A control method applied to automatic filing of pathological slice storage
By employing a rotating storage structure and the Hungarian method in the pathology slide scanner, automated sorting and optimal path control of pathology slides were achieved, solving the problems of cumbersome automatic archiving and slide collision, and improving the efficiency of automated organization of pathology slide storage.
Patent Information
- Application Number
- CN202311240717.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-25
- Publication Date
- 2025-12-26
- Estimated Expiration
- 2043-09-25
AI Technical Summary
The existing pathology slide scanning process and storage management cannot achieve automatic archiving, resulting in cumbersome operation, easy damage and mis-storage of pathology slides, and the possibility of slide collisions during the scanning process.
A rotating S-shaped storage structure is adopted, with each face divided into L layers, and each layer being a warehouse. Combining the Hungarian method and recursion, the automatic sorting and optimal path control of pathological slides are achieved by constructing warehouse state queues and archiving queues, ensuring that pathological slides are archived in sequence and that empty warehouses are reused.
It achieves efficient and automated archiving of pathology slides, ensuring no slide collisions during scanning, and optimizes path planning to improve the efficiency of automated organization of pathology slide storage.
Smart Images

Figure CN117262574B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of pathological section archiving, more particularly, it relates to a control method applied to automatic archiving of pathological section storage. BACKGROUND
[0002] Pathological examination is the gold standard for disease diagnosis and is irreplaceable in the diagnosis of diseases. The pathological section scanner is an important device for scanning and digitally storing pathological sections. The multi-channel pathological section scanner can realize one-time multi-section scanning, greatly simplifying the operation process of doctors. However, the existing scanning process and storage management cannot meet the automatic archiving of pathological sections. Since each patient often has multiple differently stained pathological sections in a pathological examination, and due to the limitations of the manufacturing process of pathological sections, it is not possible to dye the pathological sections of the same patient together in advance. Therefore, doctors need to manually check each section and store and record the location of the same patient in order. This process is not only tedious but also prone to accidents, which can easily cause human damage to pathological sections and misplacement and misrecording.
[0003] Therefore, it is necessary to realize the automatic archiving of pathological section storage, to automatically realize the sorting and archiving of pathological sections by the scanner, and to simplify human participation. In the automatic archiving process, it is not only necessary to simply reorder and place the pathological sections in the designated location. Rather, it is necessary to consider the scanning task and to realize automatic archiving in the same storage. And in the scanning process, the storage will be at least one empty position, i.e. the pathological section being scanned. Therefore, while realizing the automatic sorting and archiving of the storage, it is also necessary to realize the optimal control of taking and placing the sections during the scanning process. To ensure that there is no section collision during the scanning process, and to realize the optimal path to avoid excessive storage rotation. Therefore, it is necessary to combine the corresponding integer programming and path planning to effectively solve the above problems and realize the automatic archiving of the pathological section scanner. SUMMARY
[0004] The present application overcomes the shortcomings of the prior art and provides a control method applied to automatic archiving of pathological section storage, which has high processing efficiency and high automation.
