Artificial intelligence blood smear cell labeling system and method

Abstract

The invention relates to the technical field of medical auxiliary diagnosis, and discloses an artificial intelligence blood smear cell labeling system and method. The artificial intelligence blood smear cell labeling system comprises an observation unit, a labeling unit, an analysis and classification unit and a central control unit; wherein the observation unit comprises a microscope, a CCD camera and a microscopic control module; the labeling unit comprises a display module and a labeling operation module, and the labeling operation module comprises a labeling line adding module and a standard line deleting module; and the analysis and classification unit comprises a cell distribution statistical template, a cell quantity statistical module, a purple cell statistical module, an orange cell statistical module, a huge cell statistical module, a cell cluster analysis module, a benign cell statistical module and an abnormal cell statistical module. According to the method, the multiple of the microscope can be automatically adjusted, the clearest image in the same visual field image is found around the set multiple and cells in the image are marked, wherein the benign cells enter a benign disease classification system, and abnormal cells enter a pathology classification system.

Description

technical field [0001] The invention relates to the technical field of medical aided diagnosis, in particular to an artificial intelligence blood smear cell labeling system and method. Background technique [0002] In recent years, the integration of artificial intelligence technology and the medical and health field has continued to deepen. With the gradual maturity of technologies such as voice interaction, computer vision and cognitive computing in the field of artificial intelligence, the application scenarios of artificial intelligence have become more and more abundant, and artificial intelligence technology has gradually become An important factor affecting the development of the medical industry and improving the level of medical services. Its application technologies mainly include: voice recording of medical records, medical image-aided diagnosis, drug research and development, medical robots, intelligent analysis of personal health big data, etc. [0003] At pres...

AI Technical Summary

Problems solved by technology

At present, artificial intelligence technology used in medical images for artificial intelligence diagnosis mainly includes artificial intelligence-assisted diagnosis applied to ordinary medical images or pathological images; among them, artificial intelligence-assisted diagnosis applied to ordinary medical images is mainly for CT images, magnetic resonance images Or ultrasonic images and other small-sized images, through artificial intelligence technology, the analysis of the whole image can be completed on a single machine at one time, and the possible lesion areas in the image can be identified to assist doctors in making further diagnoses, but This method can only be applied to ordinary medical images, and its corresponding analysis model cannot be transferred to the diagnosis of pathological images; while the artificial intelligence-assisted diagnosis applied to pathological images is usually combined with special pathological diagnosis and analysis equipment such as digital pathological scanners. It digitally processes the pathological images and uploads them to the pathological diagnosis and analysis equipment, and then completes the analysis of the pathological images locally to assist doctors in making further diagnoses. However, this method is only applicable to pathological images with ordinary resolutions. Unable to process pathological full-scan images with ultra-high resolution

[0004] For this reason, the patent application number CN201910081438.X discloses an artificial intelligence pathological labeling system, although the artificial intelligence pathological labeling system can simultaneously solve the problem of reading multi-level, large-size and ultra-high-resolution full-scan pathological images and labeling problems, it can realize the fast storage and reading operation of the full-scan pathological image, so that the reading speed of the full-scan pathological image is no longer limited to the hardware equipment conditions of the system, and the system also adopts a more user-friendly and A convenient image annotation tool, which can assist doctors to conveniently and quickly annotate suspected lesion areas on full-scan pathological images, and can also store the annotated and modified full-scan pathological images in real time, thereby improving artificial intelligence image analysis technology assistance. The efficiency and accuracy of the doctor's diagnosis results; however, there are still the following problems in this solution: the system is only based on the pathological annotation of the full-scan pathological image, and cannot observe and label blood smears at different multiples, and cannot automatically Adjust the magnification of the microscope, and find the clearest image in the same field of view around the set magnification, and mark the cells in the image, and cannot automatically enter benign cells into the benign disease classification system, and abnormal cells into pathology classification system

Images