CN114748162A

CN114748162A - Path planning method and readable storage medium

Info

Publication number: CN114748162A
Application number: CN202210332086.2A
Authority: CN
Inventors: 张维; 王伟伟; 付晓璇
Original assignee: Beijing Yone Galaxy Technology Co ltd; Beijing Yinhe Fangyuan Technology Co ltd
Current assignee: Beijing Yinhe Fangyuan Technology Co ltd
Priority date: 2022-03-30
Filing date: 2022-03-30
Publication date: 2022-07-15

Abstract

The embodiment of the invention discloses a path planning method and a readable storage medium, belonging to the field of medical image data processing. The path planning method comprises the following steps: step S1, acquiring a to-be-evaluated track set according to the target point set in the target point region and the implantation point set in the skull implantation point region; step S2, carrying out track filtering on the track set to be evaluated to obtain a filtering track set; and step S3, carrying out risk assessment on the filtering track set to obtain an optimized path set. The path planning method and the readable storage medium provided by the invention can at least partially realize the path planning of the skull electrode implantation by utilizing a computer-aided doctor, so that the time for the path planning process of the skull electrode is reduced to a minute level, and the manual planning of the electrode implantation time of the doctor is greatly saved.

Description

Path planning method and readable storage medium

Technical Field

The present invention relates to the field of medical image data processing, and in particular, to a path planning method and a readable storage medium.

Background

Epilepsy is a brain function network disorder disease caused by repeated abnormal synchronous discharge of brain neurons. In current therapy, about 2/3 patients can effectively control seizures by reasonably taking antiepileptic drugs, but 1/3 patients still have poor control of medication.

The epilepsy area is effectively positioned by adopting a stereotactic electroencephalogram (SEEG) technology through preoperative comprehensive evaluation of a part of patients, and the epilepsy area is expected to be cured through operations and other modes.

Stereotactic electroencephalography (SEEG) is a technique for locating epileptic regions by implanting deep electrodes in the intracranial of an epileptic patient to record the discharge in the brain of the epileptic patient. Generally, 7-12 electrodes are implanted in the intracranial of one patient on average, and the manual planning of the electrode path by a doctor takes about 2-3 hours on average, so that the intracranial electrode implantation is time-consuming, and the working efficiency of the doctor is greatly reduced.

Therefore, it is necessary to provide a new path planning method and a readable storage medium.

Disclosure of Invention

In order to solve at least one aspect of the above problems and disadvantages in the prior art, the present invention provides a path planning method and a readable storage medium, which can at least partially implement path planning for implanting a cranial electrode by using a computer-aided physician, so that the time for path planning process of the cranial electrode is reduced to a minute level, thereby greatly saving the time for manually planning electrode implantation by the physician. The technical scheme is as follows:

According to an aspect of the present invention, there is provided a path planning method including the steps of:

step S1, acquiring a trajectory set to be evaluated according to the target point set in the target point area and the implantation point set in the skull implantation point area;

step S2, carrying out track filtration on the track set to be evaluated to obtain a filtration track set;

and step S3, carrying out risk assessment on the filtering track set to obtain an optimized path set.

According to another aspect of the present invention, there is provided a storable medium having stored thereon a program or instructions that when executed by a processor performs the path planning method described above.

The path planning method and the readable storage medium according to the embodiments of the present invention have at least one of the following advantages:

(1) the path planning method and the readable storage medium provided by the invention can complete the path planning of skull electrode implantation by utilizing a computer-aided doctor, so that the time consumption of the path planning process of the skull electrode is reduced to a minute level, and the manual planning of the electrode implantation time of the doctor is greatly saved;

(2) in the path planning of intracranial electrode implantation, the path planning method and the readable storage medium provided by the invention exclude the track intersecting with the intracranial danger area from the planned path;

(3) According to the path planning method and the readable storage medium, in the path planning process, through target filtering, multiple kinds of track filtering and track grading planning, the risk of the planned path is effectively reduced, and therefore the reliability of the path planning scheme is greatly improved;

(4) according to the path planning method and the readable storage medium, in the path planning process, a plurality of constraint conditions are added to the selection of the electrode entry point of the temporal lobe part, so that a safer optimized path is obtained;

(5) in the path planning process, the multi-electrode dynamic path planning method obtains a safe path planning scheme which not only obtains the optimal path planning scheme of a single target point from the perspective of the single target point, but also avoids the path conflict among a plurality of target points.

Drawings

These and/or other aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the preferred embodiments, taken in conjunction with the accompanying drawings of which:

FIG. 1 is a flow diagram of a path planning method according to one embodiment of the invention;

FIG. 2 is a flow diagram of the trajectory filtering shown in FIG. 1;

FIG. 3 is a flow chart of risk assessment shown in FIG. 1.

Detailed Description

The technical scheme of the invention is further specifically described by the following embodiments and the accompanying drawings. In the specification, the same or similar reference numerals denote the same or similar components. The following description of the embodiments of the present invention with reference to the accompanying drawings is intended to explain the general inventive concept of the present invention and should not be construed as limiting the invention.

