TWI681752B - Method for identifying and locating nerves - Google Patents

Abstract

The present invention relates a method for identifying and locating at least one nerve in a determined operation area. In the method, an image detector is utilized to scan the determined operation area according to a nerve direction forecast from a nerve identifying program, capture multiple 2D cross-sectional images of the determined operation area in different parts, and determine whether the 2D cross-sectional images belong to a target nerve by the nerve identifying program. If the 2D cross-sectional images are identified as the target nerve by the nerve identifying program, the nerve identifying program constructs a 3D nerve image from the 2D cross-sectional images by 3D image reconstruction.

Description

Neural identification and positioning method

The invention relates to a nerve identification and positioning method, in particular to a nerve identification and positioning method for reconstructing a three-dimensional nerve image.

In the surgical operation, it is necessary to confirm the direction of various nerves to avoid accidentally damaging or compressing the nerve. Taking an orthopedic trauma surgery as an example, when performing minimally invasive surgery, a bone plate needs to be implanted into a limb through all mouths In order to fix the broken bone, however, the nerve must be avoided during the implantation of the bone plate to prevent the bone plate from compressing the nerve, resulting in postoperative pain or paralysis.

Because the nerve direction of each person is different, in order to avoid nerves and blood vessels during surgery, doctors can only predict the nerve direction based on anatomical knowledge and surgical experience. Therefore, there is a risk of bone plate compression of the nerve in each operation.

The main purpose of the present invention is to confirm the nerve direction of the position to be operated in advance before the operation to avoid damage to the nerve during the operation (such as the nerve being compressed by the bone plate).

A nerve identification and positioning method of the present invention includes providing an image display device with a nerve identification program and a nerve identification module; providing a detection device that detects It is provided with an inspection machine and an image detector, and the image detector is arranged on the inspection machine; and a scan identification procedure is performed. The scan identification procedure includes: a first step, scanning a pending operation with the image detector Parts to confirm an initial position of a nerve to be located; in a second step, the image detector captures a plurality of two-dimensional cross-sectional images of the initial position of the nerve to be located, the nerve identification program uses the nerve identification mode The group recognizes the two-dimensional cross-sectional images and provides the detection machine with a predicted nerve direction; in a third step, the detection machine drives the image detector to move to an area of the site to be operated according to the predicted nerve direction, and Acquiring a plurality of two-dimensional cross-sectional images of the area; and a fourth step, the nerve identification program identifies whether the two-dimensional cross-sectional images of the area are the nerve to be located by the nerve identification module, when the nerve is identified When the program recognizes that the two-dimensional cross-sectional images are the nerve to be located, the detection machine then drives the image detector to scan the nerve to be located in the next area of the site to be operated according to the predicted nerve direction, and extracts the area A plurality of two-dimensional cross-sectional images and the two-dimensional cross-sectional images to identify whether the area is the nerve to be located, if the two-dimensional cross-sectional images are the nerve to be located, repeat the fourth step until the The image detector scans to the last area.

In the present invention, the two-dimensional cross-sectional images of the nerve to be located are acquired by the image display device, the detection device, and the scanning identification program, and according to the predicted nerve direction, and the two-dimensional images are reconstructed by a three-dimensional image reconstruction technology. The three-dimensional cross-sectional image is constructed as a three-dimensional image, so that the doctor can confirm the nerve position before the operation, so as to reduce the risk of bone plate compression of the nerve.

100‧‧‧Neural identification and positioning method

110‧‧‧Provide an image display device

110a‧‧‧Image display device

110b‧‧‧Monitor

110c‧‧‧computer

120‧‧‧Provide a detection device

120a‧‧‧Detection device

120b‧‧‧ testing machine

120c‧‧‧Image detector

120d‧‧‧The first track piece

120e‧‧‧The second rail

120f‧‧‧Rotating plate

120g‧‧‧vehicle

120h‧‧‧Dock

120i‧‧‧accommodating space

120j‧‧‧Perforation

120k‧‧‧ transmission rod

120m‧‧‧screw

120n‧‧‧ First motor

120p‧‧‧Second motor

120q‧‧‧third motor

120r‧‧‧Encoder

130‧‧‧ Perform a scan identification process

131‧‧‧ First step

132‧‧‧The second step

133‧‧‧The third step

134‧‧‧The fourth step

135‧‧‧ fifth step

136‧‧‧Step 6

200‧‧‧ site to be operated

200a‧‧‧Region

210‧‧‧ nerve to be located

210a‧‧‧Initial position

220‧‧‧Bone

P‧‧‧Two-dimensional cross-sectional image

P1‧‧‧3D image

P2‧‧‧ Image of the area to be operated

Figure 1: Schematic diagram of a radial nerve in a limb wound around a bone of an upper limb.

