Preserving privacy for data analysis

A computer program product for anonymizing a plurality of statistical data for a secure transfer, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a hospital computer processing system to cause the hospital computer processing system to perform a method comprising:

• calculating, by each of a plurality of privacy processors managing access to a plurality of isolated patient databases of a plurality of hospital computer processing systems, statistical information for each of the plurality of statistical data for a patient, the privacy processors being respective portals through which the statistical data is anonymized for the secure transfer, each of the plurality of privacy processors connected to a respective one of the plurality of isolated patient databases of a respective one of the plurality of hospital computer processing systems;
• aggregating, by the plurality of privacy processors, the statistical information to calculate a valid range for each of the statistical data;
• removing, by the plurality of privacy processors, outlier data based on the valid range for each of the plurality of statistical data;
• creating, by the plurality of privacy processors, pair lists from each of the plurality of statistical data and target data, the pair lists having a respective member from both the plurality of statistical data and the target data, the target data relating to a specific health status test;
• replacing, by the plurality of privacy processors, each respective member of the target data by a random number existing in a range of a corresponding one of a plurality of target data bins;
• swapping, by the plurality of privacy processors, each pair in each of the pair lists in a random order using a randomized number, wherein the randomized number used for swapping is different for different ones of the pair lists;
• performing, by a wireless communication device, the secure transfer for patient privacy protection, by sending the swapped pairs over multiple communication paths from the plurality of hospital computer processing systems to a remote statistical analysis system, wherein the wireless communication device transforms the swapped pairs into a wireless signal;
• and
• performing, by a statistical processor, a statistical significance test on the swapped pairs at the remote statistical analysis system.

A system for anonymizing a plurality of statistical data for a secure transfer, the system comprising:

• a memory device for storing program code;
• a plurality of isolated patient databases in a plurality of hospital computer processing systems for storing the plurality of statistical data;
• and
• a plurality of privacy processors, each connected to a respective one of the plurality of isolated patient databases of a respective one of the plurality of hospital computer processing systems for running the program code and providing managing access to the respective one of the plurality of isolated patient databases to
• calculate statistical information for each of the plurality of statistical data for a patient;
• aggregate the statistical information;
• create pair lists from each of the plurality of statistical data and target data, the pair lists having a respective member from both the plurality of statistical data and the target data, the target data relating to a specific health status test;
• replace each respective member of the target data by a random number existing in a range of a corresponding one of a plurality of target data bins;
• swap each pair in each of the pair lists in a random order using a randomized number, wherein the randomized number used for swapping is different for different ones of the pair lists;
• and
• a wireless communication device for
• performing the secure transfer for patient privacy protection by sending the swapped pairs over multiple communication paths from the plurality of hospital computer processing systems to a remote statistical analysis system, wherein the wireless communication device transforms the swapped pairs into a wireless signal;
• wherein a statistical processor performs a statistical significance test on the swapped pairs at the remote statistical analysis system.