Stereotactic electroencephalography (SEEG) is a technique for locating an epileptic attack area by implanting deep electrodes in the brain of an epileptic patient to record the discharge condition in the brain of the epileptic patient. Generally, 7-12 electrodes are implanted in the cranium of a patient on average, and a doctor needs to manually plan the implantation path of the electrodes. Because the cranium contains a plurality of key structures (such as complicated blood vessels, ventricles and the like) which need to be avoided when the electrode is implanted, the safety requirement on the planning of the electrode path is very high, and meanwhile, because the path is planned manually, a plurality of paths often conflict with each other, so that the re-planning is needed, the intracranial electrode implantation is very time-consuming, the time is usually 2-3 hours, and the working efficiency of a doctor is seriously reduced. To solve at least one of the above problems, we have devised a fully automatic path planning scheme. The method comprises the following specific steps:

Referring to fig. 1, a path planning method according to an embodiment of the invention is shown. The path planning method comprises the following steps:

step S1, acquiring a to-be-evaluated track set according to the target point set in the target point region and the implantation point set in the skull implantation point region;

step S2, carrying out track filtering on the track set to be evaluated to obtain a filtering track set;

In one example, data preprocessing is required prior to determining the target region and/or the skull implant point region. The method specifically comprises the following steps:

providing a Computed Tomography (CT) image;

providing a Magnetic Resonance Imaging (MRI) image;

providing a CT angiography (CTA) image; and/or

Providing a Magnetic Resonance Angiography (MRA) image; and/or

Providing a Magnetic Resonance angiography (MRV) image; and/or

Obtaining a mask of at least one preset structure in a preset structure set; and/or

Obtaining a temporal lobe mask; and/or

Obtaining a skull mask; and/or

Obtaining a mask of a middle plane between the left brain and the right brain; and/or

Obtaining a blood vessel image subjected to cerebral vessel segmentation in a preset structure set; and/or

Obtaining blood oxygen level dependent functional magnetic resonance imaging (BOLD fMRI) data;

respectively registering the CT image, the MRI image, the CTA image, the MRA image and/or the MRV image into a template space of magnetic resonance T1 weighted imaging (T1WI) to obtain a registered medical image, wherein the registered medical image comprises a registered CT image, a registered T1WI image, a registered CTA image and/or a registered MRA image and/or a registered MRV image;

temporal lobe data was registered into the template space of T1 weighted imaging (T1 WI).

And (3) segmenting the cerebrovascular image through the registered CTA image and/or the registered MRA image and/or the registered MRV image to obtain the vascular image.

In one example, one or any combination of temporal lobe, vessel, gray matter, ventricle, cerebellum, and brain functional network are included in the preset set of structures. The at least one pre-configured mask includes at least one of a gray matter mask, a ventricle mask, a cerebellum mask, and a brain function network boundary mask.

In one example, the segmentation of the cerebrovascular image is performed by performing vessel segmentation based on a whole brain mask (e.g., a brainstem mask, a ventricle mask, a cerebellar mask, etc.).

In one example, the mask of the preset structure is obtained by extracting the preset structure from the registered medical image, and can also be generated by T1WI image reconstruction.

In one example, a mask of the middle plane between the left and right brain is obtained by extracting the middle plane from the registered cine images, and may also be generated by T1WI image reconstruction.

In one example, the skull mask of an individual is registered to a medical image (e.g., an MRI image of the individual) for a skull template.

In one example, the temporal lobe skull mask is obtained by extracting the anatomy of the temporal lobe of the skull from the registered medical image, and can also be generated by T1WI image reconstruction.

In one example, the mask of the temporal lobe is obtained by extracting the anatomical structure of the temporal lobe from the registered medical image, and may also be generated by T1WI image reconstruction.

In one example, the CT image may be obtained directly by electron computed tomography, and the T1WI image may also be obtained by generative confrontation network (GAN) transformation.

In one example, the brain functional network boundary mask is obtained from functional magnetic resonance imaging (BOLD fMRI) data.

In one example, the target region is presented on a medical image after image pre-processing. Preferably, it is shown on the T1WI image after registration and image pre-processing.

In one example, step S1 further includes:

screening the target points according to the relationship between the second distance between all the target points in the target point set in the target point region and all the preset structures in the preset structure set and the threshold distance range of the target points to obtain a screened target point set;

and constructing a track set to be evaluated between all the screened target points in the screened target point set and all the implantation points in the implantation point set respectively.

In one example, risk assessment may be performed on all target points in the target point region first by setting the threshold distance range of the target points, so as to delete the target points with higher risk and keep the target points with lower risk. Therefore, the calculation amount in the subsequent path planning track is reduced, and the speed and the efficiency of path planning are improved. Deleting the target point when a second distance (a first distance will be described in detail below) between the target point and any preset structure in the preset structure set falls within a threshold distance range of the target point; and when the second distances between the target point and all the preset structures in the preset structure set are positioned outside the threshold distance range of the target point, reserving the target point and adding the target point into the screened target point set. And iterating the step until all the targets in the target point set are screened, obtaining all screened target points, and adding all the screened target points into the screened target point set.

For example, the target point threshold distance range is set to 5mm or less, and preferably, the target point threshold distance range may be set to 3mm or less. And when the second distance falls in the range of less than or equal to 5mm (preferably less than or equal to 3mm), determining the target point as a high-risk target point, and deleting the target point. When the second distance is within a range greater than 5mm (preferably greater than 3mm) (i.e., not within a range of less than or equal to 5mm, preferably less than or equal to 3mm), then the target is identified as a low risk target, retained and added to the post-screening set of targets.

