Approximate-model closed-loop minimal sampling method for HIV viral-load minima detection.

Area of Science:

• Mathematical Modeling and Simulation
• Infectious Disease Dynamics
• Pharmacometrics

Background:

• Monitoring viral load is crucial for managing Human Immunodeficiency Virus (HIV) infection.
• Therapy transitions in HIV management require careful monitoring to prevent viral rebound.
• Optimizing sampling frequency can reduce patient burden and healthcare costs.

Purpose of the Study:

• To develop a closed-loop method for identifying the minimum point of a viral rebound curve in HIV patients.
• To optimize the timing of viral load measurements during HIV therapy transitions.
• To reduce the number of required patient samples compared to traditional fixed-interval sampling.

Main Methods:

• A reduced approximate solution was fitted to viral load measurements using a Simulated Annealing direct search algorithm.
• Gaussian white noise was introduced to simulate measurement variability, generating a family of fits.
• A safety tolerance measure was applied to the family of fits to determine the optimal next sampling time.

Main Results:

• The closed-loop method demonstrated robust performance on noisy data generated from identified patient models.
• The proposed method significantly reduced the number of samples needed compared to fixed-interval sampling.
• The approach effectively determined the minimum of the viral rebound curve.

Conclusions:

• The developed closed-loop method provides an efficient and robust approach for monitoring viral load in HIV patients.
• This method optimizes sampling strategies, potentially leading to reduced patient monitoring frequency and associated costs.
• The technique shows promise for personalized medicine in HIV management.