TWI819936B - Vertebral canal abnormality determination system and determination device thereof - Google Patents

Abstract

A vertebral canal abnormality determination system includes an image capturing device and a vertebral canal abnormality determination device. The image capturing device is configured to capture a magnetic resonance imaging (MRI) of the vertebral canal. The vertebral canal abnormality determination device is configured to obtain a spinous process area, a pedicle area, a vertebral foramen area and an intervertebral disk area of the MRI, numericalize the spinous process area, the pedicle area, the vertebral foramen area and the intervertebral disk area of the MRI to obtain a plurality of parameter values of the areas, and determine whether the vertebral canal is normal or not.

Description

Spinal canal abnormality diagnosis system and its interpretation device

The present disclosure relates to a spinal canal abnormality diagnosis system and a spinal canal abnormality diagnosis device.

Current medical institutions rely on doctors to manually observe magnetic resonance imaging (MRI) to determine whether there are abnormalities in the magnetic resonance imaging. However, this type of artificial interpretation method is time-consuming and labor-intensive, and there are problems such as human misjudgment. Therefore, how to improve the aforementioned conventional problems is one of the goals of those in this technical field.

Therefore, the present disclosure proposes a spinal canal abnormality diagnosis system and a spinal canal abnormality diagnosis device, which can improve the aforementioned conventional problems.

An embodiment of the present disclosure provides a spinal canal abnormality diagnosis system. The spinal canal abnormality diagnosis system includes an image capturing device and a spinal canal abnormality judgment device. The image capturing device is used to capture a magnetic resonance image of a spinal canal. The spinal canal abnormality interpretation device is used to: obtain the spinous process area, a pedicle area, a vertebral foramen area and an intervertebral disc area in magnetic resonance imaging; to obtain the spinous process area, vertebral pedicle area, vertebral foramen area and intervertebral disc area. Magnetic resonance imaging Numericalize to obtain several parameter values in these areas; and, based on these parameter values, determine whether the spinal canal is abnormal.

Another embodiment of the present disclosure provides a device for diagnosing spinal canal abnormalities. The spinal canal abnormality diagnosis device includes a region acquisition unit, a digitization unit and an interpretation unit. The area acquisition unit is used to acquire a spinous process area, a pedicle area, a vertebral foramen area and an intervertebral disc area in a magnetic resonance image of the spinal canal. The numerical unit is used to digitize the magnetic resonance images of the spinous process area, pedicle area, vertebral foramen area and intervertebral disc area to obtain several parameter values of these areas. The interpretation unit is used to determine whether the spinal canal is abnormal based on these parameter values.

In order to have a better understanding of the above and other aspects of the present disclosure, embodiments are given below and described in detail with reference to the accompanying drawings:

100: Spinal Canal Abnormality Diagnosis System

110:Image capture device

120: Spinal canal abnormality diagnosis device

121: Area acquisition unit

122: Numerical unit

123: Interpretation unit

C:Spinal canal

FV: Feature vector

IN: Intervertebral disc area

M1: magnetic resonance imaging

PE: pedicle area

S1: abnormal signal

SP: spinous process area

S110~S150: steps

T2: axial

VE: vertebral foramen area

Figure 1 illustrates a functional block diagram of a spinal canal abnormality diagnosis system according to an embodiment of the present disclosure.

Figure 2 is a schematic diagram of a magnetic resonance image captured by the image capturing device in Figure 1 .

Figure 3 shows a schematic diagram of the spinous process area, pedicle area, vertebral foramen area and intervertebral disc area obtained from the magnetic resonance imaging in Figure 2.

Figure 4 illustrates a flow chart of the spinal canal abnormality diagnosis method of the spinal canal abnormality diagnosis system in Figure 1.

Please refer to Figures 1 to 3. Figure 1 illustrates a functional block diagram of the spinal canal abnormality diagnosis system 100 according to an embodiment of the present disclosure. Figure 2 illustrates the magnetic field captured by the image capture device 110 in Figure 1. A schematic diagram of the magnetic resonance image M1, and Figure 3 shows a schematic diagram of the spinous process area SP, pedicle area PE, vertebral foramen area VE and intervertebral disc area IN obtained from the magnetic resonance image M1 in Figure 2.

As shown in Figures 1 to 3, the spinal canal abnormality diagnosis system 100 includes an image capturing device 110 and a spinal canal abnormality diagnosis device 120. The image capturing device 110 is used to capture the magnetic resonance image M1 of the spinal canal (vertebral canal) C (as shown in Figure 2). The spinal canal abnormality diagnosis device 120 is used to: obtain the spinous process area SP, the pedicle area PE, the vertebral foramen (vertebral formulas) area VE, and the intervertebral disk (intervertebral disk) area IN of the magnetic resonance image M1 (As shown in Figure 3); Numericalize the magnetic resonance imaging of the spinous process area SP, pedicle area PE, vertebral foramen area VE and intervertebral disc area IN to obtain several parameter values of the above areas; and, Based on these parameter values, it is judged whether the spinal canal C is abnormal. In this way, the spinal canal abnormality diagnosis system 100 can automatically determine whether the spinal canal C is abnormal, thereby improving the problem of high misjudgment rate in manual interpretation.

