Algorithm for automatically identifying distal radius fracture palmar inclination angle ruler deflection angle radius height

Abstract

The invention discloses an algorithm for automatically identifying distal radius fracture palmar inclination angle and ruler deflection angle radius height, and relates to the technical field of medical image processing, comprising the following steps: step 1, image preprocessing, step 2, key structure positioning, step 3, palmar inclination angle calculation, step 4, ruler deflection angle calculation, step 5, radius height calculation, and step 6, early warning and verification. According to the method, the skeleton contour can be reliably reserved through image preprocessing, the skeleton edge is highlighted, and all key points of the radius can be reliably positioned through key structure positioning; the palm inclination angle, the ulna deflection angle and the radius height of the radius are intelligently calculated through palm inclination angle calculation, ulna deflection angle calculation and radius height calculation, and the method is reasonable in design and suitable for application and popularization.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to an algorithm for automatically identifying the palmar inclination angle, ulnar deviation angle, and radius height of a distal radius fracture. Background Art

[0002] Distal radius fracture is a common fracture type in the upper limb, accounting for about 17.5% of all fractures. Since the distal radius is the junction of cancellous bone and compact bone, the bone cortex is thin and the mechanical structure is relatively weak, so it is prone to fracture. Distal radius fracture involves multiple key angles and height parameters, including palmar inclination, ulnar deviation, and radial height. These parameters are crucial for assessing the degree of fracture displacement, formulating reduction plans, and predicting prognosis. Traditional measurement methods mainly rely on manual measurement on X-rays, which is cumbersome and easily affected by subjective factors. Manual measurement is not only time-consuming and labor-intensive, but also difficult to ensure the accuracy and consistency of the measurement.

[0003] In summary, an algorithm was designed to automatically identify the palmar inclination angle, ulnar deviation angle and radial height of distal radius fractures. Summary of the invention

[0004] In order to overcome the above-mentioned shortcomings, the present invention provides an algorithm for automatically identifying the palmar inclination angle, ulnar deviation angle and radius height of distal radius fracture.

[0005] The present invention achieves the above-mentioned purpose through the following technical solutions: An algorithm for automatically identifying palmar inclination angle, ulnar deviation angle, and radial height of distal radius fractures comprises the following steps: Step 1: Image preprocessing, used to preprocess standard wrist anteroposterior and lateral X-ray films; Step 2: Key structure positioning, used to automatically detect the key points of the distal radial articular surface, styloid process tip, and ulna sigmoid notch; Step 3: palmar inclination angle calculation, used to calculate the palmar inclination angle of the radius; Step 4: Calculation of ulnar deviation angle, used to calculate the ulnar deviation angle of the radius; Step 5: Radius height calculation, used to calculate the height of the radius; Step 6: Warning and verification; Step 7: Output the results.

[0006] Preferably, the step 1 comprises: S11, image enhancement processing; S12, scale calibration, converting the pixel spacing in the X-ray into millimeter units.

[0007] Preferably, the step S11 comprises: S111, read the image and convert it into grayscale; S112, Gaussian filtering denoising, used to smooth the image to reduce noise while preserving the bone contour; S113, CLAHE contrast enhancement, is used to adaptively enhance local contrast and highlight bone edge details.

[0008] Preferably, the step 2 comprises: S21, edge detection, extracting bone contours through the Canny operator; S22, detecting the long axis of the radius and the tangent line of the articular surface, and extracting the long axis of the radius and the tangent line of the articular surface through Hough line transformation; S23, key point positioning, Tip of radial styloid process: the extreme point of the lateralmost contour of the radius in the AP radiograph; Ulnar angle of lunate fossa: the intersection of the ulnar side of the articular surface and the radial axis on the AP view; Palmar / dorsal articular surface points: the highest and lowest points of the articular surface on the lateral view.

[0009] Preferably, the step three comprises: S31, lateral radiograph processing, locate the palmar point (A) and dorsal point (B) of the articular surface, draw the long axis (L) along the middle of the radial shaft, and draw a vertical line (V) through the midpoint of the articular surface; S32, palm inclination angle calculation, connect points AB to form the joint surface line (J), calculate the angle between J and V: , where negative values ​​of θ represent dorsal tilt and positive values ​​represent palm tilt.

