December 4, 2019
Risk Management Plan
Cybersecurity Awareness Principle (encryption at rest, HTTPS,etc.)
Evidence of The Surveillance Process
Timely Detection and Management of Cybersecurity Threat
2. Software Version Control
3. Clarify if the software incorporates Machine Learning (ML) based algorithm. If yes, please provide the following information:
Input data (data type such as image, patient age, patient’s hospital records, physiological signals, etc.)
Rationale and justification for selection of the source and size of the training and validation date sets used.
Additional intervals for training data update cycle (e.g in months or years) within the projected useful life.
Briefly describe the machine learning system, the model and the algorithm used.
Provide the rationale and selection process of the particular ML model (e.g. based on cross-validation method).
The input data (data type and size) used to generate the corresponding output.
The source and size of training and validation data sets used, including rationale for selection.
The performance of the ML system output (e.g. accuracy, sensitivity, specificity), if any. If the output is in a range, please provide the output in range.
The test protocol and report for verification and validation of the machine learning system, including the acceptance limits.
The interval for training data update cycle (e.g. in months or years), if applicable.
The procedure or plan implemented to trace the software version for subsequent iterations.
Product Owner’s intended clinical workflow. [The clinical workflow should include the any human intervention (if any) recommended by the Product Owner.]
Requirements for Software AI
Central and South America
+65 9067 1432
Design & Development
+65 6909 0396