing services.

Studies have shown how a majority of care professionals experience burnout at their workplaces. Integrating NLP with electronic health record systems will help take off workload from doctors and make analysis easier. Already, virtual assistants such as Siri, Cortana, and Alexa have made it into healthcare organizations, working as administrative aids, helping with customer service tasks and help desk responsibilities.

Soon, NLP in Healthcare might make virtual assistants cross over to the clinical side of the healthcare industry as ordering assistants or medical scribes.

Empower Patients with Health Literacy

With conversational AI already being a success within the healthcare space, a key use-case and benefit of implementing this technology is the ability to help patients understand their symptoms and gain more knowledge about their conditions. By becoming more aware of their health conditions, patients can make informed decisions, and keep their health on track by interacting with an intelligent chatbot.

In a 2017 study, researchers used NLP solutions to match clinical terms from their documents with their layman language counterparts. By doing so, they aimed to improve patient EHR understanding and the patient portal experience. Natural Language Processing in healthcare could boost patients’ understanding of EHR portals, opening up opportunities to make them more aware of their health.

Address the Need for Higher Quality of Healthcare

NLP can be the front-runner in assessing and improving the quality of healthcare by measuring physician performance and identifying gaps in care delivery.

Research has shown that artificial intelligence in healthcare can ease the process of physician assessment and automate patient diagnosis, reducing the time and human effort needed in carrying out routine tasks such as patient diagnosis. NLP in healthcare can also identify and mitigate potential errors in care delivery. A study showed that NLP could also be utilized in measuring the quality of healthcare and monitor adherence to clinical guidelines.

Identify Patients who Need Improved Care

Machine Learning and NLP tools have the capabilities needed to detect patients with complex health conditions who have a history of mental health or substance abuse and need improved care. Factors such as food insecurity and housing instability can deter the treatment protocols, thereby compelling these patients to incur more cost in their lifetime.

The data of a patient’s social status and demography is often hard to locate than their clinical information since it is usually in an unstructured format. NLP can help solve this problem. NLP can also be used to improve care coordination with patients who have behavioral health conditions. Both, Natural Language Processing & Machine Learning can be utilized to mine patient data and detect those that are at risk of falling through any gaps in the healthcare system.

Since the healthcare industry generates both structured and unstructured data, it is crucial for healthcare organizations to refine both before implementing NLP in healthcare.

How Would Healthcare Benefit from NLP Integration?

Natural Language Processing in the healthcare industry can help enhance the accuracy and completeness of ehrs by transforming the free text into standardized data. This could also make documentation easier by allowing care providers to dictate notes as NLP turns it into documented data.

Computer-aided coding is another excellent benefit of NLP in healthcare. It can be viewed as a silver bullet for the issues of adding significant detail and introducing specificity in clinical documentation. For providers in need of a point-of-care solution for highly complex patient issues, NLP can be used for decision support. An often-quoted example and an epitome of NLP in healthcare is IBM Watson. It has a massive appetite for academic literature and growing expertise in clinical decision support for precision medicine and cancer care. In 2014, IBM Watson was used to investigating how NLP and Machine Learning could be used to flag patients with heart diseases and help clinicians take the first step in care delivery.

Natural Language Processing algorithms were applied to patient data and several risk factors were automatically detected from the notes in the medical records. Since there is this explosion of data in healthcare which pertains not only to genomes but everything else, the industry needs to find the best way to extract relevant information from it and bring it together to help clinicians base their decisions on facts and insights.

What the Future of NLP in Healthcare Looks Like

NLP in Healthcare is still not up to snuff, but the industry is willing to put in the effort to make advancements. Semantic big data analytics and cognitive computing projects, which have foundations in NLP, are seeing significant investments in healthcare from some recognizable players.

Allied Market Research has predicted that the cognitive computing market will be worth USD 13.7 billion across industries by 2020. The same company has projected spending of USD 6.5 billion on text analytics by 2020. Eventually, natural language processing tools might be able to bridge the gap between the insurmountable volume of data in healthcare generated every day and the limited cognitive capacity of the human brain.

NLP has found applications in healthcare ranging from the most cutting-edge solutions in precision medicine applications to the simple job of coding a claim for reimbursement or billing. The technology has far and wide implications on the healthcare industry, should it be brought to fruition. However, the key to the success of introducing this technology will be to develop algorithms that are intelligent, accurate, and specific to ground-level issues in the industry. NLP will have to meet the dual goals of data extraction and data presentation so that patients can have an accurate record of their health in terms they can understand. If that happens, there are no bars to the improvement in physical efficiency we will witness within the healthcare space.

At Maruti Techlabs, we are truly committed to transforming the healthcare space by building solutions like contextual AI assistants as we realize that conversations with patients or internally at hospitals are rarely just one question and answer. Our chatbot solutions and NLP models have helped leading hospitals within India and abroad, overhaul their patient and staff experience through use cases like automation of appointment booking, feedback collection, optimization of internal process like medical coding and data assessment as well as data entry. It has been truly exhilarating for us to see our clients & partners go live with their chatbots and AI based models, enhance & train over time, and meet their organizational goals.

PATIENT EXPERIENCE

The bringing together of significant information from different sources into a single resource. Then various health providers will be given easy access to it so that they can find specific patient information, without the need to dig through piles of paperwork. This information resource includes symptom documentation that can be analysed to reveal hidden trends based on things that patients mention to their doctor in passing, which currently are not noted as they do not fall into the check-boxes listed on a chart.

To improve the communication between providers and patients. This tool can help educate providers with respect to suitable end-of-life conversations with patients. It can also provide advice regarding the most effective ways to establish goals and priorities, with both patients and their families, while they are in appropriate mindsets.