Using a Calculated Pulse Rate with an Artificial Neural Network to Detect Irregular Interbeats.

Heart rate is an important clinical measure that is often used in pathological diagnosis and prognosis. Valid detection of irregular heartbeats is crucial in the clinical practice. We propose an artificial neural network using the calculated pulse rate to detect irregular interbeats. The proposed system measures the calculated pulse rate to determine an "irregular interbeat on" or "irregular interbeat off" event. If an irregular interbeat is detected, the proposed system produces a danger warning, which is helpful for clinicians. If a non-irregular interbeat is detected, the proposed system displays the calculated pulse rate. We include a flow chart of the proposed software. In an experiment, we measure the calculated pulse rates and achieve an error percentage of < 3% in 20 participants with a wide age range. When we use the calculated pulse rates to detect irregular interbeats, we find such irregular interbeats in eight participants.

Double-Brillouin-frequency spaced multiwavelength generation in one loop structure.

We demonstrate a simple and unique double-Brillouin-frequency spaced multiwavelength structure. It uses the reflection property of an isolator and fiber loop mirror which is made of a 50:50 coupler. This scheme has only one loop, instead of two or more loops, to separate the odd and even Stokes waves generated by Brillouin scattering. The light going into loop cavity can be absorbed twice to produce a Brillouin Stokes wave twice in one round loop, so a double-Brillouin-frequency spaced multiwavelength harmonics wave can be achieved in the output of the structure. Finally, our experiment successfully achieves a double-Brillouin-frequency space of about 0.1475 and 0.1734 nm in two different type structures.