The skilled person can also set that when the second distance between the target point and any one of the preset structures in the preset structure set falls within the threshold distance range of the target point, the target point is determined as a target point with lower risk, the target point is retained, and the target point is added into the screened target point set; and when the second distances between the target point and all the preset structures in the preset structure set are located outside the threshold distance range of the target point, determining the target point as the target point with higher risk, and deleting the target point. And iterating the step until all the targets in the target point set are screened, obtaining all screened target points, and adding all the screened target points into the screened target point set.

For example, the target point threshold distance range is set to be greater than 2mm, and preferably, the target point threshold distance range may be set to be greater than 3 mm. When the second distance falls within a range of more than 2mm, preferably more than 3mm, the target is identified as a low risk target, retained and added to the set of screened targets. When the second distance is within a range of 2mm or less (preferably 3mm or less) (i.e., not within a range of more than 2mm, preferably more than 3mm), the target point is determined to be a target point with higher risk, and the target point is deleted.

In one example, when the user adds a target area or adds a target, the path planning algorithm performs the target risk assessment again on all the targets in the added target area or the added targets according to the added target area or the added target.

In one example, a preset number of screened targets may be set to ensure that when the routing algorithm run is complete, there are a sufficient number of paths to meet the user's requirements. When the number of the screened target points is smaller than the preset number, the user can newly add a target point area or a target point and perform target point screening again.

In one example, since the muscles of the head cortex (e.g., temporal muscle, occipital muscle, posterior auricular muscle) are easily deformed, the entry point of the path trajectory obtained by the path planning is inaccurate, so that the medical device (e.g., electrode) implanted according to the location of the entry point will deviate from the location of the preset trajectory, resulting in an increased implantation risk. Therefore, the implantation point of the medical instrument (such as an electrode) needs to be determined on the skull, so that the position deviation of the implanted medical instrument is avoided, and the safety of the implanted medical instrument is increased.

The method for determining the skull implantation point region comprises the following steps:

setting the head cortex in a medical image (such as a registered CT image, a registered MRI image and a reconstructed cerebral cortex surface) as an entry point region, wherein all points on the head cortex are all entry points, and then adding all obtained entry points into an entry point set;

establishing connection lines between all screened target points in the screened target point set and all entry points respectively to form a connection line set;

and respectively arranging all the connecting lines in the connecting line set at the intersection points with the skull, wherein all the obtained implantation points are all implantation points, and then adding all the obtained implantation points into the implantation point set.

In one example, connecting lines are respectively constructed between all screened target points in the screened target point set and all entry points, and a many-to-many relationship may be established between all screened target points and all entry points, or a one-to-many relationship may be designed between all screened target points and all entry points.

In one example, all entry points in the entry point region may be divided into a left brain entry point located on the left lateral half brain side and a right brain entry point located on the right lateral half brain side by a mask at the middle plane between the left and right brains, and the left and right brain entry points may be marked. And simultaneously dividing all the target points in the target point region into a left brain target point positioned on the left half brain side and a right brain target point positioned on the right half brain side, and marking the left brain target point and the right brain target point. By means of the division, in the subsequent path planning process, path planning can be carried out between the left brain target point and the left brain entry point only, and path planning can be carried out between the right brain target point and the right brain entry point only, so that the situation that the path track obtained through screening traverses the left half brain and the right half brain of the brain is avoided.

In one example, when the implantation point and/or the entry point is located at or near the temporal lobe, the location of the temporal lobe may be determined from the temporal lobe mask, thereby increasing the safety of implanting a medical device (e.g., an electrode) at or near the temporal lobe.

In one example, it is determined whether the implantation point and/or entry point is located at the cerebellum and/or ventricles based on the cerebellum mask and the ventricles mask, and the implantation point and/or entry point is deleted when the implantation point and/or entry point is located at the cerebellum and/or ventricles.

In one example, a trajectory to be evaluated is constructed between all the screened target spots in the screened target spot set and all the implantation points in the implantation point set, and all the constructed trajectories to be evaluated are added to the trajectory set to be evaluated.

As shown in fig. 2, in step S2, the trajectory filtering includes at least one of preset structure trajectory filtering, edge trajectory filtering, trajectory length filtering, and trajectory angle filtering, or any combination thereof, for example, the preset structure trajectory filtering, and trajectory length filtering may be combined.

In one example, the preset structure track filtering, track length filtering and track angle filtering methods may be performed in any combination therebetween. For example, those skilled in the art can design the track length filtering, the track angle filtering and the preset structure track filtering in sequence according to requirements. The method can also be designed to sequentially carry out track angle filtering, preset structure track filtering, track length filtering and the like. The precedence order of the trace filtering is not limited thereto.

In one example, the track filtering for the preset structures includes track intersection filtering, where the track intersection filtering is to determine whether all tracks to be evaluated intersect with at least one preset structure in a preset structure set according to position information of each preset structure in the preset structure set. When the trajectory to be evaluated intersects at least one preset structure, the trajectory to be evaluated is excluded (i.e., deleted).