[0010] Preferably, the step 4 comprises: S41, anteroposterior radiographs were processed to locate the midpoint of the sigmoid notch (E) and the tip of the radial styloid process (F), and the long axis (L) was drawn along the radial shaft, and a vertical line (V) was drawn; S42, Calculate the scale deflection angle. Connect the EF points to form a scale deflection line (R). Calculate the angle between R and V: .

[0011] Preferably, the step five comprises: S51, positive radiograph, locating the tip of the styloid process (F) and the ulnar angle of the lunate fossa (G); S52. For vertical distance measurement, draw two parallel lines perpendicular to the long axis of the radius (L), passing through points F and G respectively, calculate the distance between the two lines and convert it into millimeters: .

[0012] Preferably, the step six comprises: S61, palm tilt angle warning, palm tilt angle >25° or <-20°, triggering "angle abnormality" warning; S62, radial height warning, radial height <5mm, marked as "severe shortening"; S63, manual review for doctors to confirm the accuracy of the measurement.

[0013] Preferably, in the Gaussian filtering denoising, the calculation formula is: , is the standard deviation, which controls the "width" of the Gaussian distribution. The larger σ is, the stronger the blurring effect is. is the coordinate of the pixel relative to the center of the convolution kernel.

[0014] Preferably, a weighted average formula is used in the Gaussian filtering denoising. Noise usually manifests itself as isolated pixel value mutations. Through weighted average, abnormal pixel values ​​are diluted by the "normal values" of surrounding pixels to achieve the effect of smoothing noise. Since the Gaussian kernel weight decreases from the center to the outside, the weight difference of pixels on both sides of the edge is small when averaged, so the edge will not be completely blurred.

[0015] The beneficial effects of the present invention are as follows: in the algorithm for automatically identifying the palmar inclination angle, ulnar deviation angle and radius height of distal radius fracture: Image preprocessing can reliably preserve the bone contour and highlight the bone edge, and combined with key structure positioning, it can reliably locate the key points of the radius; The palm inclination angle, ulnar deviation angle and radius height are calculated intelligently to calculate the palm inclination angle, ulnar deviation angle and radius height. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will now be described by way of example with reference to the accompanying drawings, in which: Figure 1 is a step diagram of the present invention; Figure 2 It is a step diagram of image preprocessing of the present invention; Figure 3 It is a step diagram of the image enhancement processing of the present invention; Figure 4 It is a step diagram of the key structure positioning of the present invention; Figure 5 It is a step diagram of palm tilt angle calculation of the present invention; Figure 6 It is a step diagram of the scale deflection angle calculation of the present invention; Figure 7 is a step diagram of radius height calculation of the present invention; Figure 8 It is a step diagram of early warning and verification of the present invention. DETAILED DESCRIPTION

[0017] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0018] like Figure 1-Figure 8 As shown, an algorithm for automatically identifying the palmar inclination angle, ulnar deviation angle, and radial height of distal radius fractures includes the following steps: Step 1: Image preprocessing, used to preprocess standard wrist anteroposterior and lateral X-ray films; Step 2: Key structure positioning, used to automatically detect the key points of the distal radial articular surface, styloid process tip, and ulna sigmoid notch; Step 3: palmar inclination angle calculation, used to calculate the palmar inclination angle of the radius; Step 4: Calculation of ulnar deviation angle, used to calculate the ulnar deviation angle of the radius; Step 5: Radius height calculation, used to calculate the height of the radius; Step 6: Early warning and verification; Step 7: Output the results.

[0019] Specifically, the step 1 includes: S11, image enhancement processing; S12, scale calibration, converting the pixel spacing in the X-ray into millimeter units.

[0020] Specifically, the step S11 includes: S111, read the image and convert it into grayscale; S112, Gaussian filtering denoising, used to smooth the image to reduce noise while preserving the bone contour; S113, CLAHE contrast enhancement, is used to adaptively enhance local contrast and highlight bone edge details.