In addition, there is no need to create an additional three-dimensional (3D) model of the spinal canal C. The spinal canal abnormality diagnosis system 100 can determine whether the spinal canal C is abnormal based on (or analyzes) the magnetic resonance image M1. Abnormalities in this article refer to diseases that may occur in the spinal canal, such as foraminal stenosis, disc herniation, etc.

As shown in FIG. 1 , the image capturing device 110 is, for example, a nuclear magnetic resonance device, which can capture a magnetic resonance image of a longitudinal section of the subject's spinal canal C (the subject is lying down). The aforementioned spinous process area SP, pedicle area PE, vertebral foramen area VE and intervertebral disc area IN belong to soft tissue. The magnetic resonance image M1 is, for example, an axial T2 image, which can show soft tissue It achieves better imaging quality and increases the accuracy of abnormal diagnosis. In addition, the aforementioned longitudinal section may be a section of any vertebral segment of the spinal canal C.

As shown in FIG. 1 , the spinal canal abnormality diagnosis device 120 includes a region acquisition unit 121 , a digitization unit 122 , and a judgment unit 123 . The area acquisition unit 121, the numerical unit 122 and the interpretation unit 123 are, for example, physical circuits formed using a semiconductor process, but the embodiment of the present invention is not limited to this. At least two of the area acquisition unit 121, the digitization unit 122, and the interpretation unit 123 can be integrated into a single unit; or, at least one of the area acquisition unit 121, the digitization unit 122, and the interpretation unit 123 can be integrated into a controller ( controller) or a processor.

As shown in Figures 1 and 3, the area acquisition unit 121 can acquire the spinous process area SP, the pedicle area PE, the vertebral foramen area VE, and the intervertebral disc area IN of the magnetic resonance image M1 of the spinal canal C. For example, the area obtaining unit 121 may use, for example, machine learning technology to obtain (or determine) the spinous process area SP, the pedicle area PE, the vertebral foramen area VE, and the intervertebral disc area IN. In one embodiment, after obtaining these regions, the region obtaining unit 121 can mark these regions in the magnetic resonance image M1 (as shown in FIG. 3 ) and display them on a display (not shown). For medical personnel to observe. In addition, the spinous process area SP, pedicle area PE, vertebral foramen area VE and intervertebral disc area IN in Figure 3 can be represented in different shapes (eg, linear shapes) respectively. The digitizing unit 122 can digitize the magnetic resonance images of the spinous process area SP, the pedicle area PE, the vertebral foramen area VE, and the intervertebral disc area IN to obtain several parameter values of the above areas. The interpretation unit 123 is used to determine whether the spinal canal C is abnormal based on these parameter values. The aforementioned machine learning technologies are, for example, Convolutional Neural Network (CNN), Generative Adversarial Network, GAN) or other suitable machine learning methods, but the embodiment of the present invention is not limited to this.

The spinal canal abnormality diagnosis device 120 is further used to: when the spinal canal C is abnormal, send an abnormality signal S1 to alert the subject or examiner of the occurrence of the abnormality. For example, the interpretation unit 123 is used to: when the spinal canal C is abnormal, send the abnormal signal S1. An indication unit (not shown) is electrically connected to the interpretation unit 123 and is used to issue an indication signal according to the abnormal signal S1. In one embodiment, the indication unit is, for example, a light emitter, and the indication signal is, for example, light; or, the indication unit is, for example, a vibrator, and the indication signal is, for example, vibration; or, the indication unit is, for example, a display, and the indication signal is, for example, text or graphics; Alternatively, the indicating unit is, for example, a speaker, and the indicating signal is, for example, a sound. The indication unit may be configured on the image capture device 110 or on the spinal canal abnormality diagnosis device 120, but the embodiment of the present invention does not limit the location and type of the indication unit.

The aforementioned parameter values include the solidity of at least one of the spinous process area SP, the pedicle area PE, the vertebral foramen area VE, and the intervertebral disc area IN, the standard deviation of the grayscale value in the area, the area, the true roundness, and the ratio of the major and minor axes of the approximate ellipse. , at least two of the approximate ellipse area and actual area ratio and the approximate ellipse perimeter. In one embodiment, the aforementioned parameter values may further include a ratio of the two areas of the spinous process area SP, the pedicle area PE, the vertebral foramen area VE, and the intervertebral disc area IN, and the standard deviation of the gray scale values in the two areas. One ratio, one ratio of the ratio of the major and minor axes of the two approximate ellipses, and/or one ratio of the ratio of the two approximate ellipse areas to the actual area.

