A data driven diagnosis tool for thyroid hormones

Thyroid hormones play a significant role in human health. Understanding their dynamics is crucial to diagnoses and maintaining the well-being of the thyroid. In this work we propose a data driven algorithm to detect a fixed point and a limit cycle in real data for thyroid hormones. This algorithm finds the maximum frequency point (fixed point) and extracts a smooth ellipse (limit cycle) from the data. These features characterize various data sets and provide interesting insights to differentiate healthy from malfunctioning thyroid data. This scheme which is backed by a solid dynamical analysis determines the size, orientation and location of a detected limit cycle and provides information about the behavior of the thyroid in its various normal and abnormal conditions. This algorithm does not require tuning any ad-hoc parameters. This approach could lead to an effective way of implementing a personal treatment strategy, and a control system to improve the performance of the thyroid.