CN-122004915-A - Mammary gland focus identification and positioning method

CN122004915ACN 122004915 ACN122004915 ACN 122004915ACN-122004915-A

Abstract

The invention discloses a breast focus identification positioning method, which relates to the technical field of medical image processing, and comprises the steps of carrying out breast focus image identification detection on a target patient through a molybdenum target technology, counting main focus position characteristics with fuzzy boundaries, marking the distribution position of fat type breast tissues as a central position if the target patient is detected as a dominant region of the fat type breast tissues, obtaining auxiliary tissue central position characteristics, marking the main focus position characteristics and the auxiliary tissue central position characteristics, obtaining a plurality of marked lines, setting the same included angle formed between the marked lines, presetting the position gradient for breast focus identification continuous detection on the target patient according to the distance value of the marked lines and the angle value formed between the main focus position characteristics and the nipple, and carrying out breast focus continuous detection on the target patient through a high-frequency acoustic wave imaging technology to obtain focus calibration positioning results. The invention can improve the focus recognition and positioning accuracy.

Inventors

Zhou Huanhao
WANG WENWU
LI PING

Assignees

衢州市人民医院(衢州中心医院)

Dates

Publication Date: 20260512
Application Date: 20260204

Claims (7)

1. The breast focus identifying and positioning method is characterized by comprising the following steps: Step S1, performing breast focus image identification detection on a target patient through a molybdenum target technology, counting breast focus position data with fuzzy boundaries to obtain main detection focus position characteristics, and if the target patient is detected to be a dominant region of fat type breast tissue, marking the distribution position of the fat type breast tissue as a central position to obtain auxiliary detection tissue central position characteristics; S2, marking the position features of the main measuring focus and the central position features of the auxiliary measuring tissue to obtain a plurality of marks, setting the included angles formed between the marks to be the same, and presetting the position gradient for performing breast focus identification and continuous detection on a target patient according to the distance value of the marks and the angle value formed between the position features of the main measuring focus and the nipple, and performing breast focus continuous detection on the target patient by a high-frequency acoustic wave imaging technology to obtain focus correction and positioning results; And S3, if the target patient is detected as the gland type breast tissue dominant region, marking the distribution positions of the fat type breast tissue and the gland type breast tissue to obtain a double distribution position region characteristic, and according to the double distribution position region characteristic and the main detection focus position characteristic, presetting a follow-up gradient set for carrying out breast focus identification follow-up on the target patient, and carrying out breast focus follow-up on the target patient through a high-frequency acoustic wave imaging technology to obtain a focus correction positioning result.
2. The method for breast lesion recognition and localization according to claim 1, wherein step S1 comprises: performing breast focus image identification detection on a target patient by a molybdenum target technology to obtain a primary detection focus image, and performing three-dimensional space position positioning on a breast focus in the primary detection focus image by adopting a multi-slice two-dimensional image stacking method according to the primary detection focus image to obtain primary detection positioning data; extracting breast lesions with fuzzy boundaries in the primary detection lesion images to obtain lesions to be detected, correspondingly counting the position data of the lesions to be detected from the primary detection positioning data according to the lesions to be detected, and obtaining the position characteristics of the primary detection lesions; and performing type detection judgment on the fat type breast or the gland type breast on the target patient, and outputting a judgment case I if the target patient is detected to be the fat type breast tissue dominant region.
3. The method of claim 2, wherein step S1 further comprises: marking the distribution position of the fat type mammary tissue according to the first judgment condition to obtain a single distribution position area characteristic; presetting a first threshold value of a to-be-measured ratio according to the main measured focus position characteristic, and extracting a single-distribution to-be-measured position characteristic from the single-distribution position region characteristic according to the first threshold value of the to-be-measured ratio; And (3) marking the central position of the single-distribution position feature to be measured to obtain the central position feature of the auxiliary tissue to be measured.
4. A breast lesion recognition and localization method according to claim 3, wherein step S2 comprises: The center position feature of the auxiliary measurement tissue and the focus position feature of the main measurement are connected in a straight line mode, so that a first marking line is obtained, and the distance value of the first marking line is counted, so that distance data I is obtained; And (3) connecting two straight lines between the main measured focus position features and edges to which the single-distribution measured position features belong to obtain a second marking and a third marking, and counting distance values of the second marking and the third marking to obtain second distance data and third distance data, wherein the third marking comprises a first marking, a second marking and a third marking.
5. The method of claim 4, wherein step S2 further comprises: counting the linear distance between the main focus position feature and the nipple to obtain parameter distance data, and counting the angle value formed between the main focus position feature and the nipple to obtain a parameter angle value; And comparing the distance data I, the distance data II and the distance data III with the reference distance data in sequence to obtain a reference decision factor I, a reference decision factor II and a reference decision factor III, and multiplying the reference decision factor I, the reference decision factor II and the reference decision factor III with the reference angle value in sequence to obtain a follow-up gradient I, a follow-up gradient II and a follow-up gradient III.
6. The method of claim 5, wherein step S2 further comprises: Performing breast focus follow-up on the target patient through high-frequency acoustic imaging according to the first marking and the follow-up gradient to obtain a first breast focus recognition result, performing breast focus follow-up on the target patient through high-frequency acoustic imaging according to the second marking and the follow-up gradient to obtain a second breast focus recognition result, and performing breast focus follow-up on the target patient through a high-frequency acoustic imaging technology according to the third marking and the follow-up gradient to obtain a third breast focus recognition result; and carrying out integration comparison on the first breast focus identification result, the second breast focus identification result and the third breast focus identification result to obtain focus calibration positioning results.
7. The method of claim 6, wherein step S3 comprises: if the target patient is detected to be the gland type breast tissue dominant region, outputting a judging condition II; Marking distribution positions of fat type breast tissues and gland type breast tissues according to the second judgment condition to obtain a first double-distribution position area feature and a second double-distribution position area feature, wherein the first double-distribution position area feature comprises a first double-distribution position area feature and a second double-distribution position area feature; Presetting a to-be-detected proportion threshold value II, extracting a double-distribution to-be-detected position feature I from a double-distribution position region feature I according to the to-be-detected proportion threshold value I, extracting a double-distribution to-be-detected position feature II from the double-distribution position region feature II according to the to-be-detected proportion threshold value II, and marking three follow-up gradient for carrying out breast focus follow-up on a target patient according to the double-distribution to-be-detected position feature I, the follow-up gradient II and the follow-up gradient III to obtain a first class of follow-up gradient set; marking three follow-up gradient for follow-up breast focus detection of the target patient according to the two-distribution position feature to be detected, the first follow-up gradient, the second follow-up gradient and the third follow-up gradient to obtain a second type of follow-up gradient set; Performing breast focus continuous detection on a target patient through a high-frequency acoustic wave imaging technology according to the first type continuous detection gradient set and the second type continuous detection gradient set to obtain a first type breast focus identification result set and a second type breast focus identification result set; and carrying out integration comparison on the first type breast focus identification result set and the second type breast focus identification result set to obtain focus calibration positioning results.