In one example, whether each trajectory to be evaluated intersects with the preset structure is determined by the blood vessel image obtained by the segmentation of the blood vessel image in the preset structure set and masks (such as a ventricle mask and a cerebellum mask) of other remaining preset structures. For example, a mask with a preset structure is projected to obtain a corresponding coordinate of the preset structure, and whether a track to be evaluated is intersected with a point corresponding to the coordinate or not can be calculated according to the coordinate; or respectively obtaining respective three-dimensional matrixes according to the blood vessel image, the ventricle mask and the cerebellum mask, and enabling the corresponding preset structure position in the three-dimensional matrixes to be 1 and the rest positions to be 0. And determining whether the coordinate has an intersection with the position 1 in the three-dimensional matrix according to all coordinate positions on the track to be evaluated, and deleting the track to be evaluated where the coordinate is located when the coordinate has an intersection with the position 1 in the three-dimensional matrix.

In one example, the preset structure trajectory filtering further comprises temporal lobe trajectory filtering. After the intersection filtering with the track, we obtain a subset of the tracks to be filtered by screening. In order to further increase the safety of the implantation trajectory of the medical device, temporal lobe trajectory filtering is performed on all the sub-trajectories to be filtered in the subset of trajectories to be filtered. Temporal lobe trajectory filtering includes the following steps:

providing a temporalis muscle thickness threshold at a temporal lobe, and obtaining temporal lobe position information according to a temporal lobe mask;

and judging the position relation between all entry points in the entry point set and the temporal lobe according to the temporal lobe position information, and judging the thickness of the temporal muscle at the intersection of the entry point and the temporal lobe when the entry point intersects the temporal lobe. And when the thickness of the temporal muscle at the intersection is larger than the threshold value of the thickness of the temporal muscle, excluding the entry point at the intersection and the sub-track to be evaluated where the entry point is located. And iterating the step, eliminating all entry points which intersect with the temporalis muscle and the thickness of the temporalis muscle at the intersection is larger than a temporalis muscle thickness threshold value, and further eliminating (namely deleting) the sub-tracks to be screened where the entry points are located, thereby reserving and obtaining the filtering track set.

It will be understood by those skilled in the art that the steps of performing the track intersection filtering and temporal lobe track filtering may be changed as desired, i.e., the temporal lobe track filtering may be performed first and then the track intersection filtering may be performed. The two filtering methods do not limit the sequence, as long as the track filtering can be realized step by step to improve the track safety.

In one example, the temporal muscle thickness threshold is an average thickness of the temporal muscle. The thickness of the temporal muscle can be obtained by extracting a CT image and an MRI image.

In one example, a three-dimensional matrix of the temporal lobe and coordinates of an entry point may be obtained through a temporal lobe mask and a skull mask, respectively, and when the coordinates of the entry point intersect with a temporal lobe position in the three-dimensional matrix of the temporal lobe or are located at the temporal lobe position, a relationship between a thickness of a temporal muscle where the entry point intersects with or at the temporal lobe position and an average thickness of the temporal muscle is determined.

In one example, to reduce subsequent path planning algorithm errors, such as reducing computational errors in subsequent track length filtering (described in detail below) and track angle filtering (described in detail below), we also perform edge track filtering on the tracks to be evaluated.

The method for filtering the edge track comprises the following steps:

obtaining position information of all first edge reference points at the edge of the skull mask according to the skull mask;

filtering the to-be-evaluated track falling at the edge of the skull in the to-be-evaluated track set according to the position information of all the first edge reference points;

obtaining position information of all second edge reference points at edges of a temporal lobe mask according to the temporal lobe mask;

And filtering the track to be evaluated which falls at the edge of the temporal lobe in the track set to be evaluated according to the position information of all the second edge reference points.

In an example, the trajectory to be evaluated to be subjected to the edge trajectory filtering may be a trajectory subjected to trajectory intersection filtering, may also be a trajectory subjected to temporal lobe trajectory filtering, may also be a trajectory subjected to trajectory intersection filtering and temporal lobe trajectory filtering, and of course, may also be a trajectory subjected to target point and implantation point filtering to obtain a trajectory to be evaluated. That is, the step of executing the edge trajectory is provided immediately after step S1, or may be provided after the trajectory intersection filtering step in the trajectory filtering of the preset structure, or may be provided after the temporal lobe trajectory filtering step, or may be provided after both the trajectory intersection filtering step and the temporal lobe trajectory filtering step.

In one example, the first edge reference point is a skull point at an edge in the skull mask. The second edge reference point is the temporal lobe point at the edge in the temporal lobe mask.

In one example, the skull mask and the temporal lobe mask are projected in the X direction, the Y direction and the Z direction in the three-dimensional space, respectively, to obtain the coordinates of the corresponding skull point and the coordinates of the temporal lobe point located at the edge. And when the coordinates of any point in the track to be filtered coincide with the coordinates of the skull point at the edge, and/or when the coordinates of any point in the track to be filtered coincide with the coordinates of the temporal lobe point at the edge, excluding the track to be filtered.

