Full-automatic view retrieving method

Abstract

The invention relates to a full-automatic view retrieval method, which comprises the following steps: S1, after a scanning module of a microscopic scanner scans a target view, recording the coordinate position of the target view and issuing the coordinate position to a lower computer, and S2, for a doubt view, using a retrieval function, the lower computer retrieves the visual field of the question according to the actual position and displays the visual field on a display screen of the upper computer, and a displayed picture is recorded as a preliminary retrieved visual field; an auditor can perform manual focusing and picture displacement adjustment on the preliminary retrieval view field and observe the preliminary retrieval view field, so that the purpose of manual auditing is achieved; under special circumstances, the auditor uses an index function and uses a built-in algorithm to index and match until a question view is matched. By using the method, the auditor can quickly find the visual field corresponding to the doubt visual field to perform observation and judgment again, and the accuracy of the detection result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of scanning film reading, and specifically relates to a fully automatic visual field retrieval method. Background Art

[0002] As disclosed in the invention patent with the authorization announcement number CN111551756B, a fully automatic slide preparation, staining and scanning system includes several major parts such as a sample bottle conveying device, a slide preparation device, a staining device, a slide storage device, and a scanning device. Each part is arranged in a housing to form an integrated machine for sampling, slide preparation, staining, and scanning. It has functions such as automatic sampling, automatic smearing, automatic staining, and automatic film reading, and is applicable to tuberculosis samples such as sputum, bronchial washing fluid, ascites, urine, and cerebrospinal fluid, meeting the needs of tuberculosis smear diagnosis in tuberculosis laboratories at all levels. Some of the scanning results are uncertain. Commonly, they are classified as negative, suspected positive, and positive. No matter which result it is, it can be questioned by the doctor responsible for reviewing the film, or some of the result pictures are not clear and need to be checked again. For example, for suspected positive cases, there is a need to observe them again. Then, how to find the required target visual field for re-observation is a problem that needs to be solved; if manually examined through an eyepiece, it is very difficult to find the required visual field. Summary of the Invention

[0003] The purpose of the present invention is to provide a fully automatic visual field retrieval method to meet the requirement of being able to find the target visual field in question for re-observation and further determine the result manually.

[0004] To achieve the above purpose, the present invention adopts the following technical solutions:

[0005] A fully automatic visual field retrieval method includes the following steps.

[0006] Step S1: After the scanning module of the microscope scanner scans out the target visual field, record its coordinate position and send it to the lower computer, and upload the pictures of each target visual field to the upper computer.

[0007] Step S2: Based on the upper computer, for a questionable visual field, the reviewer can use the "retrieval" function. The lower computer retrieves the questionable visual field according to the actual position and displays it on the display screen of the upper computer. The displayed picture is recorded as the initially retrieved visual field. The initially retrieved visual field and the questionable visual field are simultaneously displayed on the display screen for the reviewer to observe.

[0008] If the initially retrieved visual field is consistent with the questionable visual field, then execute Step S3: If the initially retrieved visual field is significantly inconsistent with the questionable visual field, then execute Step S4:

[0009] Step S3: The reviewer can manually focus and adjust the picture displacement of the initially retrieved visual field to further observe the initially retrieved visual field, achieving the purpose of manual review.

[0010] Step S4: The reviewer uses the "index" function and performs indexing using the built-in template matching algorithm until the questionable field of view is matched.

[0011] Further, in the step S4, the template matching algorithm takes the preliminary retrieved field of view as the center point, traverses an N*N rectangular area, performs template matching on multiple images in this area and the questionable field of view, and displays the matching position on the image scanned from the N*N rectangular area in real time by drawing a red solid circle. The reviewer zooms in on this area, moves to this position by double-clicking on the field of view where the red solid circle is located, and the field of view displayed in real time is the preliminary retrieved field of view.

[0012] Further, N can be an integer within 5 to 15.

[0013] Further, the host computer can be a computer or an intelligent terminal with a touch screen. The reviewer can click on the corresponding function keys and operation screens using a mouse, or touch the corresponding function keys and operation screens with a hand.

[0014] Further, during the process of retrieving the questionable field of view, the lower computer can automatically control the focusing and translation of the scanning device.

