Towards Implementation of Ethical Principles: Privacy and Confidentiality in Medical Data Mining—Indian Scenario

Datasets constitute a fundamental part of any medical research. Medical data sets (ie, medical records) are generally noted to be large, comp lex, heterogeneous, hierarchical, t ime related, and of varying quality. Such frag mented and distributed data sets challenge efficient data integration and interpretation. Application of data min ing techniques for healthcare related data will be h ighly rewarding. Lack of do main knowledge within healthcare providers calls for active collaboration between the doctor and data miner. Ethical concerns, especially issues of consent and confidentiality have resulted in the introduction of stringent regulations in doing this form o f research in Ind ia. However we found that only 16% of the reviewed publications in medical data mining within India reported of having undergone ethical rev iew. This suggests that the ethical considerations while p racticing data mining is very much undermined. It is mandatory to obtain ethical and legal clearance fro m reg ional as well as Institutional Ethics Rev iew Board (IERB), before using data min ing tools in health care research.