How Does The CM-Connect Work?

A surface level overview of the Collective Minds CM-Connect software

The common configuration of this software allows for anonymization/ pseudonymization of DICOM files that are then pushed to our platform for education, research, or clinical collaboration purposes....but how?

The CM-Connect does not fetch data, it listens for data being sent to it typically from a PACS archive. This means that it does not access any of your databases on its own. Rather, it is listening or waiting to receive data from a PACS or a DICOM node that is configured to send to it. Operating in this way ensures that only the intended data is used and sent to the platform.

Once data is sent to the CM-Connect, the data is temporarily stored for processing on the computer or server that the CM-Connect is installed on. This is why we ask that hard disk requirements be adhered to. Then the anonymization/ pseudonymization process will begin. The original data will be held here until this process is completed.

Anonymization / Pseudonymization of DICOM data

The objective of the Default Pseudonymization Profile is to remove any data which can be traced back to an individual Data Subject (patient). Unique identifiers in the form or keyed hashes are kept within each study to enable linking back to the original patient and study.

An important aspect of the pseudonymization process is that all UIDs (Unique Identifiers) are replaced with non-deterministic entries during each pseudonymization process. Hence, in order for a Study to remain intact as one study after the process, the sending of the DICOM data to the CM-Connect needs to happen at the Study level.

On the contrary, if a study is sent twice to the CM-Connect, two individual copies of the pseudonymized study will come out on the other end.

The process of pseudonymization removes all direct identifiers such as name, birthdate, phone number, etc. from the DICOM images and removes any metadata tag which could indirectly enable the identification of the Data Subject.

More details about the exact data modifications which take place under the Default Pseudonymization Profile can be found in our Privacy Policy cmrad.com/privacy.

Beyond the Default Pseudonymization Profile, we have the ability to create Custom Pseudonymization Profiles. Any such modifications will only be configured upon your specific request.

After the anonymization/ pseudonymization process is completed, the anonymized/ pseudonymized data will be sent to the platform and the CM-Connect will clean itself up by deleting the original data. This process secures patient data within your institution and behind your firewalls as the original data never leaves.

Connected Hospital

Further, as part of the pseudonymization process, the CM-Connect enables traceability of an institutions data into the cloud to comply with regulatory requirements such as "The right to be forgotten" in GDPR. We use a keyed hashing technology where the data owner (you), holds a key which, along with the original patient information, allows you to trace your own data stored within Collective Minds.

Therefore, if your institution wants to trace a patient ID in our database, you can do so using the Patient Hash Calculator (see below). We do not store the Patient IDs (they are hashed in the CM-connect at institution level, before transfer), but we store the hash secret, and we make it available to the members of the respective institution through this calculator.

Input is Patient ID, and output will be the corresponding Hashed ID in our database.

A depiction of the keyed hashing process used to securely trace your data.