Natural Language Processing: A breakthrough technology in healthcare

Natural Language Processing, popularly known as NLP, is the field of Artificial Intelligence which gives the computers the ability to understand, read, analyze and interpret the human language.
Image: Shutterstock

Have you ever tried to converse with a computer in the language you speak?

Sounds like a fascinating concept? Well, it is! Lately, there have been noteworthy discoveries in engaging computers to comprehend the language similarly as we do, well known as Natural Language Processing. However, It is anything but a simple errand training machines to see how we, humans convey.

What is Natural Language Processing?
Natural Language Processing, popularly known as NLP, is the field of Artificial Intelligence which gives the computers the ability to understand, read, analyze and interpret the human language. Simply put, it is the relation between the computers and human language. It has several sub-disciplines, including Natural Language Understanding (NLU), Natural Language Generation (NLG), and Natural Language Query (NLQ). Combining the intensity of Artificial Intelligence, computational phonetics, and computer science, NLP permits a machine to comprehend human language which only humans could possibly do, until now. It is viewed as a difficult issue in computer science due to the nature of human language, which makes it troublesome.

The global market size of NLP is growing at a CAGR of 18.6% for period of 2020-2026 and by 2026, it is estimated to reach to USD 27.6 Billion, from USD 9.9 Billion in 2020. The NLP market consists of major growth factors like the increase in smart device usage, growth in the adoption of cloud-based solutions and NLP-based applications to improve customer service, as well as the increase in technological investments in the healthcare industry.

Levels of NLP

 


NLP in Healthcare
Natural Language Processing is rapidly being adopted in Healthcare industry. It has realized the potential of this cutting edge technology in streamlining the processes. Healthcare industry is quick in understanding the significance of information, gathering data from EMRs/EHRs and other sources. Huge volumes of unstructured patient data is inputted into EHRs on a daily basis, but it’s hard for a computer to help physicians aggregate that critical data. Structured data like claims or CCDAs / FHIR APIs may help determine disease burden, but gives us a limited view of the actual patient record. With the fragmented nature of the healthcare industry, large amount of data is collected in silos having as much as 80 percent of data unstructured and of poor quality. This brings us to a pertinent challenge of data extraction and utilization in the healthcare space through NLP in Healthcare. This unstructured data will take a lot of time and effort if humans try to structure it manually, making the data unusable. This hinders us from taking effective decisions through analytics because of the form of our data. Therefore, NLP can help to leverage this unstructured data as we make a shift gradually from fee for service model to value based care.

For healthcare and life sciences, by 2025 the global market size of NLP is assessed to reach USD 3.7 billion from current size in 2020 that is USD 1.5 billion. Also, it is growing at CAGR of 20.5%. The healthcare and life sciences NLP market include the factors such as increasing use of predictive analytics to enhance health outcomes and growing demand for improving Electronic Health Record (EHR) data usability to enhance patient care. The worldwide healthcare and life sciences NLP market consists of important vendors such as the vendors from United States of America include 3M, Lexalytics, AWS, Google, Nuance, Microsoft, IBM and others. Clinithink is a vendor from Georgia. From Germany, Averbis is a major vendor. Linguamatics is a vendor from the United Kingdom. [7]

Use Cases
Primarily, this technology is helping the healthcare in the following ways

1.      Comprehending human speech and extracting its meaning

2.      Unlocking unstructured data in databases and documents by mapping out essential concepts and values and allowing physicians to use this information for decision making and analytics

Likewise, some of the Use cases in healthcare can be broadly highlighted in the following 3 groups:

1.      Mainstay Use Cases for NLP which has a proven ROI 

A.     Speech Recognition: NLP allows the transcription of huge amount of clinical notes from speech to text. This reduces the task of physicians to dictate notes and therefore, saves time by avoiding duplication. Many companies are working in medical transcription space like Acusis, SmartMD, IKS Health, Aquity and many more

B.     Clinical Documentation: NLP helps the physicians freeing up from the manual and complex structure of EMRs allowing them to focus more on patient care. Nuance and M*Modal have technologies that work in tandem with their speech recognition technologies to capture structured data at the point of care and standardized terminologies for future use

In the future, NLP tools could be applied to social media and other public data sets to determine social determinants of health (SDOH) as well as the effectiveness of wellness-based programs and initiatives.

C.     Computer Assisted Coding (CAC): Computer-assisted coding extracts information about procedures and therapies to capture every code and maximize claims

D.     Data Mining: NLP helps to mine the unstructured patient data allowing organizations to reduce the levels of subjectivity in decision-making and improve the quality of patient care

2.      Emerging Use cases of NLP
A.     Clinical Trial matching: Using NLP and machine learning in healthcare to identify patients for a clinical trial with the help of NLP engines for trial matching

B.     Pharmacovigilance: IQVIA is trying to apply machine learning with natural language processing to transform many Pharmacovigilance functions for greater accuracy and speed [10]

C.     Prior Authorization: NLP modules can be used by Payer to determine prior authorization rapidly. Companies like IBM Watson and Anthem are working on these NLP modules

D.     Clinical Decision Support (CDS): CDS help physicians to make better decisions. Also, it is being used to aid clinicians in checking symptoms and diagnosis

E.      Risk Adjustment and Hierarchical Condition Categories (HCC): HCC relies on ICD-10 coding to assign risk scores to each patient. NLP can help assign patients a risk factor and use their score to predict the costs of healthcare

3.      Next gen Use cases of NLP

A.     Ambient Virtual Scribe: NLP can be used to develop a speech recognition software for clinical documentation giving rise to Virtual scribes limiting the need for human scribes

B.     Precision Medicine: NLP can be used for Computational Phenotyping and Biomarker Discovery

C.     Population Surveillance: An application of NLP to EMRs can be identifying a subset of an ethnic or racial group for eventually documenting and mapping health disparities

Benefits of leveraging NLP in Healthcare Organizations 

Challenges
A key challenge in widespread application of NLP is adapting existing systems to new clinical settings

 
The Future
NLP technologies finds its applications for a range of purposes in Healthcare and Research, including Clinical documentation, medical transcription, clinical trials, Decision Support System and many more. However, the real-life implementation is still facing obstacles. Even though, NLP has the potential to revolutionize with their breakthrough technology and change the landscape of healthcare industry. It is creating new and exciting opportunities in healthcare delivery and patient experience. Natural Language Processing is here to stay, a technology that gets smarter with time, as it empowers the providers to positively influence the health outcomes.

-     The article has been contributed by Dr. Disha Patil, a final year student of PGDM Healthcare Management Batch 2019-21, S.P. Mandali’s Prin. L. N. Welingkar Institute of Management Development and Research (WeSchool), Mumbai.

[This article has been reproduced with permission from Welingkar Institute of Management Development and Research (WeSchool)]