Newly developed MIT software will help to allay patients’ fears about the privacy of their confidential records so that the data can be used for medical research. In the July 24 issue of the journal “BMC Medical Informatics and Decision Making”, researchers describe a computer program capable of automatically deleting details from medical records that may identify patients, while leaving important medical information intact.
Patient records that are shared with the research community must have any identifying information removed, according to HIPAA. However, manual removal of identifying information is prohibitively expensive, time consuming, and prone to error constraints. This has prompted considerable research toward developing automated techniques for “de-identifying” medical records.
The principal investigator Roger G. Mark, a professor in HST and MIT’s Department of Electrical Engineering and Computer Science and his colleagues tested their censoring software on 1,836 nursing notes. Using multiple experts and additional algorithms, they were able to replace all personal information with fake data.
The researchers have reported that the software successfully deleted more than 94% of the confidential information while wrongly deleting only 0.2% of the useful content. This is significantly better than one expert working alone and at least as efficient as two trained medical professionals checking the work. It is also many times faster.
The researchers are providing other researchers with access to the evaluation dataset together with the software to allow others to improve their systems, and to allow the software to be adapted to other data types that may exhibit different qualities.