Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Towards Developing Safety Assurance Cases for Learning-Enabled Medical Cyber-Physical Systems

Maryam Bagheri, Josephine Lamp, Xugui Zhou, Lu Feng, Homa Alemzadeh

TL;DR

A safety assurance case for ML controllers in learning-enabled MCPS, with an emphasis on establishing confidence in the ML-based predictions, is developed and a detailed analysis by implementing a deep neural network for the prediction in APS.

Abstract

Machine Learning (ML) technologies have been increasingly adopted in Medical Cyber-Physical Systems (MCPS) to enable smart healthcare. Assuring the safety and effectiveness of learning-enabled MCPS is challenging, as such systems must account for diverse patient profiles and physiological dynamics and handle operational uncertainties. In this paper, we develop a safety assurance case for ML controllers in learning-enabled MCPS, with an emphasis on establishing confidence in the ML-based predictions. We present the safety assurance case in detail for Artificial Pancreas Systems (APS) as a representative application of learning-enabled MCPS, and provide a detailed analysis by implementing a deep neural network for the prediction in APS. We check the sufficiency of the ML data and analyze the correctness of the ML-based prediction using formal verification. Finally, we outline open research problems based on our experience in this paper.

Towards Developing Safety Assurance Cases for Learning-Enabled Medical Cyber-Physical Systems

TL;DR

A safety assurance case for ML controllers in learning-enabled MCPS, with an emphasis on establishing confidence in the ML-based predictions, is developed and a detailed analysis by implementing a deep neural network for the prediction in APS.

Abstract

Machine Learning (ML) technologies have been increasingly adopted in Medical Cyber-Physical Systems (MCPS) to enable smart healthcare. Assuring the safety and effectiveness of learning-enabled MCPS is challenging, as such systems must account for diverse patient profiles and physiological dynamics and handle operational uncertainties. In this paper, we develop a safety assurance case for ML controllers in learning-enabled MCPS, with an emphasis on establishing confidence in the ML-based predictions. We present the safety assurance case in detail for Artificial Pancreas Systems (APS) as a representative application of learning-enabled MCPS, and provide a detailed analysis by implementing a deep neural network for the prediction in APS. We check the sufficiency of the ML data and analyze the correctness of the ML-based prediction using formal verification. Finally, we outline open research problems based on our experience in this paper.
Paper Structure (14 sections, 7 figures, 6 tables)

This paper contains 14 sections, 7 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Artificial Pancreas Systems
  3. Overall Safety Assurance Case
  4. Safety Assurance Case Template for learning-enabled controllers in MCPS
  5. Instantiating Learning-Enabled Controller Assurance Case for APS
  6. Safety Assurance Case for the Glucose Prediction Component
  7. Sufficiency of the ML Development
  8. Sufficiency of the ML Model
  9. Sufficiency of the ML Data
  10. Concrete Evidence
  11. Open Research Problems
  12. Conclusion
  13. Appendix
  14. Glucose Prediction Model

Figures (7)

  • Figure 1: The structure of APS (modified from apsStructure). The APS controller consists of a prediction algorithm to predict the BG values and a controller algorithm to adjust the insulin dosage.
  • Figure 2: A safety assurance case template for a learning-enabled controller in MCPS.
  • Figure 3: A general safety assurance case for the ML glucose prediction component of APS.
  • Figure 4: Argument to ensure the sufficiency of the ML glucose prediction model.
  • Figure 5: Argument to ensure the sufficiency of the ML data.
  • ...and 2 more figures