In one embodiment, the aforementioned parameter values may further include at least one of the true roundness of each of the vertebral foramen area VE and the intervertebral disc area IN, the ratio of the major and minor axes of the approximate ellipse, the ratio of the area of the approximate ellipse to the actual area, and the circumference of the approximate ellipse. In one embodiment, the aforementioned parameter values include Including at least one of the solidity of each of the spinous process area SP, the pedicle area PE, the vertebral foramen area VE, and the intervertebral disc area IN, the standard deviation of the grayscale value in the area, and the area.

The following explains the definition of "parameter value".

"Solidity" is, for example, the ratio of the actual area of a region to the minimum coverage area (recesses of the region are not considered). "Standard deviation of grayscale values within a region" is, for example, the standard deviation distribution of grayscale values of all pixels in the region. "Area" is, for example, the area of a region. "True roundness" is, for example, the roundness of the outer contour of the region. The "major and minor axis ratio of the approximate ellipse" is, for example, the ratio of the major axis and the minor axis of an ellipse that approximates the area (can completely cover the area). The "ratio of the approximate ellipse area to the actual area" is, for example, the ratio of the area of an ellipse (which can completely cover the area) of the approximate area to the actual area of the area. The "approximate ellipse perimeter" is, for example, the circumferential length of an ellipse that approximates the area (can completely cover the area). The "ratio of two areas" is, for example, the ratio of the "areas" of two regions. The "ratio of the standard deviation of gray-scale values in two regions" is, for example, the ratio of the "standard deviation of gray-scale values in two regions". "The ratio of the major and minor axes of two approximate ellipses" is, for example, the ratio of the "major and minor axes ratio of the two approximate ellipses" of the two regions. "The ratio of the major and minor axes of two approximate ellipses" is, for example, "the ratio of the area of two approximate ellipses to the actual area" of the two regions. For example, it is the ratio of the "ratio of the approximate ellipse area to the actual area" of the two regions. The aforementioned "parameter value" is, for example, the numerical unit 122 using image analysis technology to obtain geometric information of the analysis area (for example, area edges, etc.), and obtain parameter values based on this.

In one embodiment, the interpretation unit 123 may include a classifier (not shown), which, for example, uses techniques such as random forest or regression analysis to obtain a feature vector FV of these parameter values. The interpretation unit 123 determines the corresponding abnormality type of the spinal canal C according to the feature vector FV. For example, when the area of the vertebral foramen area VE of the fourth and fifth vertebrae and When the "ratio of two areas" of the area of the spinous process area SP is greater than 0.194023, it means that the magnetic resonance image M1 shows vertebral foramen stenosis. For another example, if the aforementioned "ratio of two areas" is less than 0.194023 and the "area" of the vertebral foramen area VE is greater than 343 and the "standard deviation of grayscale values within the area" of the vertebral foramen area VE is greater than 41.858385, it also represents magnetic resonance imaging M1 Shows the condition of vertebral foraminal stenosis. However, depending on the diagnostic needs, the spinal canal abnormality diagnosis system 100 can use the parameter value of at least one of all the aforementioned parameters to generate the corresponding feature vector, and determine the status of the corresponding spinal canal C accordingly. The judgment result is, for example, an abnormality category (vertebral foramen stenosis, disc herniation, etc.) or a health category (no abnormality).

Please refer to Figure 4, which illustrates a flow chart of the spinal canal abnormality diagnosis method of the spinal canal abnormality diagnosis system in Figure 1.

In step S110, the image capturing device 110 captures the magnetic resonance image M1 of the spinal canal C.

In step S120 , the region acquisition unit 121 acquires the spinous process region SP, the pedicle region PE, the vertebral foramen region VE, and the intervertebral disc region IN of the magnetic resonance image M1 of the spinal canal C.

In step S130 , the numerical unit 122 digitizes the magnetic resonance images of the spinous process area SP, pedicle area PE, vertebral foramen area VE, and intervertebral disc area IN to obtain several parameter values of the above areas.

In step S140, the judgment unit 123 judges whether the spinal canal C is abnormal based on several parameter values. If yes, the process proceeds to step S150, the interpretation unit 123 can send an abnormality signal S1 to indicate that an abnormality is found in the magnetic resonance imaging M; if not, the interpretation unit 123 can send a normal signal (not shown) to indicate that an abnormality is found in the magnetic resonance imaging M. No abnormalities were found on image M.

