01
Age and sex
TraceRead is for adults, 18 and over, so age matters. Some limits that TraceRead AI applies when it reads an ECG also depend on age and sex. TraceRead asks for no name.
Privacy and security
TraceRead asks for very little about a patient
TraceRead needs a patient's age and sex and a photo of the printed ECG with the header covered. It keeps them in one place, never sends them to an outside AI service and never sells them.
In short
- Age and sex only. Never a name.
- The header is covered on the phone.
- Kept in the AWS Cape Town region.
- Never sent to an outside AI service.
- TraceRead staff cannot see patient data by default.
What TraceRead needs
Three things, and no more
That is all TraceRead takes for each ECG.
02
A photo of the printed ECG
The technician photographs the page on a phone. Each photo is kept near 1.5 MB. The server checks its type and size when it arrives.
03
The header, covered
Before sending, the technician drags a box over the printed header, where a machine prints the patient's details. The box is painted into the photo on the phone, so the covered part never leaves it. The technician chooses what to cover, so cover anything that names the patient.
Where it is kept
In the AWS Cape Town region
Patient data is kept in Amazon Web Services' Cape Town region, called af-south-1.
This is hosting that TraceRead pays for, like renting space for servers. Amazon Web Services has no say in how TraceRead works and does not endorse it.
Cape Town is in South Africa, so patient data is held outside Nigeria. The Nigeria Data Protection Act has rules for sending personal data out of the country. Ask your own data protection adviser how they apply to your organisation. TraceRead signs a data processing agreement with each customer.
The reading runs there too. In testing on a small server in Cape Town with no graphics card, 95 in 100 ECGs were read in 7.85 s or less (Spike S7, 97 photos made by computer).
Who can see it
Your people, and no one else by default
Patient data is for the people who capture and read your ECGs.
Your organisation
The technicians who capture the photos and the doctors who review the AI first reads.
TraceRead staff
TraceRead staff cannot see patient data by default.
The patient
Nothing reaches a patient until a doctor has checked and signed it.
What is never done
Three things TraceRead never does
Never sent to an outside AI service
Every step of an AI first read runs on TraceRead's own software, on servers that TraceRead runs in Cape Town. No generative AI writes clinical text.
Never a patient's name
TraceRead needs age and sex. It does not ask for a name, an address, a phone number or an ID number.
Never sold
TraceRead does not sell patient data.
Who is responsible
You decide, TraceRead handles
Under the Nigeria Data Protection Act, health data is sensitive personal data. It needs a clear legal basis and safeguards.
Controller
Your organisation is the data controller
It decides why and how patient data is used, and it is accountable for that use.
Processor
TraceRead is the data processor
It handles the data for you, on your instructions, under a data processing agreement that TraceRead signs with each customer.
This website
No patient data is on this website
Everything on this website comes from public research ECGs. TraceRead has not used patient data from any clinic.
Public research ECGs are recordings published for research, with no names. They come from Germany (PTB-XL), China (Chapman Shaoxing Ningbo) and Brazil (CODE-15%). Data and software credits
Ask us how your data is handled
Tell us what your data protection officer needs to know, and what your clinicians need to see.