Figure 2: Flow chart of the nerve identification and positioning method of the present invention.

Figure 3: A schematic diagram of the image display device of the present invention.

Figure 4: A perspective view of the detection device of the present invention.

Figure 5: Another perspective view of the detection device of the present invention.

Figure 6: Flow chart of the scan identification process of the present invention.

Figure 7: A schematic diagram of a two-dimensional cross-sectional image of the present invention.

Figure 8: A schematic diagram of the three-dimensional image of the present invention.

Figure 9: A schematic diagram of an image reconstructed from a three-dimensional neural image of the present invention to a site to be operated on.

Before the orthopedic surgeon performs minimally invasive surgery, it is necessary to pay attention to the following main nerves, such as the radial nerve (radial nerve, its cross-sectional area is about 4mm), the median nerve (median nerve, its cross-sectional area is about 4mm), and the axillary nerve (axillary nerve, its Cross-sectional area is about 3.6mm), ulnar nerve (ulnar nerve, its cross-sectional area is about 3mm), etc., in this embodiment, the identification and positioning of a radial nerve is used as an example, but not limited to this, please refer to Figure 1 in a limb A radial nerve is wound around a bone of the upper limb. When a radius or a bone of the upper limb is broken, a bone plate (not shown) must be used to fix the broken bone. However, due to the radial nerve of each person It is assumed that the bone positions of the upper limbs are not the same, which increases the risk of surgery.

In order to reduce the risk of surgery, please refer to FIG. 2, the present invention adopts a neural identification and positioning method 100, which includes: "providing an image display device" 110, "providing a detection device" 120, and "performing a Scan identification process" 130.

Please refer to FIG. 3, in the "providing an image display device" 110, an image display device 110a includes a display 110b and a neural recognition program and a neural recognition module in a computer 110c, the neural recognition module It is an identification module composed of at least one nerve section.

Please refer to Figures 1, 3, 4 and 5, in the "provide a detection device" 120, a detection device 120a has a detection machine 120b and an image detector 120c, the image detector 120c is provided in the detection machine The table 120b, in this embodiment, the inspection machine 120b has a first rail member 120d, a second rail member 120e, a rotating plate 120f, a carrier 120g, a base 120h and an accommodating space 120i, The accommodating space 120i is used to accommodate a site to be operated 200. In this embodiment, the site to be operated 200 is an arm, but not limited to this, the first rail member 120d is disposed on the base 120h, the The second rail member 120e is movably coupled to the first rail member 120d, and the second rail member 120e surrounds the accommodating space 120i, the rotating plate 120f is movably coupled to the second rail member 120e, and the carrier 120g is coupled On the rotating plate 120f and swinging with the rotating plate 120f, the carrier 120g has a perforation 120j and a transmission rod 120k, the transmission rod 120k is movably penetrated through the perforation 120j, and the image detector 120c is coupled to the transmission rod 120k, and the image detector 120c is selectively directed toward or away from the site to be operated 200 by the transmission rod 120k, the image detector 120c is electrically connected to the computer 110c to transmit image signals, the image detector 120c is selected from Yu Yi ultrasound detector.

Please refer to FIGS. 4 and 5. Preferably, the testing machine 120b further includes a screw 120m, a first motor 120n, a second motor 120p, a third motor 120q and at least one encoder 120r. The motor 120n is used to drive the screw 120m to rotate, the second rail member 120e is engaged with the screw 120m, and the first motor 120n is used to drive the screw 120m to rotate to drive the second rail member engaged with the screw 120m 120e moves laterally along the first rail member 120d, the second motor 120p is used to drive the rotating plate 120f to swing, so that the image detector 120c can move around the site to be operated 200, and the second motor 120p drives the rotation The way in which the plate 120f swings is selected from gear and rack transmission, transmission wheel and belt transmission, etc. The third motor 120q is used to drive the transmission rod 120k, so that the image detector 120c can be selectively moved toward or away from the standby In the surgical site 200, the manner in which the third motor 120q drives the transmission rod 120k to move is selected from Used in gear and rack drive, drive wheel and belt drive, etc.

