By: Theresa Marie Romualdez, Georgetown College, Class of 2019
Big data is a seemingly ubiquitous term these days, relevant to conversations from politics to business to social entrepreneurship. While the term itself simply denotes the large store of information to which the public has access via the Internet, it is the use of this data that is particularly groundbreaking. Big data arrives unstructured, meaning data analysts must process this disorganized data by sorting the online words, actions, conversations, and engagements, eventually drawing connections between them. Data analysts sift through this “information dump” to learn about user interests and behavior patterns, which is then implemented to direct the organization’s future actions and processes.
While there are certainly dangers related to this idea of big data, particularly in the accessibility and security of large stores of personal information, there are also endless possibilities that this data could afford us if put to good use. The healthcare industry, for instance, has progressed significantly with access to these new stockpiles of information. Since 2014, hospitals have begun to implement Electronic Health Record (EHR) systems, with an adoption rate of 80% in the United States, making this information readily available to both doctors and patients. Health records themselves are not a new phenomenon, but with the evolution of big data and data analytics, companies and organizations can capitalize on these stores of information to advance healthcare, especially for women.
In 2013 Shirley Pepke, a genomics researcher who started her career as a physicist and data scientist for NASA, was diagnosed with Stage 3 ovarian cancer. Luckily, she had access to a machine learning method called Correlation Explanation (CorEx), developed by data mining expert Greg Ver Steeg to sift through large health data sets. Through this procedure, Pepke found data sets of publicly available gene expression data from ovarian cancer patients through the Cancer Genome Atlas. Pepke had previously undergone chemotherapy to no avail, so upon the recurrence of her cancer, she opted for an immunotherapy drug that was aggregated by CorEx’s results in conjunction with her own tumor data. Although her oncologist’s recommendation called solely for additional chemotherapy, Pepke began taking this immunotherapy drug in tandem with her second round of chemo. Her MRI was clear within two months, and she is still cancer-free to this day.
Pepke treated her cancer strategically with the use of big data. Today, researchers and geneticists are using big data to make cancer prevention resources more accessible. In 2015, Color Genomics launched The Color Test, a comprehensive genetic sequencing test for breast and ovarian cancer. While most risk-assessment tests for breast and ovarian cancer can cost several thousand dollars, The Color Test can be purchased online for only $249. The test analyzes the 19 major genes associated with breast and ovarian cancer, looking for patterns in consumer data to predict whether the patient is prone to certain risks. After purchasing a kit online and sending in a saliva sample, customers have access to genetic counseling and their test results, both of which are available online and included in the cost of The Color Test.
With increasing access to this “big data,” we are at the forefront of technological innovation. We have never had access to so much human information in one place. As big data increases our understanding of women’s health in particular, greater data fluency is likely to revolutionize the entire healthcare industry at an extremely rapid pace. So what could be next? Self-driving cars, Facebook-optimized mental health resources, or household robot nurses? If we continue to use this data for social good, in ways like the example described, the possibilities are truly endless.