In one example, the temporal lobe margin trajectory may be determined first, and then the skull margin trajectory may be determined. The sequence of the two methods for judging the edge tracks is not limited to this, as long as the tracks at the skull edge and/or temporal lobe edge can be deleted, and the error of the subsequent path planning algorithm can be reduced.

In one example, the track filtering further comprises track length filtering, and the method of track length filtering comprises the steps of:

and comparing the relationship between the lengths of all the tracks to be evaluated and a preset track length range, judging whether the lengths of the tracks to be evaluated are within the preset length range, and when the lengths of the tracks to be evaluated are out of the preset length range, excluding the tracks to be evaluated.

In one example, the trajectory to be evaluated may be a trajectory filtered by an edge trajectory, and may also be a trajectory filtered by a preset structure. Due to the fact that the preset structure track filtering, the edge track filtering, the track length filtering and the track angle filtering are different in sequence, the tracks to be evaluated are different. The input track to be evaluated in each track filtering method is the track output after being filtered by the previous method.

In one example, the preset length ranges from 80mm to 140mm, and preferably, the preset length ranges from 80mm to 120 mm. When the track to be evaluated is larger than, for example, 120mm, the track to be evaluated is excluded. Of course, it can also be set that when the trajectory to be evaluated is larger than 130mm or 140mm, the trajectory to be evaluated is excluded.

In one example, the trajectory filtering further comprises trajectory angle filtering, and the method of trajectory angle filtering comprises the steps of:

acquiring an included angle set according to included angles formed between all tracks to be evaluated and normal vectors of tangent planes where the corresponding entry points are located;

comparing the relation between all included angles in the set of included angles and a preset implantation angle range respectively, judging whether the included angles are within the preset implantation angle range or not,

and when the included angle is within the preset implantation angle range, excluding the track to be evaluated corresponding to the included angle.

In one example, the tangent plane is the tangent plane of the curved surface on which the skull lies at the entry point. The entry point is a tangent point.

In one example, the angle is the angle formed between the trajectory to be evaluated and the normal vector to the tangent plane at the corresponding entry point.

In one example, the predetermined implantation angle range is set to 0-40 °, preferably 0-30 °, i.e. the maximum angle of electrode insertion is 30 °, when the angle is larger than, for example, 30 °, the trajectory to be evaluated is excluded.

In one example, the preset implantation angle range is set to 0-10 ° when the implantation site is located in the temporal lobe, and the trajectory to be evaluated is excluded when the insertion angle of the medical device (e.g., electrode), i.e., the included angle, is greater than 10 °.

As shown in fig. 3, the step S3 further includes:

providing a risk assessment index set;

and obtaining the relationship between all the filtering tracks in the filtering track set and each risk assessment index in the risk assessment index set according to the filtering track set and the risk assessment index set, and performing risk assessment.

In one example, the risk assessment indicators in the set of risk assessment indicators include a vascular risk assessment range and a gray matter sampling rate assessment range. The vascular risk assessment comprises the following steps:

obtaining a first distance set according to first distances between all the filtering tracks and the blood vessels respectively;

mapping all first distances in the first distance set into a blood vessel risk assessment range to obtain first assessment values of all filtering tracks.

In one example, samples are taken on a filtered trace, and the distance of each sample point on the filtered trace from the vessel is calculated. For example, the vascular risk assessment range is set to 0 to 1.

In one example, the method of mapping the first distance to the blood vessel risk assessment range is: when the first distance d between the filtering track and the blood vessel is larger than 10mm, the first evaluation value V of the filtering track₁A value of 0 indicates that the filtered trace is further away from the vessel, which is a low risk trace. When the distance d between the filtering track and the blood vessel is within the range of d being more than or equal to 0 and less than or equal to 3mm, the first evaluation value V of the filtering track₁Setting the filtering track to be 1, wherein the filtering track is close to the blood vessel and is a high risk track; when the distance d between the filtering track and the blood vessel is within the range of 3mm < d ≦ 10mm, the first evaluation value V of the filtering track₁The calculation method comprises the following steps:

wherein d represents the distance between the collection point and the blood vessel on the filtering track, d_minRepresenting the minimum distance between the preset filtering trajectory and the vessel.

In one example, d_minSet to 3mm, e.g. when d is 5mm, V₁＝(5-3)/(10-3)≈0.28。

In one example, the distance d between the filtering trace and the blood vessel is the distance between the closest acquisition point to the blood vessel among all the acquisition points in the filtering trace.

In one example, the gray matter sampling rate evaluation includes the steps of:

obtaining a gray matter sampling rate set of all filtering tracks according to the filtering track set;

And mapping all gray matter sampling rates in the gray matter sampling rate set to the gray matter sampling rate evaluation range to obtain second evaluation values of all filtering tracks.

In one example, a method of obtaining a gray matter sampling rate includes the steps of:

providing medical instrument parameters, and obtaining the number of contacts of the medical instrument intersected with the gray matter according to the medical instrument parameters;

and obtaining the ratio of the number of the contacts intersected with the gray matter to the total number of the contacts of the medical instrument according to the number of the contacts, wherein the ratio is the gray matter sampling rate.

