The Challenge of Big Data Analytics in Clinical Medicine
The Challenge of Big Data Analytics in Clinical Medicine
Clinical medicine and technology haven't always been on the same page. But now that more clinicians are getting more comfortable with technology, what role does Big Data play? Sisense's Elana Roth Katzor sets us up to begin considering the challenges that are being faced.
Join the DZone community and get the full member experience.Join For Free
Access NoSQL and Big Data through SQL using standard drivers (ODBC, JDBC, ADO.NET). Free Download
For a long time, the field of clinical medicine has had the reputation of a technology laggard. But today’s clinicians are gradually getting more comfortable in the digital data age. That’s great timing because it would be a shame to allow the data captured and gathered from patients, equipment and support services remain unused and go to waste – rather than be used to improve clinical care and ultimately save lives.
As in many industries, Big Data analytics methods and software have been trotted out as solutions to the problem of too much data and too little insight. In an April 2015 survey by the Aberdeen group, 40 percent of healthcare professionals identified analytics as a solution to their need for evidence-driven decision making. But let’s start with what’s driving the changes in healthcare data use and analytics.
New Methods to Save Money and Provide Better Care
You can summarize the drivers of BI for healthcare in four words: better care and cost control.
In a recent survey, 59% of healthcare decision makers said the rising cost of healthcare was their top motivator to use analytics methods and tools.
This is not vague pessimism. The U.S. Department of Health and Human Services notified healthcare providers of a plan that links 30% of fee-for-service Medicare payments to alternative models, such as accountable care organizations and bundled payments by the end of 2016.
An Explosion of Healthcare Data Sources
We’ve written before about healthcare data sources in the context of visualizing healthcare data, but now is a chance to take a more in-depth look at this data’s growth rate and where it all comes from.
Depending on whose statistics you trust most, healthcare-related data grows from 33 to 67 percent annually. It comes from a variety of sources: more patients, the move from paper to electronic records and new technology, especially in imaging.
In even the smallest medical practice, clinicians and office workers generate many types of data every day. Here are the most familiar medical systems which generate clinical data and information:
- Electronic health records (EHR): Think of an electronic version of a paper-based system used to document a patient’s condition, treatment, and care. To this collection of mostly structured information, add a wide variety of data that describes methods to measure, improve and maintain patient health. Make this collection of structured and unstructured information storable and portable, and you have an EHR system.
- Patient monitoring systems: Equipment that monitors heart rates, blood pressure, and oxygen rates, measures breathing rates, heart and brain function and many other functions—all use sensors that deliver unstructured data for notification.
- Laboratory systems. Everything you can think of that measures the current condition of a patient can be represented with lab procedures. Structured data (such as iron levels in the blood) and unstructured information (lab slides) are produced in the lab and added to the EHR system.
- Imaging systems: Imaging devices visualize the condition and sometimes the function of patients’ organs. They produce some of the most data-dense information in medicine. Computed tomography (CT), magnetic resonance imaging (MRI), X-ray and ultrasound machines are just some of the systems that provide unstructured data as images.
- Wave-form processors. This type of processing analyzes waves—heartbeats (EKG), brain function (EEGs) and blood volume surging through organs. This data-intensive processing adds more unstructured data to patient records.
- Operations support systems. The many operations tasks that make lab tests, EKGs, and physical therapy possible add an enormous burden onto medical information systems. Making these tasks more efficient and patient-centered continues to be a big challenge for healthcare data managers.
And when healthcare organizations collect all this information, what do they do with it? Generally, they store it, apply algorithms to it and put the results back into storage. But the details—which storage methods, algorithms, and handling processes—make a big difference in the value that hospitals and other organizations can extract from their clinical data.
Applying New Analytical Methods
Big Data analytics helps data specialists find, compile, manage and analyze large volumes of structured, and unstructured data. These powerful tools and methods make it easier for medical data specialists to:
- Combine different kinds of data from many different sources. As can be seen above, healthcare data originates from a large variety of systems and comes in wildly different formats, and in some cases the bulk of it is unstructured. Analyzing this data alongside traditional structured data requires unique expertise and technology.
- Process high volumes of data quickly and accurately. Clinical data is generated fast, and in copious amounts, while clinicians often need to see the results of analysis within a short timeframe. Traditional Big Data technologies might take days or more to provide results for a new analytical query, which creates a pressing need to expedite the process of accessing and querying data.
- Get different types of data technology to work smoothly together. To provide better diagnosis and treatment, healthcare professionals are using sophisticated tools and methods based on analytics interoperability. In other words, different technologies that can communicate, exchange data, and use the information that has been exchanged. Optimizing this process is achieved by creating standards, which healthcare and biotech organizations are developing now.
Accurate diagnosis, treatment, and aftercare require a combined approach, in which structured and unstructured data come from many types of equipment and procedures. Three of the newest uses include image processing – which can make diagnosis more accurate and precision medicine (personalized treatment) possible; signal processing – converting data-dense sensor signals to useful information; and IoT devices – commercial bio-sensor products that are being used in and beyond clinical settings, including wearable devices.
From taming the explosion of data volume to novel storage methods and analytics tools, innovative data technologies are ushering in remarkable improvements in medical care. Progress in machine learning, application interoperability and advanced processing methods used on the Internet of Things will continue to contribute to higher-quality, more personalized patient care.
Published at DZone with permission of Elana Roth Katzor , DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.