Big Data Mining in Medicine
One of the main themes of last Friday’s lecture was big data and its impact across numerous different industries. In today’s blog post, I’d like to focus on how big data and the associated process’ impact on the healthcare industry.
A phrase commonly interchanged with big data is data mining. In an article by Jennifer Bresnick in HealthIT Analytics titled “Data Mining, Big Data Analytics in Healthcare: What’s the Difference”, the author discusses the applications of big data and data mining to medicine. But first, what exactly does data mining mean? “At first blush, the term ‘data mining’ sounds like it should mean ‘the act of finding and extracting data from disparate systems’ in the same way coal, gold, or diamonds are found and extracted from the earth. But data mining may actually presume that data extraction step…Data scientists or informaticists must already have access to a relevant and meaningful dataset – even if it is large and messy – in order to begin mining it,” writes Bresnick. Although the phrases are often interchanged, they actually have separate meanings. Data mining is the process of analyzing big data; a more appropriate term to interchange with datamining is knowledge discovery in data (KDD).
According to the article, the following areas have significant potential for major savings by incorporating big data and data mining:
- Eliminating unnecessary tests. The United States is one of the world leaders in prescribing unnecessary diagnostics due to the profit-driven nature of the healthcare industry.
- Optimizing staffing levels at hospitals, clinics, and urgent care telephone numbers
- Monitoring the opioid (or marijuana) prescription rates of providers to ensure there’s no abuse
- Measuring the effectiveness of certain providers in administering treatments or conducting procedures
Andy Patrizio, in “Big Data and Healthcare” published in Datamation on July 26th, 2017, discusses more easily implementable applications of big data analysis in healthcare. One of the largest drivers of healthcare costs in the United States are administrative expenses; with numerous different insurance providers and health care plans, tens of thousands of hours are spent filling out and processing forms at doctor’s offices. By transitioning to Electronic Medical Records (EMR), the healthcare system can digitize patient data collection, reducing the number of hours spent recording this data. If this data, along with doctor’s notes about patients are centralized and anonymized, it can be used to better predict the outcome of surgeries and side effects of medications for certain patients.
During the 2008 recession, the federal government injected billions of dollars worth of incentives into the healthcare industry. Although we’re beginning to reap the rewards of the initial investment almost a decade ago, there’s still significant room to apply big data tools to improve our healthcare system.
2 comments on “Big Data Mining in Medicine”
Comments are closed.
Jega is also a graduate student in the Department of Aeronautics & Astronautics at Stanford University. He graduated from the University of California, Berkeley in 2015 with a B.S in Mechanical Engineering (Honors) and a minor in Public Policy. Before joining Boeing full-time in August 2015, he interned at NASA, Jaguar Land Rover, and General Motors.
His interests include aircraft design, aircraft interiors, commercial aircraft product strategy, science & technology policy, and intellectual property law. In his free time, Jega enjoys traveling, reading, and coaching high school debate.
Hi Jega, thank you for your overview of the role of data mining in health care! I especially find the part of your essay about Electronics Medical Records (EMR) and data collection interesting, since most machine learning and data mining methods need large datasets to achieve high performance. One of the major problems with machine learning and data mining based applications for healthcare right now is that in most medical domains there is a lack of medical data available. Most hospitals and clinics rarely share their data with researchers, and they almost never open source their anonymized patient data online. Most of this data protection is a result of the Health Insurance Portability and Accountability Act (HIPAA) that requires healthcare providers to appropriately protect their patients’ information (https://www.hhs.gov/hipaa/for-professionals/privacy/laws-regulations/index.html). It will be interesting to see how machine learning and data mining within healthcare will develop over the next couple of years. Will healthcare providers share their patient data in protected form with private companies (from for example Silicon Valley) that help help them uncover more useful insight that can help them keep their patients healthy? Or will they continue to be reluctant to share their data but lose out on the benefits that data mining and machine learning can provide? Will regulation change and make it easier for hospitals and clinics to share their data? You might find this article interesting: https://harvardsciencereview.com/2017/05/16/machine-learning-the-future-of-healthcare/
This is a very interesting topic Vega, and well written post!
You wrote about Big Data and healthcare, and we are currently going through a shift of healthcare applications. As you mentioned in the post, administrative expenses in healthcare is driving up the cost significantly in United States.
I would also add another dimension to your arguments that medical professional could also have the great benefit by being supported by Big Data analytics and machine learning, to help set a diagnose of the patient to both reduce time and also verify that the medical professional is actually correct. Unfortunately, we often see that patient are not given the right treatment from their medical professional due to error in setting the diagnose . However, by a data-driven approach to health with a focus on preventive and proactive medicine by the use og Big Data, I would argue that 1) Big Data can AI can support and help the medical professionals to set the correct diagnose of patients and give them the right treatment; and 2) Reduce the time/cost for medical professionals by the use of new approaches.