In one example, the parameters of the medical instrument include, for example, the length of the medical instrument, the number of contacts of the medical instrument. And estimating the three-dimensional coordinates of the contact points of the medical instrument in the gray mask according to the length of the medical instrument and the number of the contact points, then respectively judging whether the contact points are superposed with the coordinates of gray in the gray mask according to the three-dimensional coordinates of all the contact points, and judging that the contact points are positioned in the gray when the contact points are superposed with the coordinates of the gray in the gray mask. And obtaining the gray matter sampling rate according to the number of the contact points positioned in the gray matter obtained by calculation and the total number of the contact points of the medical instrument.

The gray matter sampling rate is mapped into a gray matter sampling rate evaluation range, and when the gray matter sampling rate evaluation range is set to 0 to 1, the gray matter sampling rate is set to a second evaluation value, for example, 0.3, that is, the second evaluation value is 0.3. When the gray matter sampling rate evaluation range is set to 0-10, the gray matter sampling rate, for example, 0.3, may be mapped to 3 by a factor of 10, i.e., the second evaluation value is 3.

In one example, an optimized path set is obtained according to the first evaluation value and the second evaluation value;

the method for obtaining the optimized path set comprises the following steps:

performing first grading sorting on all filtering tracks according to the first evaluation value to obtain sorted tracks;

and dividing the sequenced tracks into at least two groups of tracks, and performing second hierarchical sequencing on each of the at least two groups of tracks according to a second evaluation value to obtain an optimized path set.

In one example, to obtain an optimal set of medical device (e.g., electrode) implantation protocols, the N tracks corresponding to each target point are ranked according to the first evaluation value and the second evaluation value.

The method of hierarchical ordering is: firstly, sorting in an ascending order according to a first evaluation value, wherein the track at the first has the lowest risk; and then dividing the sequenced tracks into K groups, and sequencing the K groups in each group according to the second evaluation value in a descending order, wherein the track positioned at the first in each group has the highest gray matter sampling rate. And selecting the track positioned in the first of all the groups as an optimized path, and adding the optimized path into the optimized path set.

In one example, the sorted trajectories are divided into groups having the same or about the number of medical devices to be implanted, for example, 10 groups, 12 groups, 30 groups, etc., and may be set to 2 groups, 7 groups, or more.

In one example, to ensure that the remaining path trajectories do not conflict with each other, the path planning method further comprises the following steps:

and step S4, performing dynamic path planning according to the optimized path set to obtain a conflict-free track set.

In one example, the dynamic path planning algorithm includes the steps of:

step S41, judging whether the two optimized paths in the optimized path set intersect with each other or not, so as to filter the two intersected paths in the optimized path set and obtain two disjoint sub-optimized track sets;

step S42, dividing the implantation points corresponding to all the sub-optimization tracks in the sub-optimization track set into at least two types of implantation points through Euclidean distances;

step S43, obtaining a sub-optimization track with a maximum third evaluation value corresponding to each of at least two types of implantation points according to preset evaluation values of all sub-optimization tracks in the sub-optimization track set;

step S44, judging the relationship between the maximum third evaluation value of the current category implantation point of the at least two categories of implantation points and the preset threshold values of other categories of implantation points of the at least two categories of implantation points to obtain a candidate track set;

step S45 determines a relationship between a third distance between the implantation points of each two candidate trajectories in the candidate trajectory set and a first preset distance range, so as to obtain a collision-free trajectory.

In one example, when any one of the optimized paths in the optimized path set intersects with at least one of the rest optimized paths, one of the intersected paths and the corresponding implantation point are deleted, the rest optimized paths are reserved, and whether the rest optimized paths intersect with each other or not is judged again; and when the optimized path is not intersected with the rest other paths, reserving the optimized path, adding the optimized path into the sub-optimized track set, and iterating the step until all paths which are not intersected with each other in pairs are obtained and added into the sub-optimized track set.

In one example, the method of screening sub-optimal trajectories screens candidate trajectories through a clustering algorithm to obtain collision-free trajectories. Preferably, the method employs a K-means algorithm for screening.

In one example, the preset threshold is a score at a preset quantile value. The preset score value may be set to 3/4 score values of all sub-optimized tracks of the same category, and may also be set to 2/3 score values, which may be set by those skilled in the art according to actual needs.

In one example, the preset evaluation value is obtained by performing weighting processing on the first evaluation value and the second evaluation value. In one example, the weight of the first evaluation value is set to, for example, 0.8, and the weight of the second evaluation value is set to, for example, 0.2.

In one example, the method of obtaining candidate trajectories is: and classifying the implantation points corresponding to all the sub-optimization tracks in the sub-optimization track set into 3 classes according to Euclidean distance by a K-means algorithm, and then calculating the maximum value of the sub-optimization track scores in each class and 3/4 quantile values of all the sub-optimization track scores in the class. For example, when the maximum score value of a first category of the 3 categories is less than the 3/4 quantile value of the second category score and simultaneously less than the 3/4 quantile value of the third category score, then the sub-optimized tracks in the first category are deleted. And then judging the relationship between the maximum score value of the second category and the 3/4 quantile value of the score of the third category, and deleting the sub-optimization tracks in the third category and adding all the sub-optimization tracks in the second category into the candidate track set when the maximum score value of the second category is greater than the 3/4 quantile value of the score of the third category.

