CN-120953381-B - Vein data processing method and system based on image analysis

CN120953381BCN 120953381 BCN120953381 BCN 120953381BCN-120953381-B

Abstract

The invention provides a vein data processing method and a vein data processing system based on image analysis, which are characterized in that a target limb image proportion conversion is carried out on a target region sequence generated by a user terminal based on a limb model through receiving the expected region sequence, the target region sequence is generated and sent to an operation terminal, a probe position is acquired according to acquisition equipment, guide information is generated based on the relative positions of the probe position and the target region sequence, the overlapping degree of the probe position and the target region sequence is determined in real time, a target image is acquired according to the overlapping degree, when the target image does not meet a puncture condition, the target image is processed according to a connectivity strategy to obtain a prediction amplitude and a prediction direction, a prediction region sequence is generated based on the prediction amplitude and the prediction direction, a recommended point is obtained based on the prediction region sequence, and the accurate positioning of a puncture point can be realized while the personalized puncture requirement of a patient is considered.

Inventors

LIU TING
WANG RAN
WANG LIN
LI YING
SHEN TONG

Assignees

首都医科大学宣武医院

Dates

Publication Date: 20260508
Application Date: 20250806

Claims (7)

1. A vein data processing method based on image analysis, comprising: The method comprises the steps of receiving a desired region sequence generated by a user side based on a limb model, performing scale conversion of a target limb image on the desired region sequence, generating a target region sequence, and sending the target region sequence to an operation side, wherein the method comprises the following steps: Responding to the customized information of the user side, and calling the limb model to be sent to the user side; Determining an expected center according to triggering information of the user side on any point position of a limb model, performing intermediate point matching positioning on an indwelling needle model based on the expected center, and generating an expected model according to angle confirmation information of the indwelling needle model; Generating extension lines to two sides by taking the expected model as a reference, generating expected areas according to the extension lines, and generating expected area sequences based on the numbering information of the user side on each expected area; Determining limb areas corresponding to the expected areas, and marking joint position points corresponding to the limb areas as first characteristic points and second characteristic points, wherein the limb areas comprise a big arm area and a small arm area; Acquiring a first distance between a middle point of the expansion line corresponding to the expected area and a first characteristic point and a second distance between the middle point and the second characteristic point, and acquiring a relative position proportion corresponding to each expansion line according to a ratio between the first distance and the second distance; Determining a corresponding target line in a target limb image based on the relative position proportion and the limb area, and generating a target area according to the target line; determining the number information corresponding to the target area according to the number information corresponding to the expected area, and generating a target area sequence based on the number information corresponding to the target area; Determining a first reference point corresponding to a first characteristic point and a second reference point corresponding to a second characteristic point in the target limb image; generating a characteristic line according to the first characteristic point and the second characteristic point, and acquiring a relative angle between the expansion line and the characteristic line; Determining a conversion point between a first reference point and a second reference point according to the relative position proportion, and generating a conversion line based on the first reference point and the second reference point; generating a target line based on the relative angle by taking the conversion point as an intermediate point, and generating a target area according to the corresponding target line; acquiring a probe position according to acquisition equipment, and generating guide information based on the relative position of the probe position and a target area sequence; determining the overlapping degree of the probe position and the target region sequence in real time, and acquiring a target image according to the overlapping degree; And when the target images do not meet the puncture condition, processing the target images according to a connectivity strategy to obtain a prediction amplitude and a prediction direction, generating a prediction region sequence based on the prediction amplitude and the prediction direction, and obtaining recommended points based on the prediction region sequence.
2. The method of claim 1, wherein determining in real time a degree of overlap of the probe position with a sequence of target regions, acquiring a target image based on the degree of overlap, comprises: acquiring equipment contours corresponding to detection equipment, determining a detection range according to the equipment contours, acquiring the overlapping area of the detection range and each target area in real time, and acquiring the overlapping degree based on the overlapping area; And when the overlapping degree is greater than or equal to a preset overlapping threshold value, controlling the detection equipment to acquire ultrasonic images corresponding to the corresponding target areas, and obtaining target images corresponding to the target areas.
3. The method of claim 1, wherein processing the target image according to a connectivity policy to obtain a predicted magnitude and a predicted direction when none of the target images satisfies a puncture condition, comprises: acquiring a first endpoint and a second endpoint of each blood vessel in the target image, generating a blood vessel line according to the corresponding first endpoint and second endpoint, and determining the blood vessel direction of the blood vessel line; Determining a target direction according to the direction of the expected model, acquiring an angle difference value between the target direction and the blood vessel direction, and marking a blood vessel line with the angle difference value smaller than a preset difference value as a target blood vessel; acquiring a width value of the target blood vessel, and determining a blood vessel segment corresponding to the width value continuously larger than a preset width as an analysis blood vessel; Acquiring the blood vessel length of each analysis blood vessel, and if no analysis blood vessel with the model length larger than the expected model exists, determining that the target image does not meet the puncture condition; and processing the target image according to a connectivity strategy to obtain a prediction amplitude and a prediction direction.
4. A method according to claim 3, wherein said processing the target image according to a connectivity policy to obtain a predicted magnitude and a predicted direction comprises: Marking an analysis blood vessel intersected with the edge of the target area as a predicted blood vessel, determining one end of the predicted blood vessel intersected with the edge of the target area as a first predicted point, and determining the other end as a second predicted point; determining a prediction direction by taking the second prediction point as a starting point and the first prediction point as an end point; And when judging that the predicted blood vessel meets the gradient change condition, acquiring a length difference value between the blood vessel length of the predicted blood vessel and the model length, and determining a predicted amplitude according to the length difference value.
5. The method of claim 4, wherein said determining that the predicted vessel satisfies a gradient change condition comprises: performing three-class splitting on the predicted blood vessel to sequentially obtain a first sub-blood vessel, a second sub-blood vessel and a third sub-blood vessel which are ordered according to the predicted direction; acquiring a first average width of the first sub-blood vessel, a second average width of the second sub-blood vessel and a third average width of the third sub-blood vessel; And if the first average width is smaller than or equal to the second average width and the second average width is smaller than or equal to the third average width, judging that the predicted blood vessel meets the gradient change condition.
6. The method of claim 4, wherein generating a predicted region sequence based on the predicted magnitude and the predicted direction, deriving a recommended point location based on the predicted region sequence, comprises: Expanding the target area according to the prediction amplitude and the prediction direction to obtain a prediction area; based on the predicted amplitude of each predicted region, ascending order is carried out on the predicted regions to obtain a predicted region sequence; Acquiring predicted images corresponding to each predicted region in sequence according to the predicted region sequence, and acquiring expanded blood vessels corresponding to predicted blood vessels in the predicted images; when the expanded blood vessel meets the puncture condition, determining the corresponding predicted blood vessel and the expanded blood vessel as recommended points.
7. The image analysis-based venous data processing system of any of the methods of claims 1-6, comprising: The generation module is used for receiving a desired region sequence generated by the user terminal based on the limb model, performing proportion conversion on a target limb image on the desired region sequence, generating a target region sequence and sending the target region sequence to the operation terminal; The acquisition module is used for acquiring the probe position according to the acquisition equipment and generating guide information based on the relative position of the probe position and the target area sequence; The determining module is used for determining the overlapping degree of the probe position and the target region sequence in real time and acquiring a target image according to the overlapping degree; And the prediction module is used for processing the target image according to a connectivity strategy to obtain a prediction amplitude and a prediction direction when the target image does not meet the puncture condition, generating a prediction region sequence based on the prediction amplitude and the prediction direction, and obtaining recommended points based on the prediction region sequence.