[0015] A computer device includes a processor and a memory. The memory is configured to store a computer program. When the computer program is executed by the processor, the processor implements any of the methods described above.

[0016] A microscope slide review and inspection system includes the above computer device and a microscopic scanner connected to the computer device.

[0017] Advantages of the present invention:

[0018] When reviewing slides, for a questionable target field of view, the reviewer can use the retrieval function to quickly find the corresponding target field of view, which is defined as the preliminary retrieved field of view. The reviewer manually adjusts the focus and the screen displacement of it, and further observes this preliminary retrieved field of view through fine adjustment, so as to achieve the purpose of manually reviewing and further determining the detection result, thereby improving the accuracy rate.

[0019] For carriers that have been taken off the shelf and then re-listed, due to coordinate position errors, it may be impossible to directly retrieve the field of view. At this time, the search function can be used and calculations can be performed with the help of a certain algorithm to find the original field of view for the reviewer to further observe. Description of the Drawings

[0020] Figure 1 is the logic block diagram of the full-automatic field of view retrieval method of the present invention. Detailed Embodiments

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0022] Embodiment:

[0023] As Figure 1 shown, a fully automatic field of view retrieval method includes the following steps.

[0024] Step S1: After the scanning module of the microscopic scanner scans out the target field of view, record its coordinate position and send it to the lower computer, and upload the pictures of each target field of view to the upper computer, that is, upload the scanning result. The microscopic scanner can adopt the fully automatic slide making, staining and scanning system cited in the background technology of this application, or other existing automatic microscopic scanners. This application focuses on providing a method, and the detection target is not limited to a certain kind, such as tuberculosis bacteria.

[0025] Step S2: Based on the upper computer, for a doubtful field of view, the reviewer can use the "retrieval" function. The lower computer retrieves the doubtful field of view according to the actual position and displays it on the display screen of the upper computer. The reviewer can slide to view and conduct a review. During the process of the lower computer retrieving the doubtful field of view, the focusing and translation of the scanning device can be automatically controlled.

[0026] The reviewer in this embodiment can be a doctor. When the doctor reads the film and has doubts about some of the results determined by automatic scanning, it can be defined as a "doubtful field of view". The doctor can view the corresponding field of view again, that is, let the microscopic scanner scan and display it again.

[0027] The displayed picture is recorded as the "preliminary retrieved field of view". The "preliminary retrieved field of view" and the "doubtful field of view" are simultaneously displayed on the display screen (there are two windows) for the reviewer to compare and observe. The "preliminary retrieved field of view" is the field of view that appears again when the "retrieval" function is manually executed.

[0028] If the preliminary retrieved field of view is consistent with the doubtful field of view, which can be understood as completely consistent or having a relatively high similarity, then step S3 is executed. If the preliminary retrieved field of view is significantly inconsistent with the doubtful field of view, that is, not similar, then step S4 is executed.

[0029] Step S3: The reviewer can manually focus and adjust the screen displacement for the initially retrieved field of view, and further observe the initially retrieved field of view through fine-tuning to achieve the purpose of manually reviewing and further determining the detection result. The displacement adjustment can be along the X-axis and Y-axis, and the focusing is along the Z-axis direction. The host computer can be a computer or an intelligent terminal with a touch screen. The reviewer can use the mouse to click the corresponding function keys and operate the screen for displacement adjustment, or use the hand to touch the corresponding function keys and operate the screen. By adjusting the screen, the reviewer can observe more clearly. For example, for tuberculosis bacteria, its shape can be observed more clearly, so that the detection result can be judged more accurately, improving the accuracy rate.

[0030] Step S4: The reviewer uses the "indexing" function and performs indexing using the built-in template matching algorithm until the questionable field of view is matched.

[0031] In the template matching algorithm described in Step S4, with the initially retrieved field of view as the center point, an N*N rectangular area is traversed, and template matching is performed between multiple images in this area and the questionable field of view. The matching position is displayed in real time as a red solid circle on the image scanned from the N*N rectangular area. The reviewer zooms in on this area, moves to this position by double-clicking on the field of view where the red solid circle is located, and the field of view displayed in real time is the initially retrieved field of view. Then, manual focusing and displacement adjustment can be continued to further observe the initially retrieved field of view.

