Second Blog

Data Mining on Healthcare

A topic as delicate as healthcare requires a mother load of data about patients, doctors, overhead cost, services provide and many more. The decision-making process becomes more and more difficult the less data that is present.

Like all data, it needs to be analyzed, and in the healthcare industry there are four types of data, patient-centric data which is directly related to patients, aggregate data which is based on performance and resource management data, transformed-based data for planning, clinical and management decision support, and comparative data for health services research and outcomes measurement. Thru the years much of this data is accumulated but information is very low, with the data warehousing of today and data analysis, the healthcare industry can use past data and turn it into better decision-making.

Data in its self has zero organizational value, but when turned into information has the ability to increase profits, control cost/expensed, speed-up services, perform technically sound procedures and maintain high quality of patient care. With the use of data mining; advanced algorithms, multiprocessor computers and massive databases, it is now possible to create patterns from data gathered, turning difficult decisions into a near resemblance of a Standard Operating Procedure. Example, if a doctor is trying to figure out what necessary steps are needed to cure a certain type of cancer attacking a particular part of the body, he would simply check previous data from the data mining, check the patterns that have appeared, age, status, blood pressure, body type, physique, height, weight, symptoms, drugs taken, blood type and the like of that patient and compare it with the other patients suffering from the same cancer, if there are 800 matches for all healthcare centers, than a decision is made quicker and more reliable, thus possibly saving a life, which is the target of a healthcare center.

With data mining, we are now able to use several techniques to further enhance the quality of service, techniques such as decision trees, neural networks, predictive modeling, fuzzy logic, rule induction and many more. These techniques, given data, can help determine survival rates, thus giving priority to patients that are in need of healthcare.

“IBM and the Mayo Foundation for Medical Education and Research, the formal name for the Rochester, Minn.-based nonprofit health care organization, announced that they were teaming up in early 2002. After initially putting the medical records of about 4.4 million patients treated at Mayo's three clinics into a DB2 database, IBM and Mayo have spent the past two years developing and testing a methodology for mining the data.”¹ More and more healthcare centers are aiming to be the best at delivering service to the clients.

In data mining it is equally important for the analyzer to have visual capabilities; the ability to spot patterns. Success of data mining tools relies on the ability of the analyzer to evaluate, analyze and detect patterns from the data.

As science of gathering data and mathematics of analyzing data, become more and more advanced, technology comes up with newer computers or software that help the analyzing of this data to a detail. “Mayo researchers will use Blue Gene to do "deep computing and deep science," including mathematical modeling and a simulation of gene structures to help predict the behavior of diseases”², it is truly amazing that the creative mind can turn prototypes into an actual useable product.

From a business standpoint, being able to predict trends or fads is always a plus with always one step ahead of the market and two steps ahead of the competition; as long as the data is well analyzed there is very minimal error. Errors can be made if even though the data was properly analyzed, given that there will always be unforeseen events, there will be slight errors, at the same time if even the errors were accounted for and there are scenarios prepared for these possible errors, than data mining for business will have close to zero errors because everything has been predicted.

In my personal opinion, data mining is definitely a growing technology. I would expect this to bloom in the future to be a “tool of the trade” for all healthcare centers as well as other industries. This technology can be used in various business by predicting the up and downs of the stock market, whether to build more building or less, to be able to understand the movement of several types of clients if passive or aggressive it would be easier to determine how to act, by clustering clients, decisions could be done by cluster instead of by person, this will control costs and maximize profits.

In conclusion, I believe data mining is an asset for the healthcare industry as well as other businesses interested in improving quality of service/products, speeding up service, improving business decision-making, and understanding the different classes of people/clients involved. If I had a long standing business, investing in data mining would be something I would definitely look into.

Reference:

1. http://www.computerworld.com/s/article/95115/IBM_Mayo_Clinic_Take_Next_Data_Mining_Step1&2