A Research On Full-Field High-Resolution Cell Morphology Analysis System And Its Application

The research and development of portable medical detection equipment for rapid detection is of great significance for emergency medical malpractice monitoring or remote areas where medical resources are scarce.Cell morphology analyzer is mainly used to observe the morphological changes of various blood cells and is one of the most widely used instruments for clinical examination of hematological diseases.The traditional cell morphology analyzer is inseparable from the amplification of optical microscope,which is expensive,large in size,requires professional experimenters to operate during detection,and does not have the function of convenient and real-time detection.At the same time,due to the optical design principle and spatial bandwidth product limitation,the optical microscope cannot meet the high resolution and large field of view requirements required during cell morphology analysis.In order to solve the above problems,this paper carries out research from both hardware and algorithm.Firstly,the resolution of the microscopic imaging module built by the new imaging device based on the ultra-small pixel size-vertical charge transfer imaging device is simulated and calibrated,which confirms that the actual resolution of the microscopic imaging system is close to the effect of optical microscope under 20 X objective,and has the advantage of large field of view,which can meet the application scenario of cell morphology analysis;then the full-field high-resolution cell morphology analysis system built by the microscopic imaging module expansion collects images of single component cell samples and complex component cell samples,respectively,verifying the imaging function of the system and laying the foundation for the subsequent cell morphology analysis;finally,the morphological analysis of single component cell samples based on the traditional algorithm and the morphological analysis of complex component cell samples based on deep learning are preliminarily realized with blood samples and urine samples,respectively,showing the broad application prospect of this system in the field of clinical examination cell morphology analysis.