Theoretical Study of Exponential Best-Fit: Modeling hCG for Gestational Trophoblastic Disease

With the removal of the hydatidiform mole it has been shown that the human chorionic gonadotropin (hCG) hormone levels drop exponentially in women diagnosed with Gestational Trophoblastic Disease (GTD). This papers aims to introduce a new method at forecasting the decrease of the hCG levels as this could reduce the number of weekly blood test that a patient would require throughout the one year of monitoring. The hCG levels are modeled as a vertically shifted exponential curve, and this paper proposes and demonstrates a mathematical solution to finding the best parameters for this model. The method is validated using synthetic data as well as real data, and the results show that it is reliable, with decent accuracy and speed.