Health Information Analysis

analytics along with the awareness regarding the costs of various drug events. In healthcare industries, data is generated from several sources and collecting this data can help better identify new potentials cures and develop effective drugs in a shorter period of time.

FDA recognizes the potential of mobile applications and how these applications can represent the benefits and risks related to health. It can be useful to use the electronic health records and the patient data from hospitals to build such systems. Science is evolving at a rapid pace and patients and their advocates have a hard time catching up to the most recent developments and consequently are unable to take part in research design. To overcome this, a new initiative is started

This year they are expecting submissions from scientists, healthcare providers, manufacturers and many more organizations. The aim of collecting this data is to convert the submissions in a particular format so that the data can be used effectively. FDA also implemented change in policies and programmatic improvements, which include:

• Implementing a clinical trials program

• New policies to include patients preferences in evaluating risks and benefits of medical devices

• Implementing new policies to promote scientific tools

The FDA has responded to the promises and challenges posed by old devices by bringing better performance effective products to market. Genetic testing provides the right treatment to the patients and helps make informed decisions. Health IT authorizes people to manage their own health by using medical applications. FDA conducted its own assessment, like the data analysis in 2010 to find the cause of the decline in the performance of its premarket program. Data analysis helped FDA in identifying underlying root causes of the steady decline. The agency implemented a number of new policies to improve its performance and to adapt new technologies. The performance of FDA's device program has significantly improved and they are making good progress. To check if the devices are safe and protected, FDA made three categories on the basis of

Data mining consists of six steps: understanding the domain of the problem, understanding the data itself, preparing the data, data mining, discovering knowledge evaluation and finally using the discovered knowledge e (Krzysztof and Kurgan, 2002).

One of the biggest problems with Data Mining in medicine is that raw heath data is not only very large but also very heterogeneous. The data is assembled from various different sources such as from laboratory results, or conversations with the patients or what the doctors interpret. All this data is important as it related to the patients health and none of it can be ignored. Therefore, one of the biggest obstacles to data mining is the scope and the difficulty of medical data.

Incorrect, inconsistent data, missing data and non-standard data such as different bits of information saved in various dissimilar formats create a major obstacle in data mining. It is difficult to process gigabytes of records and formulate rational decisions. Therefore, stored information becomes less useful if that information is not present in an easily comprehendible format. For example, specialists in the same area of medicine don’t agree on some universal terms that can indicate the status of a patient. Not only are different names being used to describe the same disease but also things get more complicated when different grammatical structures are used to explain the relationship