[0005] A control method applied to automatic archiving of pathological section storage, including a rotating S-surface, and each surface is divided into L layers, each layer represents a warehouse, and each warehouse contains M rows and N columns of pathological section placement positions; the specific operation includes the following steps:
[0006] 1) Acquiring storage information step: acquiring various initial information in the storage, including the information of the storage, the information of the warehouse, and the initial pathological number information of each storage position;
[0007] 2) Taking out step: control the mechanical arm to take out the pathological section of the specified warehouse from the storage device and place it under the electron microscope;
[0008] wherein the mechanical motion time spent by the pathological section in the same warehouse is set to be negligible, the mechanical motion time across adjacent layers in the same face is denoted as w1, and the mechanical motion time across adjacent faces in the same layer is denoted as w2, w1 = 1, w2 = 100;
[0009] The mechanical time of adjustment between all warehouses constitutes the metric matrix wareMetricMat, the number of warehouses is wareNum, and the vector formed by the number of each warehouse is as follows:
[0010] wareID = [1, 2,..., i*S+j*L, i*S+(j+1)*L,..., S*L]
[0011] wherein i represents the ith face, and j represents the jth layer;
[0012] At the same time, the vector formed by the number of empty warehouses of each warehouse is wareEmptyNum, and the total number of all empty warehouses is as follows:
[0013]
[0014] wherein wareEmptyNum t represents the number of empty positions of the tth warehouse;
[0015] The individual information and overall information of the same pathological number are counted;
[0016] The specific matching process of the position is as follows:
[0017] The warehouse state queue wareCapacity and the warehouse archiving queue adjustIDList are constructed; wherein,
[0018] wareCapacity ∈ R wareNum
[0019] to record the number of remaining adjustment positions of each warehouse in real time; wareCapacity i represents the number of remaining positions of the ith warehouse, and the initial size is wareVolume; when the value is 0, it indicates that the warehouse is full and no longer accepts adjustment; in addition,
[0020] adjustIDList ∈ R wareNum×wareVolume
[0021] to record the final archived pathological section original coordinates, i.e. positionCoord value, of each warehouse in real time; adjustIDListij positionCoord value of the archived pathology slide in the jth position of the ith warehouse; its initial value is -1, and if its value is still -1 after the warehouse adjustment, it means that the position is empty after the warehouse adjustment and does not store any pathology slide;
[0022] 3) Scanning step: scanning the pathology slide of step 2), the mechanical arm taking out the next pathology slide to be scanned from the storage;
[0023] 4) Replacement step: taking out the scanned pathology slide and putting in the pathology slide to be scanned for scanning;
[0024] 5) Storage step: putting the scanned pathology slide back into the storage device;
[0025] 6) End step: repeating steps 2) to 5) until the scanning is completed.
[0026] Further, in step 1), the position is empty, and the pathology number of the position is recorded as 0; therefore, the initial information of all positions is recorded as:
[0027] oriMat∈R S×L×N×M
[0028] wherein oriMat represents the initial pathology number information of the s face, the l layer, the n column, and the m row, the value is the same for the same patient, and the value is 0 for the empty warehouse.
[0029] Further, the storage size of the pathology slide storage device is as follows:
[0030] warehouseSize=(S,L,N,M)
[0031] wherein S is the number of rotating faces of the storage, L is the number of layers of each face, N is the number of position columns of each layer, and M is the number of position rows of each column; the warehouse size is wareSize=(N,M), the number of warehouses is wareNum=S·L, and the warehouse capacity is wareVolume=N·M.
[0032] Further, the individual information includes: the original position coordinate positionCoord of each pathology slide, and the warehouse adjustment distance wareDistance of each pathology slide to each warehouse;
[0033] The overall information includes: the total warehouse adjustment distance slideSetMetricMat of each pathology number overall to each warehouse, and the number of pathology slides slideSetNum of each pathology number overall.