Description

Mammary gland focus identification and positioning method Technical Field The application belongs to the technical field of medical image processing, and particularly relates to a breast focus identification and positioning method. Background The existing dynamic evaluation of breast lesions depends on CT/MRI, and has the problems of high cost, long period and the like. Molybdenum targets (two-dimensional images) are sensitive to structural abnormalities such as microcalcifications, but are difficult to accurately position under the influence of pressing deformation, and color Doppler ultrasound (three-dimensional imaging) can identify morphological and blood flow characteristics, but has poor measurement consistency due to operator manipulation differences. The two positioning systems are not synchronous, namely, the molybdenum target is divided into areas in quadrants, the color Doppler ultrasound adopts a dial method to record the position, and the same focus is difficult to correspond to space under two modes due to deformation and activity of mammary tissue, and the focus is easy to miss diagnosis or repeat marking due to subjective comparison depending on experience of doctors. According to the invention, by fusing the structural positioning information of the molybdenum target with the ultrasonic tissue attribute detection, a unified three-dimensional coordinate reference is established, so that the accurate tracking and the volume change quantification of the trans-modal focus are realized, the human error and the background coincidence positioning error are obviously reduced, and the diagnosis consistency is improved. Disclosure of Invention In order to overcome the defects in the prior art, the application provides a breast focus identification and positioning method. The application provides a breast focus identification and positioning method, which comprises the following steps: Step S1, performing breast focus image identification detection on a target patient through a molybdenum target technology, counting breast focus position data with fuzzy boundaries to obtain main detection focus position characteristics, and if the target patient is detected to be a dominant region of fat type breast tissue, marking the distribution position of the fat type breast tissue as a central position to obtain auxiliary detection tissue central position characteristics; S2, marking the position features of the main measuring focus and the central position features of the auxiliary measuring tissue to obtain a plurality of marks, setting the included angles formed between the marks to be the same, and presetting the position gradient for performing breast focus identification and continuous detection on a target patient according to the distance value of the marks and the angle value formed between the position features of the main measuring focus and the nipple, and performing breast focus continuous detection on the target patient by a high-frequency acoustic wave imaging technology to obtain focus correction and positioning results; And S3, if the target patient is detected as the gland type breast tissue dominant region, marking the distribution positions of the fat type breast tissue and the gland type breast tissue to obtain a double distribution position region characteristic, and according to the double distribution position region characteristic and the main detection focus position characteristic, presetting a follow-up gradient set for carrying out breast focus identification follow-up on the target patient, and carrying out breast focus follow-up on the target patient through a high-frequency acoustic wave imaging technology to obtain a focus correction positioning result. Preferably, performing breast focus image identification detection on a target patient by a molybdenum target technology to obtain a primary detection focus image, and performing three-dimensional space position positioning on a breast focus in the primary detection focus image by adopting a multi-slice two-dimensional image stacking method according to the primary detection focus image to obtain primary detection positioning data; extracting breast lesions with fuzzy boundaries in the primary detection lesion images to obtain lesions to be detected, correspondingly counting the position data of the lesions to be detected from the primary detection positioning data according to the lesions to be detected, and obtaining the position characteristics of the primary detection lesions; And performing type detection judgment on the fat type breast or the gland type breast of the target patient, and outputting judgment case one if the target patient belongs to one type of fat type breast tissue or gland type breast tissue. Preferably, according to the first judgment condition, the distribution position of the fat type mammary tissue is marked to obtain a single distribution position area characteristic; presetting a first threshold value