“80% of all healthcare information is unstructured data which is so large and complex that there is dire need for a specialized tool and methods to handle it and derive insights from the data.”
Healthcare data is among the most complex and voluminous data produced in the world today. Lying among this huge pile of healthcare data are precious insights that can directly impact and improve the quality of human lives. While we lacked means of analyzing this data until as recently as a decade ago, progress in Big Data Analytics has made Healthcare Analytics a distinct reality today!
In this blog post, let us examine the problems that Big Data analytics can solve in the healthcare domain. Let us also look at a few case studies of the application of Big Data Analytics in healthcare and the tools that are used.
Why Big Data Analytics in Healthcare?
The foremost benefits of applying Big Data analytics in healthcare are:
- Early discovery and check of epidemics
- Accurate detection and cure of diseases which have low treatment success
- Discovery of new treatments based on genomics and patient profiling
- Prevention of insurance and medi-claim fraud
- Increase in profitability of healthcare institutions
The advent of wearable devices has made collection of healthcare data easier than ever before. From tracking of fitness data to geriatric care and intensive care, wearable technology has revolutionized data collection in healthcare. In fact, Global Connected Health Market 2016-2020 report forecasts the global connected health market to grow at a CAGR of 26.54% during the period 2016-2020!
The data so collected can be stored using Hadoop and analyzed using MapReduce and Spark.
Big Data in Healthcare – Use Case
One of the most well-known implementations of Big Data in Healthcare in recent times is IBM Watson, a powerful cognitive computing platform for healthcare analytics. It is equipped with natural language capabilities, hypothesis generation, and evidence-based learning to support medical professionals as they make decisions.
This is how a physician can use Watson to assist in diagnosing and treating patients:
Step 1: Physician poses a query describing symptoms of the patient and related factors.
Step 2: Watson parses the inputs by mining available patient data for relevant factors such as family health history, medications, test reports etc. and also considers doctor’s notes, clinical studies, research articles and other such data.
Step 3: Watson puts out a list of diagnoses with corresponding scores that indicate the confidence level for each hypothesis. This helps the doctor — and patient — make more informed and accurate decisions.
Evidence-Based Diagnosis – Implementation:
One of the well-known applications of IBM Watson has been the ‘Watson for Oncology’ application which IBM developed in partnership with New York’s Memorial Sloan Kettering Cancer Center (MSK).
- Premise: The basic premise on which the application is built is this – MSK oncologists are known experts in certain types of cancers. If IBM Watson can be trained to take on their expertise, then the knowledge becomes available to any doctor from any corner of the world.
- Program: The Watson for Oncology app is a one-stop application for elite cancer care that can run on an iPad or other tablets.
- Application: Let’s take a hypothetical case of a patient in a far corner of Asia who is suffering from a rare form of lung cancer that is genetically linked. The doctors in the hospital where the patient is getting treated may not have the necessary expertise to treat this specific strain of lung cancer, but Watson for Oncology does with help from MSK Cancer Center data.
The significance of this app is far-reaching as any doctor from anywhere in the world can access the app by just getting a license for the program and give their patients access to world-class cancer treatment. Such is the magic of healthcare analytics born out of access to Big Data in healthcare!
You can find more such use cases linked to predictive analysis and evidence-based treatments
Role of Hadoop in Healthcare Analytics
Hadoop is the underlying technology that is used in many healthcare analytics platforms. This is because, Apache Hadoop is the right fit to handle the huge and complex healthcare data and effectively deal with the challenges plaguing the healthcare industry. A few arguments for using Hadoop to work with Big Data in Healthcare are:
- Hadoop makes data storage less expensive and more available:
Currently, 80% of all healthcare information is unstructured data. This includes physicians’ notes, medical reports, lab results, X-ray, MRI images, vitals and financial data among others. Hadoop provides doctors and researchers the opportunity to find insights from data sets that were earlier impossible to handle.
- Storage capacity and handling:
Most healthcare organizations can store no more than three days’ worth of data per patient, limiting the opportunity for analysis of the produced data. Hadoop can store and handle humongous amount of data, making it the ideal candidate for the job.
- Hadoop can serve as a data organizer and also as an analytics tool:
Hadoop helps researchers find correlations in data sets with many variables, a difficult task for humans. This is why it is the right framework to work with healthcare data.