[0034] Further, the specific position matching step of taking out the pathology slide in step 2) is as follows:
[0035] 2.1) Empty warehouse setting and recovery steps: The recovery of empty warehouses is carried out in units of columns of each warehouse position, and when a warehouse is completely empty, it indicates that the column can be recovered; the number of recoverable empty warehouse columns is calculated as EN = emptyNum % M; according to wareEmptyNum, the specific warehouse is selected as the warehouse reserved for empty warehouses from large to small, and at the same time, the remaining warehouse position value of the warehouse is reduced by M in wareCapacity, until EN columns are selected as recoverable empty warehouses;
[0036] 2.2) Overall warehouse adjustment step based on the same pathological number: In order to ensure that pathological slides of the same pathological number are archived in the same warehouse, first, adjust the warehouse as a whole; and according to the sand bottle principle, according to the value of slideSetNum, sort from large to small, and preferentially adjust the pathological number with more quantity; the matching principle of adjustment is to select the warehouse with the least distance sum as the target warehouse according to slideSetMetricMat, which is recorded as wareID p , if the remaining warehouse position wareCapacity p of the corresponding warehouse is less than the number of pathological slides of the pathological number, then select the suboptimal warehouse as the target warehouse, until the wareCapacity p is greater than the number of pathological slides of the pathological number; when the warehouse position is matched, record the positionCoord coordinate value of all pathological slides of the pathological number in adjustIDList p ; at the same time, the remaining warehouse position value of the target warehouse is adjusted by the corresponding number, that is,
[0037] wareCapacity p = wareCapacity p - slideSetNum k
[0038] where k represents the kth pathological number; if all remaining warehouses are not enough to hold all pathological slides of the pathological number, then record that all pathological slides of the pathological number will be recorded;
[0039] 2.3) Individual pathological slide adjustment step: all pathological slides that cannot be adjusted as a whole are collected together, and the corresponding wareDistance value is taken out; 2 times of Hungarian matching are implemented for individuals, and the matching principle is to preferentially select the pathological slide with the smallest distance and its corresponding warehouse for matching according to wareDistance, and the warehouse is recorded as the target warehouse wareID p , and the positionCoord coordinate value of the pathological slide is recorded in adjustIDList p , and wareCapacity p= wareCapacity p -1; thus reciprocating until matching is completed;
[0040] 2.4) Collate target warehouse information step: according to the final adjustIDList, construct two groups of queues inIDList and outIDList; inIDList sequentially records the coordinate position source before archiving, which is changed from adjustIDList dimension to one-dimensional vector; while outIDList sequentially records the coordinate position destination after archiving.
[0041] Further, for different pathological numbers with the same number of pathological sections, the Hungarian method is used to achieve optimal matching of the warehouse positions.
[0042] Further, the automatic scheduling of the scanning process in step 3) requires that the specific automatic scheduling control be constructed according to the obtained inIDList and outIDList, and the scheduling queue be implemented according to the order of the scheduling queue. Specifically, two groups of queues need to be constructed: a section taking queue and a section placing queue. The section taking queue is used to sequentially record the warehouse position coordinates taken from the warehouse, denoted as takeOutQueue; while the section placing queue is used to sequentially record the warehouse position coordinates placed back after scanning, denoted as takeBackQueue; according to the final state dstMat of the scheduling, each position to be archived must be emptied in advance, and the pathological section on this position can be used as the next round of pathological section to be scanned for archiving. Therefore, the specific operation is as follows:
[0043] 3.1) Obtain initial archiving position step: select the initial archiving position according to the inIDList warehouse number order, and record the archiving position coordinates as back_id;
[0044] 3.2) Archiving position validity judgment step: if the selected position has not been archived and is not empty in the final (back_id≠0), it is recorded as valid and step 3.3) is entered; otherwise, it is recorded as invalid and step 3.4) is entered;
[0045] 3.3) Queue update step: store back_id in takeBackQueue in order, and store inIDList in takeOutQueue in order; then, search for the next archiving position; select outIDList as the next archiving position and update back_id=outIDList, then enter step 3.2); back_id back_id back_id
[0046] 3.4) End judgment step: Determine whether all pathological slides have been archived; if archiving is completed, end the construction of the scheduling queue and return takeOutQueue and takeBackQueue; otherwise, proceed to step 3.1) until the end.
[0047] The beneficial effects of this invention are as follows:
[0048] This invention primarily combines the Hungarian method to achieve optimal matching of storage compartments and a recursive approach to automate the scanning of pathology slides. This ensures optimal path control during scanning even when pathology slides are stored in random order; it also ensures ordered archiving of pathology slides even when they are stored in random order; and it enables centralized storage of empty compartments for reuse. Attached Figure Description
[0049] Figure 1 This is an example image of the rotating warehouse of the pathology slide scanner of the present invention;
[0050] Figure 2 This is an unfolded view of the rotating warehouse of the pathology slide scanner of the present invention;
[0051] Figure 3 This is a schematic diagram of the mechanical time measurement matrix for warehouse transfers between different warehouses in this invention;
[0052] Figure 4 This is a flowchart of the scheduling queue construction in this invention. Detailed Implementation
[0053] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.
