This technical note presents a sensor fault detection and isolation scheme for a class of Lipschitz nonlinear systems with unstructured modeling uncertainty. It significantly extends previous results by considering a class of system nonlinearities which are modeled as functions of the system input and partially measurable state variables. A new sensor fault diagnosis method is developed using adaptive estimation techniques. Adaptive thresholds for fault detection and isolation are derived, and several important properties are investigated, including robustness, stability and learning capability, and fault isolability. A robotic example is used to show the effectiveness of the method.