The article introduces a new method for identifying outliers in nonlinear regression, combining robust nonlinear regression with a false discovery rate (FDR) approach. The method, called ROUT, first fits the data using a robust nonlinear regression method based on the assumption that the scatter follows a Lorentzian distribution. This robust fit serves as a baseline to detect outliers. The ROUT method then removes the identified outliers and performs ordinary least-squares regression on the remaining data. The FDR approach is adapted to handle multiple comparisons, allowing for the detection of one or multiple outliers while controlling the False Discovery Rate (FDR) at a specified threshold (Q). Simulated data analyses show that the ROUT method effectively identifies outliers with a low false discovery rate, even in the presence of multiple outliers. The method is implemented in GraphPad Prism and is designed to improve the accuracy of nonlinear regression by removing influential outliers.The article introduces a new method for identifying outliers in nonlinear regression, combining robust nonlinear regression with a false discovery rate (FDR) approach. The method, called ROUT, first fits the data using a robust nonlinear regression method based on the assumption that the scatter follows a Lorentzian distribution. This robust fit serves as a baseline to detect outliers. The ROUT method then removes the identified outliers and performs ordinary least-squares regression on the remaining data. The FDR approach is adapted to handle multiple comparisons, allowing for the detection of one or multiple outliers while controlling the False Discovery Rate (FDR) at a specified threshold (Q). Simulated data analyses show that the ROUT method effectively identifies outliers with a low false discovery rate, even in the presence of multiple outliers. The method is implemented in GraphPad Prism and is designed to improve the accuracy of nonlinear regression by removing influential outliers.