[0054] like Figures 1 to 4 As shown, a control method for automatic archiving of pathology slides includes a rotated S-shaped solid, with each face divided into L layers. Each layer represents a warehouse, and each warehouse contains M rows and N columns of pathology slide placement positions. The specific operation includes the following steps:
[0055] 1) Steps for obtaining warehouse information: Obtain various initial information within the warehouse, including warehouse information, storage area information, and initial pathology number information for each storage location. Specifically, as follows... Figure 1 The image shows the storage compartment of a rotating hexahedral pathology slide scanner. Each face is divided into three layers (top, middle, and bottom), and each layer contains K compartments (M rows and N columns). The entire hexahedron is referred to as a storage compartment, and the N×M compartments in each layer are collectively referred to as a warehouse (e.g., ...). Figure 1The part of the middle frame is called a position, and the part of the middle frame where the pathological section is stored is called a position. When each pathological section is stored in the position, the pathological number of the pathological section is recorded. If the pathological section comes from the same patient, the pathological numbers are the same. At the same time, the position can be empty, and the pathological number of the position is recorded as 0 at this time. Therefore, the initial information of all positions is recorded as a matrix:
[0056] oriMat∈R S×L×N×M
[0057] Where oriMat represents the initial pathological number information of the s-th surface, the l-th layer, the n-th column, and the m-th row. The value is the same for the same patient, and the value is 0 for the empty position.
[0058] The storage size of the pathological section storage device is as follows:
[0059] warehouseSize=(S,L,N,M)
[0060] Where S is the number of rotating surfaces of the storage, i.e. 6 in this embodiment, 6 surfaces. L is the number of layers of each surface, i.e. 3 layers in this embodiment. N is the number of columns of positions in each layer, which is 3 columns by default in this embodiment, and M is the number of rows of positions in each column, which is 3 rows by default in this embodiment. The size of the warehouse is wareSize=(N,M), the number of warehouses is wareNum=S·L, and the capacity of the warehouse is wareVolume=N·M.
[0061] 2) Taking the section step: controlling the mechanical arm to take out the pathological section in the specified position from the storage device and place it under the electron microscope.
[0062] Where the mechanical movement time spent by the pathological section when adjusting in the same warehouse is ignored, the mechanical movement time across adjacent layers in the same surface is recorded as w1, and the mechanical movement time across adjacent surfaces in the same layer is recorded as w2, w1=1, w2=100. That is, when calculating the section taking strategy, the basic information of the warehouse needs to be known. As shown in FIG. 1, the mechanical time required to transfer from warehouse A to warehouse B is 2w1+1w2. Figure 2
[0063] The mechanical time of adjusting between all warehouses constitutes the metric matrix wareMetricMat, and the number of warehouses is wareNum. As shown in the case of FIG. 2, the vector formed by the numbers of each warehouse is as follows: Figure 3
[0064] wareID=[1,2,...,i*S+j*L,i*S+(j+1)*L,...,S*L]
[0065] Where i represents the i-th surface, and j represents the j-th layer.
[0066] Meanwhile, the vector of the empty warehouse number of each warehouse is wareEmptyNum, and the total empty warehouse number is as follows:
[0067]
[0068] where wareEmptyNum t represents the empty position number of the tthwarehouse.
[0069] The individual information and overall information of the same pathological number are counted. The individual information includes the original position coordinate positionCoord of each pathological slice and the warehouse adjustment distance wareDistance of each pathological slice to each warehouse.
[0070] The overall information includes the total adjustment distance slideSetMetricMat of each pathological number to each warehouse and the pathological slice number slideSetNum of each pathological number.
