Artificial intelligence blends with data quality strategies for smarter management of the healthcare data deluge
By Barley Laing, UK Managing Director at Melissa
Data adds great value in healthcare. Yet with huge amounts of complex data driven by increased digitisation, the health sector needs viable strategies to make sense of it all.
Health professionals are faced with many different types and variations of data, such as drug doses and molecular structures, in many forms such as text, numbers and images. At the same time, health systems are continually acquiring new types of data delivered by innovative technology, for example GPS enabled inhalers and Fitbit logs. While this fast-growing and diverse volume of data can drive excellent patient care, it simultaneously challenges time-pressured health professionals to quickly review and make vital decisions on patients and resources.
This is where artificial intelligence (AI) can play a critical role, with its ability to handle tasks with greater accuracy and speed, while at less cost than traditional handling by a human workforce.
This begs the question – what AI is appropriate for the health service and its need to bring together diverse datasets to make new learnings?
The good news is that AI has been evolving – and the relatively new arena of informatics combines AI, computer, and cognitive science to help bring a new level of intelligence to the industry. Semantic technology, a form of AI that associates words with meanings and understands relationships between them, is the icing on this cake and will arguably have the greatest impact on the heath sector.
Defining semantic technology
Semantic technology, also called semantic web or “semtech,” is an extension of the current web standard that has been defined over the past decade by the World Wide Web Consortium (W3C) in collaboration with MIT, Stanford University and others. W3C is the same institution that manages specifications for HTML, the standardised language for web documents. Where HTML is designed to enable universal access to documents on the web, semantic technologies are designed to enable universal data interoperability.
In addition to setting a standard for global data interoperability, semantic technology builds relationships between data. When applied to any number of data formats and sources, semtech creates links and context within the data. It’s this capability that’s set to provide health professionals with additional understanding of their patients, by making connections and learnings from the data in patient records in real-time. By uniquely describing all content and relationships within and across datasets, semantic technology makes machine reasoning possible. The result is greater efficiency and depth in uncovering hidden insight in the data. For health professionals this means they can quickly identify possible risk for patients, enabling them to intervene with preventative measures at an early stage, improving patient outcomes and saving lives.
Looking at the bigger picture, semantic technology can integrate data from virtually any source and can spot trends, such as an increase in the types of cancer in a certain geographic region or demographic, and can predict its possible future growth. This type of semantically enabled pattern recognition makes preventative intervention possible through the targeted deployment of health resources. This same knowledge helps to inform and improve public policy decisions around health spending, informing smart judgments on where to target precious health resources and budgets.
One of the most important aspects of semantic technology is that it can update and deliver new data and learnings as soon as additional data is input into the system. This groundbreaking technology can redefine concepts and relationships as new information becomes available – raising its value as a flexible and powerful tool that evolves in step with data.
Accurate data and intuitive platform essential for healthcare professionals
It’s important to note that semantic technology is only effective with accurate data. To access true value, health services must prioritise their investments in data quality operations that continually cleanse, standardise, and enhance patient information. These processes must be coupled with an easy to use interface, enabling users to visually explore and analyse the data without any knowledge of coding. Semantic technologies optimised for 
healthcare include built-in, in-depth knowledge of drugs, genes, and diseases – recognising and validating drug names, variants, dosages, and spellings. The right tools are designed from the ground up to prevent any confusion and mitigate errors in what is a complex and critical sector.
AI delivered by semantic technologies opens a wealth of opportunity to improve efficiency in healthcare. In this very powerful new era, errors are reduced, data insights are faster and more sophisticated, and staff are freed to focus on excellent care. Leaders in the health service must take notice of technology that can both inform their spending targets and save patient lives.
