Please login to the form below

Not currently logged in

Smart Thinking blog

Insights and expert advice on the key issues facing today’s pharma marketer

Pharma’s big data dilemma

How best to contexualise health information? It’s a question of semantics
It's a question of semantics

The big data 'dilemma' in life-sciences has received an incredible amount of attention over the last few years. The 'dilemma' is this: with ever-growing and vast amounts of data becoming available to us, how do we harness this information in order to, among other things, better diagnose disease, deliver healthcare, control costs, listen to the voice of the patient and enhance drug discovery.

Well, a few companies like Google and Apple are working feverishly away to develop a health solution. And not the solution that receives all the attention (wearable health devices and smartphones as cardiac monitors) — but a solution that addresses the dilemma itself of providing relevance and contextualisation to the mountains of health data out there. Then there are other companies who, by virtue of their lack of scale and size - and the natural media attention that comes along with scale and size - are making giant leaps in this field.

The answer to the data dilemma is semantic search. Semantic search seeks to improve search accuracy by understanding searcher intent and the contextual meaning of terms as they appear in the searchable dataspace, whether on the Web or within a closed system, to generate more relevant results. Late in 2013, I became one of the first 50 worldwide alpha users of a new semantic search platform exclusively for the life-sciences industry. A Canadian-based company called Symanta has developed a platform that allows users (bench researchers, R&D teams, commercial sales and marketing individuals, lawyers and regulatory folk) to extract relevant meaning and context from a variety of life-sciences sources. Imagine being able to 'crawl' social media to see what patients and doctors are saying about your drug. Or imagine being a bench researcher interested in a new pathway for drug development but unsure as to the current peer-reviewed literature in that space. Maybe your market access team is looking for relevant reference cases from NICE to make a cost-effectiveness argument with a payer. And imagine doing all this without the 'spammy' and advertising-laden interface that you get with other search platforms.

This is the future and our industry is leading other industry verticals in replacing outdated keyword search or some hybrid keyword-semantic model with a 100 per cent semantic model. In his book Google: Semantic Search, author David Amerland concludes that semantic search is a game-changer. And if this sounds familiar it's because Google is incorporating this into its search algorithm under the 'hummingbird' project moniker. The underlying foundation of Google and Symanta and all the other semantic search companies is the idea of knowledge graphs.

The simple idea behind knowledge graphs is that each subject of interest is represented by a 'node' and these nodes form connections with other nodes (sometimes referred to as parent-child relationships) and the connections between these nodes grows stronger and stronger as more user data is incorporated and internalised by the algorithm. And the beauty of semantic search platforms is that the knowledge graphs are individualised based on the user's 'nodal strength'.

If I search for information related to diabetes, soon the semantic search platform will understand that I am interested in information on diabetes and a truly powerful platform like the one from Symanta will dynamically continue to populate my dashboard with anything and everything that is diabetes-related. Diabetic retinopathy. Diabetic neuropathy. Diabetic hand syndrome. Patient information on diabetes. Clinical trial information on diabetes. News about diabetes. Social media chatter about diabetes medications. You don't need to keep going back and 'querying' the term 'diabetes' any longer. 

Pharma leads other verticals in replacing outdated search models with a 100% semantic model 

But the question still remains: ok, so I can search with more context and my intent is understood, but isn't the data still piling up? In theory it is. In practicality however, the less relevant the search results are to you, the less that is returned through a search query. So, getting back 1.4 million search results in less than a second is infinitely less important than getting back 500 meaningful search results in 3 seconds. The 2 second difference is literally a few blinks of the eye. And almost everyone, including patients who are searching for important information on disease management (lest we make this technology sound like it's only for Fortune 500 companies), is willing to trade off a few seconds for more meaningful data. And the less data you have to sift through, the faster you can make decisions. And in healthcare, whether it's drug discovery or disease diagnosis, speed matters. The irony should not be lost on the reader: by taking more time we save time. Now there's a novelty.

Article by
Rohit Khanna

is president and managing director, Catalytic Health. He can be reached at

12th June 2014

From: Marketing, Healthcare


Featured jobs

Subscribe to our email news alerts


Add my company

Hamell is a full-service agency with a clear focus on delivering evidence-based, sustained behaviour change. So, whether you are looking...

Latest intelligence

Precision paediatrics: Treating patients with CAR-T
Dr Stuart Adams specialises in using T-cell therapy to treat paediatric patients at Great Ormond Street Hospital. Here, he explains what it was like to develop and deliver a groundbreaking...
What does it mean to be an agile organisation
We spoke with Philip Atkinson to learn how healthcare and pharmaceutical companies can rapidly respond to changes in the market....
Battling breast cancer with precision medicine (Part 2)
Dr Mark Moasser treated breast cancer survivor Laura Holmes-Haddad (interviewed in part one) with an innovative precision medicine, which at the time was yet to be approved. Here he gives...