In one example, the first predetermined distance range is set to 0-10 mm. When the third distance between the implanted points of every two candidate tracks is within the first preset distance range, deleting one of the every two candidate tracks; and when the third distance between the implanted points of every two candidate tracks is out of the first preset distance range, keeping the every two candidate tracks until the corresponding candidate tracks with the third distances out of the first preset distance range in all the candidate tracks are screened out, and determining the candidate tracks as conflict-free tracks.

In an example, the number of implantation points may be further filtered, and when a candidate track corresponding to the same implantation point in the candidate tracks is greater than 2, one of the candidate tracks is deleted.

In one example, the dynamic programming algorithm may also be implemented by a method comprising:

sorting according to the number of the optimized paths corresponding to each target in the target set to obtain an optimized target set;

and obtaining the conflict-free track according to the relation between the fourth distance between the optimized paths corresponding to every two optimized target points in the optimized target point set and a second preset distance range.

In one example, the target points corresponding to all the optimized paths in the optimized path set are sorted according to the optimized path number thereof, for example, sorted in an ascending order, so that the target point with the least optimized path can be preferentially selected, thereby reducing the conflict between every two paths in the optimized path set.

In one example, the second preset distance range is set to 0-10mm, when the distance between any one of the optimization paths in the set of optimization paths and at least one of the remaining optimization paths is within the second preset distance range, one of the optimization paths is deleted, and the step is iterated until all the optimization paths whose fourth distances from each other are outside the second preset distance range are obtained, where the optimization paths are collision-free trajectories.

In one example, a readable storage medium is provided in accordance with another embodiment of the present invention. "readable storage medium" of embodiments of the present invention refers to any medium that participates in providing programs or instructions to a processor for execution. The medium may take many forms, including but not limited to, non-volatile media, and transmission media. Non-volatile media includes, for example, optical or magnetic disks, such as storage devices. Volatile media includes dynamic memory, such as main memory. Transmission media includes coaxial cables, copper wire and fiber optics, including the wires that comprise a bus. Transmission media can also take the form of acoustic or light waves, such as those generated during Radio Frequency (RF) and Infrared (IR) data communications. Common forms of readable storage media include, for example, a floppy disk, a flexible disk, hard disk, magnetic tape, any other magnetic medium, a CD-ROM, DVD, any other optical medium, punch cards, paper tape, any other physical medium with patterns of holes, a RAM, a PROM, and EPROM, a FLASH-EPROM, any other memory chip or cartridge, a carrier wave as described hereinafter, or any other medium from which a computer can read.

The readable storage medium stores thereon a program or instructions that when executed by the processor performs the path planning method described above.

The path planning method and the readable storage medium provided according to various examples of the present invention have at least one of the following advantages:

(2) in the path planning of intracranial electrode implantation, the path planning method and the readable storage medium provided by the invention exclude the track intersecting with the intracranial dangerous area from the planned path;

(5) In the path planning process, the path planning method and the readable storage medium obtain a safe path planning scheme which not only obtains the optimal path planning scheme of a single target point from the perspective of the single target point, but also avoids mutual path conflict among a plurality of target points through the multi-electrode dynamic path planning method.

Although a few embodiments of the present general inventive concept have been shown and described, it will be appreciated by those skilled in the art that changes may be made in these embodiments without departing from the principles and spirit of the general inventive concept, the scope of which is defined in the appended claims and their equivalents.

Claims

1. A path planning method, the path planning method comprising the steps of:

2. The path planning method according to claim 1,

in step S2, the trajectory filtering includes a preset structure trajectory filtering for determining a relationship between all the trajectories to be evaluated in the trajectory set to be evaluated and all the preset structures in the preset structure set.

3. The path planning method according to claim 2,

the preset structure track filtering comprises:

judging whether all the tracks to be evaluated are respectively intersected with at least one preset structure in a preset structure set or not according to the position information of each preset structure in the preset structure set,

and when the track to be evaluated is intersected with the at least one preset structure, excluding the track to be evaluated.

4. The path planning method according to claim 2 or 3,

the preset structure track filtering further comprises the following steps:

providing a temporal lobe mask and a temporal muscle thickness threshold at a temporal lobe, and obtaining temporal lobe position information according to the temporal lobe mask;

providing a set of entry points located at the head cortex;

judging the position relation between all entry points in the entry point set and the temporal lobe according to the temporal lobe position information,

when an entry point in the set of entry points intersects the temporal lobe, determining a thickness of a temporal muscle at the intersection of the entry point and the temporal lobe,

when the thickness of the temporal muscle at the intersection is larger than the threshold value of the thickness of the temporal muscle, excluding an entry point at the intersection and a track to be evaluated in a set of tracks to be evaluated, wherein the track to be evaluated corresponds to the entry point at the intersection, so as to obtain the set of filtering tracks.

5. The path planning method according to claim 4,

the trajectory filtering may also include edge trajectory filtering,

the method for filtering the edge track comprises the following steps:

providing a skull mask;

obtaining position information of all first edge reference points positioned at the edge of the skull mask according to the skull mask;

and filtering the to-be-evaluated track which falls at the edge of the skull in the to-be-evaluated track set according to the position information of all the first edge reference points.