[0021] Specifically, the step 2 includes: S21, edge detection, extracting bone contours through the Canny operator; S22, detecting the long axis of the radius and the tangent line of the articular surface, and extracting the long axis of the radius and the tangent line of the articular surface through Hough line transformation; S23, key point positioning, Tip of radial styloid process: the extreme point of the lateralmost contour of the radius in the AP radiograph; Ulnar angle of lunate fossa: the intersection of the ulnar side of the articular surface and the radial axis on the AP view; Palmar / dorsal articular surface points: the highest and lowest points of the articular surface on the lateral view.

[0022] Specifically, the step three includes: S31, lateral radiograph processing, locate the palmar point (A) and dorsal point (B) of the articular surface, draw the long axis (L) along the middle of the radial shaft, and draw a vertical line (V) through the midpoint of the articular surface; S32, palm inclination angle calculation, connect points AB to form the joint surface line (J), calculate the angle between J and V: , where negative values ​​of θ represent dorsal tilt and positive values ​​represent palm tilt.

[0023] Specifically, the step 4 includes: S41, anteroposterior radiographs were processed to locate the midpoint of the sigmoid notch (E) and the tip of the radial styloid process (F), and the long axis (L) was drawn along the radial shaft, and a vertical line (V) was drawn; S42, Calculate the scale deflection angle. Connect the EF points to form a scale deflection line (R). Calculate the angle between R and V: .

[0024] Specifically, the step five includes: S51, positive radiograph, locating the tip of the styloid process (F) and the ulnar angle of the lunate fossa (G); S52. For vertical distance measurement, draw two parallel lines perpendicular to the long axis of the radius (L), passing through points F and G respectively, calculate the distance between the two lines and convert it into millimeters: .

[0025] Specifically, the step six includes: S61, palm tilt angle warning, palm tilt angle >25° or <-20°, triggering "angle abnormality" warning; S62, radial height warning, radial height <5mm, marked as "severe shortening"; S63, manual review for doctors to confirm the accuracy of the measurement.

[0026] Specifically, in the Gaussian filter denoising, the calculation formula is: , is the standard deviation, which controls the "width" of the Gaussian distribution. The larger σ is, the stronger the blurring effect is. is the coordinate of the pixel relative to the center of the convolution kernel.

[0027] Specifically, the weighted average formula is used in the Gaussian filtering denoising. Noise usually appears as isolated pixel value mutations. Through weighted average, abnormal pixel values ​​are diluted by the "normal values" of surrounding pixels to achieve the effect of smoothing noise. Since the Gaussian kernel weight decreases from the center to the outside, the weight difference of pixels on both sides of the edge is small when averaged, so the edge will not be completely blurred.

[0028] The implementation steps of Gaussian filtering are as follows: Generate a Gaussian convolution kernel and calculate a matrix of size (2k+1)×(2k+1) according to the σ value. Usually k=3σ. If σ=1, the kernel size is generally 7×7.

[0029] Convolution operation: Align the center of the Gaussian kernel to each pixel of the image; Calculate the weighted average of each neighborhood pixel in the kernel and replace the original pixel value; Weighted average formula: ,Gaussian filtering suppresses noise through weighted averaging, which can strike a balance between denoising and preserving anatomical structures, ,laying the foundation for subsequent enhancement and contour extraction.

[0030] The above is based on the present invention as an inspiration. Through the above description, relevant staff can make various changes and modifications without departing from the technical idea of ​​this invention. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. An algorithm for automatically identifying palmar inclination, ulnar deviation and radial height of distal radius fractures, characterized by: The following steps are involved: Step 1: Image preprocessing, used to preprocess standard wrist anteroposterior and lateral X-ray films; Step 2: Key structure positioning, used to automatically detect the key points of the distal radial articular surface, styloid process tip, and ulna sigmoid notch; Step 3: palmar inclination angle calculation, used to calculate the palmar inclination angle of the radius; Step 4: Calculation of ulnar deviation angle, used to calculate the ulnar deviation angle of the radius; Step 5: Radius height calculation, used to calculate the height of the radius; Step 6: Early warning and verification; Step 7: Output the results.

