CN-121994655-A - Sample analyzer and body fluid sample detection method

CN121994655ACN 121994655 ACN121994655 ACN 121994655ACN-121994655-A

Abstract

The application relates to a sample analyzer and a body fluid sample detection method, which comprise the steps of acquiring optical signal data of a liquid to be detected prepared based on a body fluid sample, wherein the optical signal data comprises at least two of forward scattering optical signals, side scattering optical signals and fluorescent signals, generating a scatter diagram of the liquid to be detected according to the optical signal data, determining an abnormal region in the scatter diagram type and/or the scatter diagram based on the scatter diagram, wherein the scatter diagram type comprises a normal type, an abnormal detection type and a quality abnormal type, and determining at least one of an auxiliary diagnosis result of the body fluid sample and a sample quality detection result according to the scatter diagram type of the abnormal region under the condition of the abnormal type of the scatter diagram type, wherein the scatter diagram characteristic is used for representing particle distribution conditions of the abnormal region. According to the method, the abnormal region of the scatter diagram is determined, the scatter characteristics in the abnormal region of the scatter diagram can be identified and judged in a targeted mode, and the sensitivity and the specificity of the auxiliary diagnosis result and/or the sample quality inspection result are improved.

Inventors

FANG JIANWEI
LIU ZHIZHI

Assignees

深圳市帝迈生物技术有限公司

Dates

Publication Date: 20260508
Application Date: 20241107

Claims (10)

1. A sample analyzer, the sample analyzer comprising: A sampling assembly for obtaining a body fluid sample from a sample container; A sample preparation assembly having at least one reaction cell and a reagent supply; the reagent supply part supplies a hemolysis reagent and a fluorescent reagent to the at least one reaction tank, so that the body fluid sample, the hemolysis reagent and the fluorescent reagent are mixed in the reaction tank to prepare a liquid to be tested; the optical detection assembly is used for irradiating the liquid to be detected flowing through the detection area and collecting various optical signal data generated by cells due to the irradiation of the light; a memory storing a computer program; And the processor is connected with the memory and is used for realizing the following steps when executing the computer program: the method comprises the steps of obtaining optical signal data of a liquid to be detected through an optical detection component, generating a scatter diagram of the liquid to be detected according to the optical signal data, determining a scatter diagram type and/or an abnormal region in the scatter diagram based on the scatter diagram, wherein the scatter diagram type comprises a normal type, a detection abnormal type and a quality abnormal type, and determining at least one of an auxiliary diagnosis result and a sample quality detection result of a body fluid sample according to the scatter diagram characteristic of the abnormal region and the scatter diagram type under the condition that the scatter diagram type is not the normal type, wherein the scatter diagram characteristic is used for representing particle distribution conditions of the abnormal region.
2. A method of testing a body fluid sample, the method comprising: Acquiring optical signal data of a liquid to be detected prepared based on a body fluid sample, wherein the optical signal data comprises at least two of forward scattering optical signals, side scattering optical signals and fluorescent signals; generating a scatter diagram of the liquid to be detected according to the optical signal data; Determining a scatter diagram type and/or an abnormal region in the scatter diagram based on the scatter diagram, wherein the scatter diagram type comprises a normal type, a detected abnormal type and a quality abnormal type; And under the condition that the scatter diagram type is not the normal type, determining at least one of an auxiliary diagnosis result and a sample quality inspection result of the body fluid sample according to the scatter diagram characteristic of the abnormal region and the scatter diagram type, wherein the scatter diagram characteristic is used for representing the particle distribution condition of the abnormal region.
3. The method according to claim 2, wherein the determining a scatter plot type and/or an abnormal region in the scatter plot based on the scatter plot comprises: Inputting the scatter diagram into an AI model for recognition to obtain the scatter diagram type; Obtaining a distribution feature map according to the identification information of the scatter diagram type in the AI model under the condition that the scatter diagram type is not the normal type; and screening out the region of interest from the distribution characteristic diagram as the abnormal region.
4. A method according to claim 3, wherein said obtaining a distribution profile from said scatter plot type in the event that said scatter plot type is not of said normal type comprises: Obtaining a characteristic layer and a predicted value through forward propagation according to the scatter diagram type; The predicted value of the scatter diagram type is back propagated to obtain gradient information which is transmitted back to the feature layer, wherein the gradient information characterizes the sensitivity degree of a channel in the feature layer to the scatter diagram type; According to the gradient information of each channel in the feature layer, determining a weight vector corresponding to each channel; And carrying out weighted summation according to the weight vector and the corresponding channel to obtain a distribution characteristic diagram.
5. The method of claim 2, wherein determining at least one of a secondary diagnostic result and a sample quality test result of the body fluid sample based on the scatter signature of the abnormal region and the scatter pattern type comprises: Acquiring a body fluid type of the body fluid sample in a case where the scatter pattern type is the detection abnormality type; acquiring the scatter characteristics matched with the body fluid type based on the abnormal region; and determining whether the auxiliary diagnosis result is abnormal according to the scattered point characteristics matched with the body fluid type.
6. The method of claim 4, wherein said determining whether said auxiliary diagnostic result is abnormal as a result of detection of said body fluid sample based on said scatter signature matching said body fluid type, further comprises: acquiring particle classification information obtained based on the scatter diagram; And according to at least one of target particles matched with the body fluid type, difference information of the scattered point characteristics and the particle classification information and the scattered point characteristics matched with the body fluid type in the particle classification information, matching with a preset diagnosis function, determining that the auxiliary diagnosis result is abnormal in the detection result under the condition of successful matching, and outputting disease diagnosis information corresponding to the diagnosis function.
7. The method of claim 2, wherein determining at least one of a secondary diagnostic result and a sample quality test result of the body fluid sample based on the scatter signature of the abnormal region and the scatter pattern type comprises: acquiring particle classification information obtained based on the scatter diagram; And under the condition that the scatter diagram type is the quality abnormality type, matching with a preset quality discriminant function according to at least one of difference information of the scatter characteristics and the particle classification information and the scatter characteristics matched with the body fluid type, and under the condition that the matching is successful, determining that the sample quality of the body fluid sample is abnormal as a sample quality of the sample quality inspection result.
8. The method according to any one of claims 2 to 7, further comprising: Outputting disease diagnosis information matched with the body fluid type of the body fluid sample based on the auxiliary diagnosis result when the auxiliary diagnosis result is that the detection result of the body fluid sample is abnormal; And outputting prompt information based on the sample quality inspection result when the sample quality of the body fluid sample is abnormal, wherein the prompt information is used for prompting that the body fluid sample is stored and/or prepared unqualified.
9. The method of claim 3, wherein the training process of the AI model comprises: the method comprises the steps of obtaining a training sample set, wherein the training sample set comprises a scatter diagram of a normal sample, a scatter diagram of a sample with abnormal detection and a scatter diagram of a sample with abnormal quality; According to the abnormal type, performing mode division on the scatter diagram of the normal sample in the training sample set, and combining the scatter diagram of the normal sample to obtain a divided training sample set; and training an initial model by using the divided training sample set, and obtaining the trained AI model when the initial model meets the iteration ending condition.
10. A method of testing a body fluid sample, the method comprising: Acquiring optical signal data of a liquid to be detected prepared based on a body fluid sample, wherein the optical signal data comprises at least two of forward scattering optical signals, side scattering optical signals and fluorescent signals; generating a scatter diagram of the liquid to be detected according to the optical signal data; determining an abnormal region in the scatter plot based on the scatter plot; And determining an auxiliary diagnosis result of the body fluid sample according to the scattered point characteristic of the abnormal region, wherein the scattered point characteristic is used for representing the particle distribution condition of the abnormal region.