6. The path planning method according to claim 5,

the method for filtering the edge track further comprises the following steps:

obtaining the position information of all second edge reference points positioned at the edge of the temporal lobe mask according to the temporal lobe mask;

7. The path planning method according to claim 4,

the track filtering also comprises track length filtering, and the track length filtering method comprises the following steps:

comparing the length of all the tracks to be evaluated with the preset track length range, judging whether the length of the tracks to be evaluated is in the preset length range,

And when the length of the track to be evaluated is out of the preset length range, excluding the track to be evaluated.

8. The path planning method according to claim 4,

the track filtering also comprises track angle filtering, and the track angle filtering method comprises the following steps:

obtaining an included angle set according to included angles formed between all tracks to be evaluated and tangent planes where the corresponding entry points are located;

and when the included angle is out of the preset implantation angle range, excluding the track to be evaluated corresponding to the included angle.

9. The path planning method according to any one of claims 1-8,

the step S3 further includes:

providing a risk assessment index set;

and obtaining the relationship between all the filtering tracks in the filtering track set and each risk evaluation index in the risk evaluation index set according to the filtering track set and the risk evaluation index set, and performing risk evaluation.

10. The path planning method according to claim 9,

The risk assessment indicators in the set of risk assessment indicators include a vascular risk assessment range,

obtaining a first distance set according to first distances between all the filtering tracks and blood vessels respectively;

mapping all first distances in the first distance set to the blood vessel risk assessment range to obtain first assessment values of all filtering tracks.

11. The path planning method according to claim 10,

the risk assessment indicators also include gray matter sampling rate assessment ranges,

obtaining gray matter sampling rate sets of all the filtering tracks according to the filtering track set;

mapping all gray matter sampling rates in the gray matter sampling rate set to a gray matter sampling rate evaluation range to obtain second evaluation values of all filtering tracks.

12. The path planning method according to claim 11,

obtaining the optimized path set according to the first evaluation value and the second evaluation value;

the method for obtaining the optimized path set comprises the following steps:

performing first grading sorting on all the filtering tracks according to the first evaluation value to obtain sorted tracks;

And dividing the sequenced tracks into at least two groups of tracks, and performing second hierarchical sequencing on each group of tracks according to the second evaluation value to obtain the optimized path set.

13. The path planning method according to claim 11,

the gray matter sampling rate obtaining method comprises the following steps:

and obtaining the ratio of the contact number to the total number of the contacts of the medical instrument according to the contact number, wherein the ratio is the gray matter sampling rate.

14. The path planning method according to claim 1,

the step S1 further includes:

15. The path planning method according to claim 14,

providing an entry point region located in the scalp of a head and a set of entry points corresponding to the entry point region;

constructing a connecting line set between all screened target points and all entry points in the entry point set according to the screened target point set and the entry point set;

and obtaining the intersection points between all the connecting lines in the connecting line set and the skull respectively according to the connecting line set to form an implantation point set.

16. The path planning method according to claim 12,

the path planning method further comprises the following steps:

17. The path planning method according to claim 16,

the step S4 further includes:

step S41, judging whether the two optimized paths in the optimized path set intersect with each other or not, so as to filter the two intersected paths in the optimized path set and obtain two non-intersected sub-optimized trajectory sets;

step S42, dividing the implantation points corresponding to all the sub-optimization trajectories in the sub-optimization trajectory set into at least two types of implantation points through Euclidean distances;

Step S43, obtaining a sub-optimization track with a maximum third evaluation value corresponding to each type of implantation point in at least two types of implantation points according to preset evaluation values of all sub-optimization tracks in the sub-optimization track set;

step S44 is to determine a relationship between a maximum third evaluation value of a current category implant point of the at least two categories of implant points and a preset threshold of another category implant point of the at least two categories of implant points to obtain the candidate trajectory set;

step S45 determines a relationship between a third distance between the implantation points of each of the candidate trajectories in the candidate trajectory set and a first preset distance range, to obtain the collision-free trajectory.

18. The path planning method according to claim 17,

in step S44, a preset threshold value of the current category is obtained according to preset grading values of the third evaluation values of all the trajectories corresponding to the implant points of the current category in each category of implant points.

19. The path planning method according to claim 16,

sorting according to the number of the optimized paths corresponding to each target point in the target point set to obtain an optimized target point set;

and obtaining the conflict-free track according to the relation between a fourth distance between the optimized paths corresponding to every two optimized target points in the optimized target point set and a second preset distance range.

20. The path planning method according to claim 19,

at least one of temporal lobe, blood vessel, gray matter, ventricle, cerebellum, and brain function network is included in the preset structural set.

21. The path planning method according to claim 1,

the step S1 further includes:

displaying the target area on the medical image after image preprocessing,

the medical image after image preprocessing is an electronic computed tomography image and a magnetic resonance imaging image after image preprocessing,

the image preprocessing comprises the following steps:

registering the electron computed tomography image and the magnetic resonance imaging image into a template space of magnetic resonance T1 weighted imaging to obtain a registered medical image; and

Obtaining a blood vessel image after the segmentation of the cerebrovascular image,

the medical image after image preprocessing is a registered medical image.

22. A storable medium characterized by a first state of storage,

the readable storage medium having stored thereon a program or instructions which, when executed by a processor, performs the path planning method of any of claims 1-21.