In summary, the embodiments of the present disclosure propose a spinal canal abnormality diagnosis system, its spinal canal abnormality judgment device and a spinal canal abnormality judgment method using the same, which use a parametric method to convert images into single-stored values, which is helpful for subsequent diagnosis. Numerical evaluation of auxiliary and postoperative follow-up will continue to optimize the performance of auxiliary diagnosis, and continue to develop the function of predicting postoperative recovery, providing doctors with better accurate judgment and precise treatment tools.

In summary, although the present disclosure has been disclosed in the above embodiments, they are not used to limit the present disclosure. Those with ordinary knowledge in the technical field to which this disclosure belongs can make various modifications and modifications without departing from the spirit and scope of this disclosure. Therefore, the protection scope of the present disclosure shall be subject to the scope of the appended patent application.

100: Spinal Canal Abnormality Diagnosis System

110:Image capture device

120: Spinal canal abnormality diagnosis device

121: Area acquisition unit

122: Numerical unit

123: Interpretation unit

FV: Feature vector

M1: magnetic resonance imaging

S1: abnormal signal

Claims (12)

A spinal canal abnormality interpretation system, including: an image acquisition device used to acquire a magnetic resonance image of the spinal canal; and a spinal canal abnormality interpretation device used to: acquire a spinous process of the magnetic resonance image area, a pedicle area, a vertebral foramen area and an intervertebral disc area; digitize the magnetic resonance images of the spinous process area, the pedicle area, the vertebral foramen area and the intervertebral disc area to obtain these A plurality of parameter values in the region; and based on these parameter values, determine whether the spinal canal is abnormal. The spinal canal abnormality diagnosis system of claim 1, wherein the spinal canal abnormality judgment device is further used to: send an abnormality signal when the spinal canal is abnormal. The spinal canal abnormality diagnosis system according to claim 1, wherein the parameter values include a solidity of at least one of the spinous process area, the pedicle area, the vertebral foramen area and the intervertebral disc area, a region The standard deviation of the grayscale value, an area, a true roundness, a ratio of the major and minor axes of the approximate ellipse, at least two of the ratio of the approximate ellipse area to the actual area, and an approximate ellipse circumference. The spinal canal abnormality diagnosis system as described in claim 3, wherein the parameter values include a ratio of two areas of the spinous process area, the pedicle area, the vertebral foramen area and the intervertebral disc area, and two One ratio of the standard deviation of the gray scale values in the area, one ratio of the major and minor axis ratios of the approximate ellipse, and two ratios of the area of the approximate ellipse to the actual area. The spinal canal abnormality diagnosis system as described in claim 1, wherein the parameter values include a true roundness of each of the vertebral foramen area and the intervertebral disc area, an approximate ellipse major and minor axis ratio, an approximate ellipse area and the actual area Ratio to at least one of the approximate ellipse circumferences. The spinal canal abnormality diagnosis system according to claim 1, wherein the parameter values include the solidity of each of the spinous process area, the pedicle area, the vertebral foramen area and the intervertebral disc area, the grayness in a region At least one of the order standard deviation and an area. A spinal canal abnormality diagnosis device includes: a region acquisition unit for acquiring a spinous process region, a pedicle region, a vertebral foramen region and an intervertebral disc region in a magnetic resonance image of the spinal canal; a digitizer A unit for digitizing the magnetic resonance imaging of the spinous process area, the pedicle area, the vertebral foramen area and the intervertebral disc area to obtain a plurality of parameter values of these areas; and an interpretation unit for Based on these parameter values, it is determined whether the spinal canal is abnormal. The spinal canal abnormality diagnosis device of claim 7, wherein the judgment unit is further used to: send an abnormality signal when the spinal canal is abnormal. The spinal canal abnormality diagnosis device according to claim 7, wherein the parameter values include a solidity of at least one of the spinous process area, the pedicle area, the vertebral foramen area and the intervertebral disc area, a region The standard deviation of the grayscale value, an area, a true roundness, a ratio of the major and minor axes of the approximate ellipse, at least two of the ratio of the approximate ellipse area to the actual area, and an approximate ellipse circumference. The device for diagnosing spinal canal abnormalities as described in claim 9, wherein the parameter values include a ratio of two areas of the spinous process area, the pedicle area, the vertebral foramen area and the intervertebral disc area, and two One ratio of the standard deviation of the gray scale values in the area, one ratio of the major and minor axis ratios of the approximate ellipse, and two ratios of the area of the approximate ellipse to the actual area. The device for diagnosing spinal canal abnormalities as described in claim 7, wherein the parameter values include the true roundness of each of the vertebral foramen area and the intervertebral disc area, an approximate ellipse major and minor axis ratio, an approximate ellipse area and the actual area Ratio to at least one of the approximate ellipse circumferences. The device for diagnosing spinal canal abnormalities as described in claim 7, wherein the parameter values include the solidity of each of the spinous process area, the pedicle area, the vertebral foramen area and the intervertebral disc area, the grayness in a region At least one of the order standard deviation and an area.

Images