Research And Implementation Of The Recognition Algorithm Of Blood Cells Pulse Using FPGA

Blood cell counting plays an important role in the modern clinical diagnosis. It is the basis to recognize the diseases or not and its classification. With the development of the technique of the pattern recognition, blood cell recognition is becoming the hotspot in the biomedicine research area. Blood parameters contain mainly hemoglobin (HGB) concentration, red blood cell count, white cell count (WBC) and platelet (PLT) and their volume distributions, etc. These parameters have become one of the most important and useful ways to diagnose the diseases.Coulter principle is widely used in blood corpuscle analyzer. When the blood cell passes the sensor, bringing M shape signals with noise disturbance,that will make it difficult to identify the sequential blood cell signals. The traditional identification way of blood cell signal which is based on Coulter principle is achieved by using analog circuit, but it has risk of misrecognition to the M shape signals. After analyzing the traditional signal recognition way, this thesis focus on the Digital Pulse Process technique, by this way to identify and count the blood cell and calculate the blood cell volume distribution.The Digital Pulse Process technique is according to the picked blood cell signals to compute some parameters, such as the pulse width, height and integration, then to identify all sorts of blood cell signals. This thesis proposes a novel digital particle size recognition (DPR) algorithm which is based on Coulter principle. This algorithm can identifies a single blood cell pulse and M shape signals, avoids risk of misrecognition to the signals efficiently which are not from the blood cell, then raises the accuracy of measurement. It is implemented on FPGA by hardware description language. We create the blood cell histogram analysis distribution recording circuits which basing on FPGA and get the measurement result about the blood cell volume distribution. Such circuits include mainly blood cell signal quantization, filtration, recognition and calculation, etc. The measurement result proves that the DPR has high recognition efficiency and it can be applied to the blood cell counting equipment.