Blog

Cambridge Healthtech Institute and Bio-IT World have awarded Roche a 2020 Innovative Practices Award in Focused Research for their use of the IQVIA Linguamatics Natural Language Processing (NLP) platform to glean patient insights from social media to improve clinical trial design.

Each year, Bio-IT World highlights outstanding examples of technology innovation in the life sciences. The Innovative Practices Awards are designed to recognize partnerships and projects pushing the industry forward, by sharing strategies that can be implemented across the industry to improve the quality, pace, and reach of the life sciences.

NLP over social media identifies clinical endpoints relevant to Parkinson’s disease patients

Bio-IT World judges selected Roche in the patient-focused research category for its work to discover if social media, particularly patient blogs and forums, can provide a good substrate to develop clinical endpoints relevant to Parkinson’s disease patients. Roche researchers established a series of NLP-based text mining queries to analyze patient discussions around Parkinson’s. The study identified symptoms confirmatory of the clinical trial endpoints, and also revealed new symptoms; a number of which have been added to the conceptual disease model used in the clinical trial.


Using NLP workflows to identify key social determinants of health

A year has gone by since I last blogged about the importance of social determinants of health (SDoH). Since then so much has happened in our lives. The COVID-19 pandemic has affected us all, in ways we could never imagine and it will continue to do so into the future.  When you review the SDoH risk factors, you realize that, due to the pandemic, responsible citizens are now enforcing one of these risk factors for safety reasons: social isolation.  It wasn’t too long ago that social isolation was branded a warning sign for mental health issues - not a recommendation for maintaining physical health.  


How research at Kaiser Permanente found that NLP can assist in identifying hospitalizations for worsening heart failure

Heart health is nothing to be taken lightly.  A healthy functioning heart is a key factor in providing every part of your body the blood it needs to provide nutrients and oxygen and remove the waste necessary for life. If there is presence of disease there are times when the heart decompensates, meaning that the heart is no longer able to maintain an efficient circulation allowing for this imperative exchange amongst tissues. According to the Center for Disease Control, in 2017 approximately 6.5 million adults in the United States had heart failure, and it was a major factor of mortality in 1 of 8 deaths. In 2017 it was reported that average wait times for a replacement heart in the U.S. is 191 days, and the cost can approach 1.4 million dollars. So, it is of no surprise that it is better to take care of the heart you have and imperative to capture decompensation as early as possible. Unfortunately, sometimes decompensation can be missed.

NLP workflows in contrast to manual abstraction

Imperative information is often hidden within free text and reports that are attachments within the Electronic Health Record (EHR). A myriad of key cardiology measures can be abstracted from EHRs, be it text based notes to test reports that are PDF attachments, for example: ejection fraction measurement; symptoms such as shortness of breath, fatigue and palpitations; New York Heart Association classifications, and B-type natriuretic peptide (BNP).


Early Scientific Intelligence pipeline gives 360 degree view of “novelty” in diabetes and obesity

We hear a lot these days about evidence-based decision making. Particularly in the current climate, it’s critical that governments, businesses, health organisations and individuals are guided by facts and evidence, not fiction and hearsay. But what do we really mean by evidence-based decision making? Ideally, before making any decision, you want to be able to gather all relevant information, synthesized from different relevant sources. This approach allows you to see the overall picture, drill down to details, understand and weigh up the evidence and therefore make the best decision possible.

Creating a hub of evidence

Getting a comprehensive view of the whole picture is something Linguamatics pharma and healthcare customers need for their decision making in many different arenas - and that often means being able to integrate information from unstructured textual data streams together with data from structured sources. Capturing and integrating the information from a range of document sources can build a landscape of knowledge, a “hub” of evidence. Evidence hubs can be developed for discovery, development, regulatory affairs, safety, patient risk; with input data sources and output dashboards or alerts tailored as needed.

Novo Nordisk are using this integrated approach for an “Early Scientific Intelligence” evidence hub. Sten Christensen & Brian Schurmann (Novo Nordisk) presented on this innovative project at our virtual NLP summit on Thursday 4th June 2020.


The value of NLP for payers and health plans

Health plans and payers rely on medical record review for multiple different business-critical processes. Manual review of the vast amounts of unstructured medical data is very labor-intensive, requiring significant staff and time investment - with related high costs - and generally, slow headway. However, the vital member insights that can be gained from medical records and other unstructured healthcare data sources are too important to be ignored. To address this challenge, payers have increasingly been assessing technology options to streamline the process of identifying and extracting these insights in a more efficient, cost-effective manner.

Natural Language Processing (NLP) as an AI (Augmented Intelligence) technique has become increasingly popular with payers and health plans in recent years. By using NLP to analyze unstructured data like PDF medical records, call center transcripts, and Electronic Health Record (EHR) exports, companies are now able to streamline business processes where manual review is needed - extracting key healthcare insights from medical records in a fraction of the time, at a fraction of the cost.

Key NLP application areas for payers and health plans

Business-critical processes requiring medical record review include NCQA HEDIS™ quality measure reporting, clinical review/medical necessity and Medicare risk adjustment. The more established use of NLP in disease coding, and especially risk adjustment, has paved the way for NLP to also be applied in new areas to enhance predictive models, identify high risk members, reduce manual chart review and streamline business audit processes that require extensive medical record review.