Please refer to FIGS. 4 and 5, the encoder 120r is used to record the transmission data (such as the number of forward or reverse rotations, etc.) of the first motor 120n, the second motor 120p and the third motor 120q, the encoder 120r may be integrated into the first motor 120n, the second motor 120p, and the third motor 120q.

Please refer to FIGS. 1, 4, 5 and 6, in the scan identification procedure 130, first, the operation site 200 is placed in the accommodating space 120i, and then, a first step 131 is performed on the first In step 131, the image detector 120c is used to scan the site to be operated 200 to confirm an initial position 210a of a nerve 210 to be located. Please refer to FIGS. 1, 3 and 4 to contact the site to be operated with the image detector 120c 200, and a cross-sectional image of the nerve 210 to be located in the site 200 to be operated is displayed on the display 110b. In this embodiment, the nerve 210 to be located is used as the description of the radial nerve, but not limited to this And set an image capture frequency parameter and a plurality of scanning area parameters for the image detector 120c, wherein the image capture frequency parameter is the number of two-dimensional cross-sectional images captured by the image detector 120c in a time, for example Capture 300 two-dimensional cross-sectional images in 1 second. The scanning area parameters are the number of times that the inspection machine 120b drives the image detector 120c to move in one direction on the surface of the site to be operated 200, please refer to page 3. As shown in the figure, preferably, the image detector 120c transmits a scanned image of the site to be operated 200 to the display 110b for confirmation.

Please refer to FIGS. 1, 4, 5 and 6, after the first step 131, a second step 132 is performed. In the second step 132, according to the image capture frequency parameter, the image detector 120c captures A plurality of two-dimensional cross-sectional images P (see FIG. 7) of the initial position 210a of the nerve 210 to be located, the two-dimensional cross-sectional images P include a section of the nerve 210 to be located, in this embodiment, the The image detector 120c surrounds the to-be-operated site 200 and contacts the to-be-operated site 200 by the second rail member 120e to capture the two-dimensional cross-sectional images P, and the image detector 120c combines the two-dimensional cross-sectional images P sent to the movie Stored in the computer 110c like the display device 110a, the neural recognition program recognizes the two-dimensional cross-sectional images P by the neural recognition module, and provides the detection machine 120b with a prediction of the nerve direction, in addition, the two In the case of a multi-dimensional cross-sectional image P, the encoder 120r transmits the transmission data of the first motor 120n, the second motor 120p, and the third motor 120q to the computer 110c for conversion by an arithmetic program into each of the two-dimensional cross-sections Each shooting coordinate of the image P, and record each shooting coordinate corresponding to the two-dimensional cross-sectional image P. In this embodiment, each shooting coordinate is the image detector 120c through the first track member 120d, A displacement coordinate when the second rail member 120e and the transmission rod 120k move to a position.

Please refer to Figures 4, 5, 6, and 7, after the second step 132, a third step 133 is performed. In the third step 133, based on the predicted nerve direction and the scan area parameters, the detection machine 120b drives the image detector 120c to move to an area 200a of the site to be operated 200. In this embodiment, the first motor 120n drives the screw 120m to rotate to drive the second rail member 120e and the rotating plate 120f, the transmission rod 120k and the image detector 120c move to the area 200a. Similarly, the image detector 120c surrounds the site to be operated 200 by the second track member 120e and contacts the site to be operated 200, and according to The image capture frequency parameter captures a plurality of two-dimensional cross-sectional images P of the area 200a (see FIG. 7), the encoder 120r combines the first motor 120n, the second motor 120p, and the third motor 120q The transmission data of is transmitted to the computer 110c, and the shooting coordinates of the two-dimensional cross-sectional image P in the area 200a are recorded. In this embodiment, the shooting coordinates are the image detector 120c through the first track A displacement coordinate of the member 120d, the second rail member 120e, the rotating plate 120f and the transmission rod 120k when moving to the area 200a.