[0071] Therefore, the specific matching process of the position is as follows:
[0072] The warehouse state queue wareCapacity and the warehouse archiving queue adjustIDList are constructed. Wherein,
[0073] wareCapacity∈R wareNum
[0074] The remaining adjustment position number of each warehouse is recorded in real time. wareCapacity i represents the remaining position number of the ithwarehouse, and the initial size is wareVolume. When the value is 0, it indicates that the warehouse is full and no longer accepts adjustment. In addition,
[0075] adjustIDList∈R wareNum×wareVolume
[0076] The final archiving positionCoord value of each pathological slice of each warehouse is recorded in real time. adjustIDList ij represents the positionCoord value of the pathological slice archived in the jthposition of the ithwarehouse. The initial value is -1, and if the value is still -1 after adjustment, it indicates that the position is empty after adjustment and does not store any pathological slice.
[0077] The specific position matching steps of the pathological slice are as follows:
[0078] 2.1) Empty warehouse setting and recycling steps: The empty warehouse recycling is carried out in units of each position column. For example, Figure 1As shown, when the boxed warehouse is full, it indicates that the column can be recycled. The number of recyclable empty warehouse columns is EN = emptyNum % M. According to wareEmptyNum, the specific warehouse is selected as the warehouse with reserved empty warehouse from large to small, and the remaining warehouse value of the warehouse is reduced by M in wareCapacity, until EN columns are selected as recyclable empty warehouses.
[0079] 2.2) Overall warehouse adjustment step based on the same pathology number: In order to ensure that the same pathology number is archived in the same warehouse, it is first adjusted as a whole. According to the sand bottle principle, according to the value of slideSetNum, it is sorted from large to small, and the pathology number with more quantity is preferentially adjusted. The matching principle of adjustment is to select the warehouse with the least distance sum as the target warehouse according to slideSetMetr Mat, which is recorded as wareID p If the remaining capacity of the corresponding warehouse wareCapacity p is less than the number of pathological sections of the pathology number, the suboptimal warehouse is selected as the target warehouse, and the process is repeated until the wareCapacity p is greater than the number of pathological sections of the pathology number. Among them, for different pathology numbers with the same number of pathological sections, the optimal matching of the warehouse is realized by using the Hungarian method, and the matching principle is as described above.
[0080] When the warehouse position matching is completed, the positionCoord coordinate value of all pathological sections of the pathology number is recorded in the corresponding warehouse adjustIDList p At the same time, the remaining warehouse value of the target warehouse is adjusted by the corresponding number, that is,
[0081] wareCapacity p = wareCapacity p - slideSetNum k
[0082] Among them, k represents the kth pathology number. If the remaining capacity of all warehouses is not enough to hold all pathological sections of the pathology number, it is recorded that all pathological sections of the pathology number will be recorded.
[0083] 2.3) Individual pathological section adjustment step: All pathological sections that cannot be adjusted as a whole are collected together, and the corresponding wareDistance value is taken out. Two Hungarian matching are performed for individuals, and the matching principle is to preferentially select the pathological section with the smallest distance and its corresponding warehouse for matching, which is recorded as the target warehouse wareID p , and adjustIDList pThe positionCoord coordinate value of the pathological section is recorded, and the wareCapacity is also recorded p = wareCapacity p -1. Thus, the matching is completed.
[0084] 2.4) Target storage position information arrangement step: According to the final adjustIDList, two queues inIDList and outIDList are constructed. The inIDList sequentially records the coordinate position source before archiving, which is changed from the dimension of adjustIDList to a one-dimensional vector. The outIDList sequentially records the coordinate position destination after archiving.
[0085] 3) Scanning step: The pathological sections in step 2) are scanned, and the mechanical arm takes out the next pathological section to be scanned from the storage. The automatic scheduling of the scanning process, i.e., the specific automatic scheduling control, needs to construct a scheduling queue according to the obtained inIDList and outIDList, and implement scheduling according to the order of the scheduling queue. Specifically, two queues need to be constructed: a take-out queue and a put-back queue. The take-out queue is used to sequentially record the storage position coordinates taken out from the warehouse, denoted as takeOutQueue. The put-back queue is used to sequentially record the storage position coordinates after scanning and putting back into the warehouse, denoted as takeBackQueue. According to the final state dstMat of the scheduling, each position that needs to be archived must be emptied in advance, and the pathological section on this position can be used as the next round of pathological sections to be scanned for archiving.
