SIIM 2015: Data mining and decision support for personalized cancer screening management

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
 - Big Data

National Harbor, Md. Consensus statements, organizational guidelines, population-based recommendations—these have traditionally guided clinical decision support in the practice of radiology. The future, as outlined in a paper presented at the Society for Imaging Informatics in Medicine (SIIM) annual meeting, may lie in mining “big data” to create individualized patient management strategies.

Author Arjun Sharma, MD, of the University of Maryland, believes an improved personalized approach to cancer screening is now possible due to the sheer volume of patient and case information. “The rise of ‘big data,’ in concordance with the recent publication of massive clinical datasets such as the results of the Prostate, Lung, Colorectal and Ovarian Cancer (PLCO) Screening Trial and the National Lung Screening Trial, allows the development of data mining tools to more precisely target such recommendations,” wrote Sharma. “Using these matched data, an individualized diagnostic decision-support system can personalize imaging, testing, follow-up intervals, intervention, and prognosis.”

To support this assertion, Sharma used a Web-based application to query a database of patient information from the PLCO trials against screening results, demographics and personalized risk factors. The matching data resulted in a detailed analysis of outcomes that can be used to guide patient-specific treatment and care management strategies. Additionally, the data mined using the tool could eventually be incorporated into computer-aided diagnosis software to better predict future outcomes based on pre-test probabilities.

“Data mining the PLCO dataset for clinical decision support can optimize the use of limited healthcare resources, focusing screening on patients for whom the benefit to risk ratio is the greatest and screening would be most efficacious,” wrote Sharma. “A data driven, personalized approach to cancer screening maximizes its economic and clinical efficacy and enables early identification of patients in which the course of disease can be improved.”