Description

Vein data processing method and system based on image analysis Technical Field The present invention relates to image processing technology, and in particular, to a vein data processing method and system based on image analysis. Background In the medical field, venipuncture is an important basic link of operations such as clinical treatment, blood collection and the like, and along with the development of medical technology, the requirement on the accuracy of venipuncture is increasingly improved. In the prior art, the determination of the puncture position point mainly depends on the experience of medical staff to determine the position point during puncture, a nurse judges the optimal puncture position subjectively by observing the surface characteristics of the skin of a patient, the elasticity and trend of a touch blood vessel and other modes, however, the patient group has great individual difference in the aspects of physical condition, blood vessel condition and the like, the indwelling needle puncture is taken as an example, some patients have thinner blood vessels and are buried under the skin deeply, some patients have arteriosclerosis and poor elasticity, and some patients have special conditions such as vascular malformation, under the condition, if the patients depend on the experience of the nurse only, the judgment error easily occurs, thus the puncture success rate can be reduced, and because the experience level and professional cognition of each nurse possibly have difference, the selection of the puncture point has stronger subjectivity and is difficult to adapt to the actual requirements of different patients accurately, so that the puncture operation effect and the treatment experience of the patients are influenced. Therefore, how to realize accurate positioning of puncture points while considering personalized puncture requirements of patients becomes a problem to be solved. Disclosure of Invention The embodiment of the invention provides a vein data processing method and a vein data processing system based on image analysis, which can realize accurate positioning of puncture points while considering personalized puncture requirements of patients. In a first aspect of the present invention, there is provided a vein data processing method based on image analysis, including: receiving an expected region sequence generated by a user side based on a limb model, performing scale conversion of a target limb image on the expected region sequence, generating a target region sequence, and sending the target region sequence to an operation side; acquiring a probe position according to acquisition equipment, and generating guide information based on the relative position of the probe position and a target area sequence; determining the overlapping degree of the probe position and the target region sequence in real time, and acquiring a target image according to the overlapping degree; And when the target images do not meet the puncture condition, processing the target images according to a connectivity strategy to obtain a prediction amplitude and a prediction direction, generating a prediction region sequence based on the prediction amplitude and the prediction direction, and obtaining recommended points based on the prediction region sequence. Optionally, in a possible implementation manner of the first aspect, the receiving user side generates the expected region sequence based on the limb model, including: Responding to the customized information of the user side, and calling the limb model to be sent to the user side; Determining an expected center according to triggering information of the user side on any point position of a limb model, performing intermediate point matching positioning on an indwelling needle model based on the expected center, and generating an expected model according to angle confirmation information of the indwelling needle model; Generating extension lines to two sides based on the expected model, generating expected areas according to the extension lines, and generating expected area sequences based on the number information of the user side on each expected area. Optionally, in one possible implementation manner of the first aspect, the performing scaling conversion on the target limb image on the desired region sequence, generating a target region sequence, and sending the target region sequence to an operation end includes: Determining limb areas corresponding to the expected areas, and marking joint position points corresponding to the limb areas as first characteristic points and second characteristic points, wherein the limb areas comprise a big arm area and a small arm area; Acquiring a first distance between a middle point of the expansion line corresponding to the expected area and a first characteristic point and a second distance between the middle point and the second characteristic point, and acquiring a relative position proportion corresponding to ea