Please refer to FIG. 6, after the third step 133, a fourth step 134 is performed. In the fourth step 134, the neural recognition program recognizes the image in the third step 133 by the neural recognition module The two-dimensional cross-sectional images P of the detector 120c moving to the site to be operated 200 and the area 200a are Whether it is the nerve 210 to be located, when the nerve recognition program recognizes that the two-dimensional cross-sectional images are the nerve 210 to be located, preferably, the computer 110c sends out a prompt signal, which may be selected from the optical signal , Sound signal, and then, the detection machine 120b drives the image detector 120c to scan the nerve 210 to be located in the next area 200a of the site 200 to be operated according to the predicted nerve direction, and extracts a plurality of areas 200a The two-dimensional cross-sectional image P and the shooting coordinates of the two-dimensional cross-sectional image P recorded in the area 200a, and identify whether the two-dimensional cross-sectional images P of the area 200a are the nerve 210 to be located, if the When the two-dimensional cross-sectional image P is the nerve 210 to be located, the fourth step 134 is repeated until the image detector 120c scans to the last area 200a. Conversely, in the fourth step 134, when the neural recognition program recognizes that the two-dimensional cross-sectional images P are not the nerve 210 to be located, the neural recognition program provides another prediction nerve direction of the detection machine 120b, and then The third step 133 and the fourth step 134 are performed until the image detector 120c scans to the last area 200a.

Please refer to FIGS. 7 and 8, the scan identification process 130 further includes a fifth step 135, after the fourth step 134, the fifth step 135 is performed, and in the fifth step 135, a three-dimensional image reconstruction technique is used The two-dimensional cross-sectional images P captured in the third step 133 and the fourth step 134 are constructed into a three-dimensional image P1. The three-dimensional image P1 includes a three-dimensional neural image. Preferably, the three-dimensional image P1 A warning area is included, which is located outside the three-dimensional neural image.

Please refer to FIG. 8 and FIG. 9, the scan recognition program 130 further includes a sixth step 136, after the fifth step 135, the sixth step 136 is performed, and in the sixth step 136, the three-dimensional neural image is reconstructed to In an image P2 of the site to be operated 200 (see FIG. 9), the image P2 of the site to be operated is an image of a broken bone 220 of the site 200 to be referred to during the operation by referring to the three-dimensional image , To avoid damage to the nerves during the operation (such as nerve compression by the bone plate).

The present invention captures the two-dimensional cross-sectional images P of the nerve 210 to be located, and uses the The three-dimensional image reconstruction technology constructs the two-dimensional cross-sectional images P into the three-dimensional image P1, so that the doctor can confirm the nerve position before the operation to effectively reduce the risk of nerve damage.

The scope of protection of the present invention shall be subject to the scope defined in the attached patent application. Any changes and modifications made by those who are familiar with this skill without departing from the spirit and scope of the present invention shall fall within the scope of protection of the present invention. .

100‧‧‧Neural identification and positioning method

110‧‧‧Provide an image display device

120‧‧‧Provide a detection device

130‧‧‧ Perform a scan identification process

Claims (18)