2. The algorithm for automatically identifying the palmar inclination angle, ulnar deviation angle, and radius height of distal radius fracture according to claim 1, characterized in that: The step one comprises: S11, image enhancement processing; S12, scale calibration, converting the pixel spacing in the X-ray into millimeter units.

3. The algorithm for automatically identifying palmar inclination angle, ulnar deviation angle, and radius height of distal radius fracture according to claim 2, characterized in that: The step S11 includes: S111, read the image and convert it into grayscale; S112, Gaussian filtering denoising, used to smooth the image to reduce noise while preserving the bone contour; S113, CLAHE contrast enhancement, is used to adaptively enhance local contrast and highlight bone edge details.

4. The algorithm for automatically identifying palmar inclination angle, ulnar deviation angle, and radius height of distal radius fracture according to claim 1, characterized in that: The second step comprises: S21, edge detection, extracting bone contours through the Canny operator; S22, detecting the long axis of the radius and the tangent line of the articular surface, and extracting the long axis of the radius and the tangent line of the articular surface through Hough line transformation; S23, key point positioning, Tip of radial styloid process: the extreme point of the lateralmost contour of the radius in the AP radiograph; Ulnar angle of lunate fossa: the intersection of the ulnar side of the articular surface and the radial axis on the AP view; Palmar / dorsal articular surface points: the highest and lowest points of the articular surface on the lateral view.

5. The algorithm for automatically identifying palmar inclination angle, ulnar deviation angle, and radius height of distal radius fracture according to claim 1, characterized in that: The step three comprises: S31, lateral radiograph processing, locate the palmar point (A) and dorsal point (B) of the articular surface, draw the long axis (L) along the middle of the radial shaft, and draw a vertical line (V) through the midpoint of the articular surface; S32, palm inclination angle calculation, connect points AB to form the joint surface line (J), calculate the angle between J and V: , where negative values ​​of θ represent dorsal tilt and positive values ​​represent palm tilt.

6. The algorithm for automatically identifying palmar inclination angle, ulnar deviation angle, and radius height of distal radius fracture according to claim 1, characterized in that: The fourth step comprises: S41, anteroposterior radiographs were processed to locate the midpoint of the sigmoid notch (E) and the tip of the radial styloid process (F), and the long axis (L) was drawn along the radial shaft, and a vertical line (V) was drawn; S42, Calculate the scale deflection angle. Connect the EF points to form a scale deflection line (R). Calculate the angle between R and V: 。 7. The algorithm for automatically identifying palmar inclination angle, ulnar deviation angle, and radius height of distal radius fracture according to claim 1, characterized in that: The step five comprises: S51, positive radiograph, locating the tip of the styloid process (F) and the ulnar angle of the lunate fossa (G); S52. For vertical distance measurement, draw two parallel lines perpendicular to the long axis of the radius (L), passing through points F and G respectively, calculate the distance between the two lines and convert it into millimeters: 。 8. The algorithm for automatically identifying palmar inclination angle, ulnar deviation angle, and radius height of distal radius fracture according to claim 1, characterized in that: The step six comprises: S61, palm tilt angle warning, palm tilt angle >25° or <-20°, triggering "angle abnormality" warning; S62, radial height warning, radial height <5mm, marked as "severe shortening"; S63, manual review for doctors to confirm the accuracy of the measurement.

9. The algorithm for automatically identifying palmar inclination angle, ulnar deviation angle, and radius height of distal radius fracture according to claim 3, characterized in that: In the Gaussian filter denoising, the calculation formula is: , is the standard deviation, which controls the "width" of the Gaussian distribution. The larger σ is, the stronger the blurring effect is. is the coordinate of the pixel relative to the center of the convolution kernel.

10. The algorithm for automatically identifying palmar inclination angle, ulnar deviation angle, and radius height of distal radius fracture according to claim 9, characterized in that: The Gaussian filter denoising adopts a weighted average formula.

Images