Description

Sample analyzer and body fluid sample detection method Technical Field The application relates to the technical field of in-vitro auxiliary diagnosis, in particular to a sample analyzer and a body fluid sample detection method. Background Body fluid detection is of great significance to screening and diagnosis of patients suffering from diseases. At present, the clinically used body fluid detection method mainly comprises routine body fluid detection, biochemical body fluid detection and body fluid tumor marker detection. However, conventional detection of body fluid and biochemical detection of body fluid are usually performed by defining specific cell particles in a gating manner, for example, a high fluorescence region is designated to identify tumor cells, however, cells with high fluorescence signals not only comprise tumor cells, but also normal cells such as mesothelial cells and macrophages, and particularly the mesothelial cells have the greatest influence on the detection of tumor cells, so that the sensitivity and the specificity of body fluid detection are relatively low, and the detection cost of body fluid tumor markers is high and depends on the diagnosis experience of medical staff. Therefore, how to achieve higher sensitivity, specificity and popularity of body fluid detection is a problem to be solved. Disclosure of Invention In view of the above, it is necessary to provide a sample analyzer and a method for detecting a body fluid sample, which can ensure sensitivity and specificity of body fluid detection. In a first aspect, the present application provides a sample analyzer comprising: A sampling assembly for obtaining a body fluid sample from a sample container; A sample preparation assembly having at least one reaction cell and a reagent supply; the reagent supply part supplies a hemolysis reagent and a fluorescent reagent to the at least one reaction tank, so that the body fluid sample, the hemolysis reagent and the fluorescent reagent are mixed in the reaction tank to prepare a liquid to be tested; the optical detection assembly is used for irradiating the liquid to be detected flowing through the detection area and collecting various optical signal data generated by cells due to the irradiation of the light; a memory storing a computer program; And the processor is connected with the memory and is used for realizing the following steps when executing the computer program: the method comprises the steps of obtaining optical signal data of a liquid to be detected through an optical detection component, generating a scatter diagram of the liquid to be detected according to the optical signal data, determining a scatter diagram type and/or an abnormal region in the scatter diagram based on the scatter diagram, wherein the scatter diagram type comprises a normal type, a detection abnormal type and a quality abnormal type, and determining at least one of an auxiliary diagnosis result and a sample quality detection result of a body fluid sample according to the scatter diagram characteristic of the abnormal region and the scatter diagram type under the condition that the scatter diagram type is not the normal type, wherein the scatter diagram characteristic is used for representing particle distribution conditions of the abnormal region. In a second aspect, the present application also provides a method of testing a body fluid sample, the method comprising: Acquiring optical signal data of a liquid to be detected prepared based on a body fluid sample, wherein the optical signal data comprises at least two of forward scattering optical signals, side scattering optical signals and fluorescent signals; generating a scatter diagram of the liquid to be detected according to the optical signal data; Determining the scatter diagram type and/or an abnormal region in the scatter diagram based on the scatter diagram, wherein the scatter diagram type comprises a normal type, a detected abnormal type and a quality abnormal type; And under the condition that the scatter diagram type is not the normal type, determining at least one of an auxiliary diagnosis result and a sample quality inspection result of the body fluid sample according to the scatter diagram characteristic of the abnormal region and the scatter diagram type, wherein the scatter diagram characteristic is used for representing the particle distribution condition of the abnormal region. In one embodiment, the determining the scatter pattern type and/or the abnormal region in the scatter pattern based on the scatter pattern includes: Inputting the scatter diagram into an AI model for recognition to obtain the scatter diagram type; Obtaining a distribution feature map according to the identification information of the scatter diagram type in the AI model under the condition that the scatter diagram type is not the normal type; and screening out the region of interest from the distribution characteristic diagram as the abnormal region. In on