[0032] N can be an integer within 5 to 15. For example, N can be 5, 9, 11, etc. In this embodiment, N is preferably selected as 11.

[0033] Generally, using the "retrieval" function can quickly find the target field of view corresponding to the questionable field of view. The reason for Step S4 is that there is a certain deviation in the coordinate position, which is caused by the physical position deviation when the target object carrier is removed from the microscopic scanner and then re-inserted into the microscopic scanner before automatic retrieval.

[0034] The template matching algorithm is a prior art and a simple method, such as the sum of squared differences; a method with higher accuracy is preferably used, such as the method of using the normalized cross-correlation coefficient for cross-correlation, which is a normalized version of the correlation coefficient matching method: the correlation between the difference between the original image and its mean value and the difference between the template and its mean value is used for matching, and the result is normalized. The similarity between the original image and the template is measured by calculating the normalized correlation coefficient. The best matching result is 1, indicating a perfect match. The worse the match, the smaller the matching value.

[0035] A computer device includes a processor and a memory. The memory is configured to store a computer program; when the computer program is executed by the processor, the processor implements the method as described above.

[0036] A microscope slide review system includes the above-mentioned computer device and a micro scanner connected to the computer device. The micro scanner is connected to the computer device through a data cable. The computer device may include a display screen and corresponding operating devices (such as a mouse and a keyboard), or a touch-screen intelligent terminal may also be used. The scanning result of the micro scanner can be displayed on the display screen, and the reviewer can review the result.

[0037] The present invention is not limited to the above-mentioned best implementation mode. Anyone can obtain other various forms of products under the inspiration of the present invention. However, no matter what changes are made in its shape or structure, as long as it has the same or similar technical solutions as this application, it falls within the protection scope of the present invention.

Claims

1. A fully automatic visual field retrieval method, characterized in that: The following steps are included: Step S1: After scanning the target field of view, the scanning module of the microscope scanner records its coordinate position and sends it to the lower computer, and uploads the image of each target field of view to the upper computer; Step S2: Based on the host computer, the auditor can use the "retrieve" function for the questionable visual field. The lower computer retrieves the questionable visual field according to the actual position and displays it on the display screen of the host computer. The displayed picture is recorded as the preliminary retrieved visual field. The preliminary retrieved visual field and the questionable visual field are displayed on the display screen at the same time for the auditor to observe; If the initial retrieved visual field is consistent with the questioned visual field, then execute step S3; if the initial retrieved visual field is obviously inconsistent with the questioned visual field, then execute step S4: Step S3: the reviewer can manually focus and adjust the image displacement of the preliminary recovered visual field, and further observe the preliminary recovered visual field to achieve the purpose of manual review; Step S4: The auditor uses the "index" function to perform indexing using the built-in template matching algorithm until the questionable field of view is matched.

2. The fully automatic visual field recovery method according to claim 1, characterized in that: In step S4, the template matching algorithm takes the preliminary recovered field of view as the center point, walks in an N*N rectangular area, and performs template matching on multiple images in the area and the question field of view, and displays the matching position in real time on the image scanned from the N*N rectangular area by drawing a red solid circle. The reviewer zooms in on the area, double-clicks the field of view where the red solid circle is located, moves to that position, and displays the field of view in real time as the preliminary recovered field of view.

3. The fully automatic visual field recovery method according to claim 2, characterized in that: The N can be an integer between 5 and 15.

4. The fully automatic visual field recovery method according to claim 1, characterized in that: The host computer can be a computer or an intelligent terminal with a touch screen, and the auditor can click the corresponding function keys and operation screens with a mouse, or touch the corresponding function keys and operation screens with his hands.

5. The fully automatic visual field recovery method according to claim 1, characterized in that: In the process of the lower computer finding the questionable field of view, the focusing and translation of the scanning device can be automatically controlled.

6. A computer device, characterized in that: The method comprises a processor and a memory, wherein the memory is configured to store a computer program; when the computer program is executed by the processor, the processor implements the method according to any one of claims 1 to 5.

7. A microscope review film review system, characterized in that: The method comprises the computer device as claimed in claim 6, and a microscopic scanner connected to the computer device.

Images