[0086] 4) Replacement step: The scanned pathological section is taken out and placed into the pathological section to be scanned.
[0087] 5) Storage step: The scanned pathological section is put back into the storage device.
[0088] 6) End step: Repeat steps 2) to 5) above until the scanning is completed.
[0089] Therefore, the specific operation is as follows:
[0090] 3.1) Initial archiving position acquisition step: The initial archiving position is selected according to the inIDList warehouse number sequence, and the coordinate of the archiving position is recorded as back_id.
[0091] 3.2) Archiving position validity judgment step: If the selected position has not been archived and is not empty in the end (back_id≠0), it is recorded as valid and step 3.3) is entered. Otherwise, it is recorded as invalid and step 3.4) is entered.
[0092] 3.3) queue updating step: storing back_id sequentially in takeBackQueue and storing inIDList back_id sequentially in takeOutQueue. Then, searching for the next archiving position. Selecting outIDList back_id as the next archiving position and updating back_id = outIDList back_id , and then entering step 3.2).
[0093] 3.4) end judging step: judging whether all the pathological sections are archived. If the archiving is completed, ending the construction of the scheduling queue and returning takeOutQueue and takeBackQueue. Otherwise, entering step 3.1) until the end.
[0094] The specific scheduling implementation is as follows: constructing the sequential section taking queue and the sequential section placing queue takeOutQueue and takeBackQueue.
[0095] Taking out the pathological section at the position coordinate recorded in the i-th record in takeOutQueue and placing it under the scanner for scanning (initial state i = 1). Taking out the pathological section at the position coordinate recorded in the i+1-th record in takeOutQueue and placing it in the scanning state. After the i-th pathological section is scanned, taking it back from the scanner and placing the i+1-th pathological section for scanning. Placing the scanned i-th pathological section back to the i-th record in takeBackQueue. Repeating the above storing and taking steps until the last record in takeOutQueue is scanned. After the last pathological section is scanned, placing it back to the last record in takeBackQueue.
[0096] In summary, the present scheme combines the Hungarian method to achieve the optimal matching of the positions and the recursive method to achieve the automatic control of the pathological section scanning. The optimal path control in the scanning process is achieved under the condition that the pathological sections are stored in disorder. The pathological sections are archived in order under the condition that the pathological sections are stored in disorder. The empty positions are centralized to be reused.
[0097] The above only describes the preferred embodiments of the present application. It should be noted that those skilled in the art can make several improvements and refinements without departing from the concept of the present application, and these improvements and refinements should also be considered within the protection scope of the present application.
Claims
1. A control method applied to automatic filing of pathological slide storage, characterized in that, The S-face body is rotated, and each face is divided into L layers, each layer representing a warehouse, and each warehouse containing M rows and N columns of pathological slide placement positions; the specific operation includes the following steps: 1) Obtain warehouse information step: obtain various initial information in the warehouse, including warehouse information, warehouse information, and initial pathological number information of each storage position; 2) Take a slice step: control the mechanical arm to take out the pathological slice in the specified storage position from the warehouse equipment and place it under the electron microscope; Where, the mechanical movement time spent by the pathological slice when adjusting in the same warehouse is ignored, the mechanical movement time across between adjacent layers in the same face is recorded as w1, and the mechanical movement time across between adjacent faces in the same layer is recorded as w2, w1=1, w2=100; The mechanical time for adjusting between all warehouses constitutes the metric matrix wareMetricMat, the number of warehouses is wareNum, and the vector formed by the number of each warehouse is as follows: wareID=[1,2,…,i*S+j*L,i*S+(j+1)*L,…,S*L] Where i represents the i-th face, and j represents the j-th layer; At the same time, the vector formed by the number of empty warehouses in each warehouse is wareEmptyNum, and the total number of all empty warehouses is as follows: wherein, wareEmptyNum t represents the number of empty positions of the tth warehouse; 3) Scanning step: scan the pathological slice in step 2), and the mechanical arm takes out the next pathological slice to be scanned from the warehouse; 4) Replace step: take out the scanned pathological slice and put in the pathological slice to be scanned for scanning; 5) Storage step: put the scanned pathological slice back into the warehouse equipment; 6) End step: repeat steps 2) to 5) above until the scanning is completed.