A nerve identification and positioning method includes: providing an image display device with a nerve identification program and a nerve identification module; providing a detection device with a detection machine and an image detector, the The image detector is installed on the inspection machine; and a scan identification procedure is performed, the scan identification procedure includes: a first step, the image detector is used to scan an operation site to confirm an initial position of a nerve to be located; In a second step, the image detector captures a plurality of two-dimensional cross-sectional images of the initial position of the nerve to be located, the nerve recognition program recognizes the two-dimensional cross-sectional images by the nerve recognition module, and provides the detection A machine predicts the nerve direction; in a third step, the detection machine drives the image detector to move to an area of the site to be operated according to the predicted nerve direction, and acquires a plurality of two-dimensional cross-sectional images of the area; and In a fourth step, the nerve recognition program recognizes whether the two-dimensional cross-sectional images of the area are the nerve to be located by the nerve recognition module, and when the nerve recognition program recognizes the two-dimensional cross-sectional images to be the nerve to be located At this time, the detection machine then drives the image detector to scan the nerve to be located in the next area of the site to be operated according to the predicted nerve direction, and acquires a plurality of two-dimensional cross-sectional images of the area and identifies the area. Whether the two-dimensional cross-sectional images are the nerve to be located, if the two-dimensional cross-sectional images are the nerve to be located, the fourth step is repeated until the image detector scans to the last area. The nerve identification localization method as described in item 1 of the patent application scope, wherein the image detection When the device contacts the site to be operated, a cross-sectional image of the nerve in the site to be operated is displayed on a display. The nerve identification localization method as described in item 1 of the patent application scope, wherein in the fourth step, when the nerve identification program identifies that the two-dimensional cross-sectional images are not the nerve to be located, the nerve identification program provides the detection again The machine predicts the nerve direction, and then performs the third step and the fourth step until the image detector scans to the last area. The nerve identification localization method as described in item 1 of the patent application scope, wherein when the nerve identification program identifies the two-dimensional cross-sectional images as the nerve to be located, it sends out a prompt signal, which can be selected from the optical signal , Sound signal. The nerve identification and positioning method as described in item 1 of the patent application scope, wherein the scan identification process further includes a fifth step, which uses the three-dimensional image reconstruction technology to extract the third and fourth steps The two-dimensional cross-sectional image is constructed as a three-dimensional image, and the three-dimensional image includes a three-dimensional neural image. The nerve identification and positioning method as described in item 5 of the patent scope, wherein the three-dimensional image includes a warning area, and the warning area is located outside the three-dimensional nerve image. The nerve identification and positioning method as described in item 1 of the patent application scope, wherein the scan identification process further includes a sixth step, which reconstructs the three-dimensional image formed in the fifth step to a pending operation of the site to be operated Part image. The nerve identification localization method as described in item 1 of the patent application scope, wherein the two-dimensional cross-sectional images include a cross-section of the nerve to be localized. The nerve identification and positioning method as described in item 1 of the patent application scope, wherein when capturing each of the two-dimensional cross-sectional images, a shooting coordinate of each of the two-dimensional cross-sectional images is simultaneously recorded. The nerve identification and positioning method as described in item 9 of the patent application scope, wherein the detection machine has a first rail member, a second rail member, a rotating plate and a carrier, the second rail member is movably combined with The first rail member, the rotating plate is movably coupled to the second rail member, the carrier is coupled to the rotating plate and swings with the rotating plate, the carrier has a perforation and a transmission rod, the transmission rod is movable The image detector is combined with the transmission rod, and the shooting coordinates are a displacement coordinate of the image detector moving to a position by the second rail member, the rotating plate, and the transmission rod. The nerve identification and positioning method as described in item 10 of the patent application scope, wherein the testing machine has a base and an accommodating space, the accommodating space is for accommodating the site to be operated, and the first rail member is disposed at The base, the second track member surrounds the accommodating space, and the image detector is selectively directed toward or away from the site to be operated by the transmission rod. The nerve identification and positioning method as described in item 11 of the patent application scope, wherein the testing machine has a screw and a first motor, the first motor is used to drive the screw to rotate, and the second track member is engaged with the screw , The first motor drives the screw to rotate to drive the second rail member to move laterally along the first rail member. The nerve identification and positioning method as described in item 12 of the patent application scope, wherein the detection machine has a second motor, the second motor is used to drive the rotating plate to swing, so that the image detector can be around the operation site mobile. The nerve identification and positioning method as described in item 13 of the patent application scope, wherein the detection machine has a third motor, the third motor is used to drive the transmission rod, so that the image detector can selectively move toward or away from the The area to be operated. The nerve identification localization method as described in item 14 of the patent application scope, wherein the detection machine has at least one encoder, the encoder is used to record the first motor, the second motor and the third motor The transmission data of is converted into the shooting coordinates by an arithmetic program. The nerve identification localization method as described in item 1 of the patent application scope, wherein in the first step, an image acquisition frequency parameter and a plurality of scanning area parameters are set for the image detector. The nerve identification localization method as described in item 16 of the patent application scope, wherein in the second step, the image detector captures the two-dimensional of the initial position of the nerve to be positioned according to the image capture frequency parameter Cross-sectional image. The nerve identification and positioning method as described in item 16 of the patent application scope, wherein in the third step, the detection machine drives the image detector to move to the site to be operated according to the predicted nerve direction and the scan area parameters In this area, the image detector captures the two-dimensional cross-sectional images of the area according to the image capture frequency parameter.

Images