2. The control method for automatic filing of pathological slide storage according to claim 1, wherein, In the case of empty position in step 1), the pathological number of the position is recorded as 0; therefore, the initial information of all positions is recorded as: oriMat∈R S×L×N×M , wherein oriMat represents the initial pathological number information of the s-th surface, the l-th layer, the n-th column, and the m-th row; the initial pathological number information of the same patient is the same; and the pathological number of the empty position is recorded as 0.
3. The control method for automatic filing of pathological slide storage according to claim 1, wherein, The size of the pathological slice warehouse equipment is as follows: warehouseSize=(S,L,N,M), where S is the number of rotating faces of the warehouse, L is the number of layers of each face, N is the number of columns of storage positions of each layer, and M is the number of rows of storage positions of each column; then the warehouse size is wareSize=(N,M), the number of warehouses is wareNum=S·L, and the warehouse capacity is wareVolume=N·M.
4. The control method for automatic filing of pathological slide storage according to claim 1, wherein, The individual information and the overall information of the same pathological number are counted; The specific matching process of the storage position is as follows: Constructing warehouse state queue wareCapacity and warehouse archive queue adjustIDList; wherein, wareCapacity ∈ R wareNum To record the number of remaining warehouse positions in real time; wareCapacity i The initial size of the i-th warehouse is wareVolume, and when its value is 0, it means that the warehouse is full and no longer accepts warehousing. In addition, adjustIDList ∈ R wareNum×wareVolume to record the positionCoord value of the final archived pathology slide in each warehouse in real time; adjustIDList ij represents the positionCoord value of the archived pathology slide in the jth position of the ith warehouse; its initial value is -1, and if its value is still -1 after the warehouse adjustment, it indicates that the position is empty after the warehouse adjustment and does not store any pathology slides; The individual information includes: the original storage position coordinates positionCoord of each pathological slice, and the wareDistance of each pathological slice to each warehouse; The overall information includes: the total wareDistance of each pathological number to each warehouse, and the number of pathological slices slideSetNum of each pathological number; The specific storage position matching step of taking pathological slices in step 2) is as follows: 2.1) Empty warehouse setting and recovery step: The recovery of empty warehouse is carried out in column units of each warehouse position, when a warehouse is empty, it means that the column can be recovered; the number of recoverable empty warehouse columns is calculated as EN = emptyNum % M; according to wareEmptyNum, the specific warehouse is selected as the warehouse reserved for empty warehouse from large to small, and at the same time the remaining position value of the warehouse in wareCapacity is reduced by M until EN columns are selected as recoverable empty warehouse; 2.2) Overall repositioning step based on the same pathological number: in order to ensure that the pathological slides of the same pathological number are archived in the same warehouse, first reposition them as a whole; and according to the sand bottle principle, according to the value of slideSetNum, sort from large to small, and give priority to pathological numbers with more quantity for repositioning; the matching principle of repositioning is to select the warehouse with the least distance sum as the target warehouse according to slideSetMetricMat, recorded as wareID p If the remaining warehouse capacity wareCapacity p of the corresponding warehouse is less than the number of pathological slides of the pathological number, then select the suboptimal warehouse as the target warehouse, and adjust the remaining warehouse capacity wareCapacity p until it is greater than the number of pathological slides of the pathological number; when the warehouse capacity is matched, record the positionCoord coordinate values of all pathological slides of the pathological number in the corresponding warehouse adjustIDList p ; at the same time, the remaining warehouse capacity value of the target warehouse is adjusted by the corresponding number, that is wareCapacity p = wareCapacity p -slideSetNum k Wherein k represents the kth pathological number; if all the remaining positions of the warehouse are not enough to put all the pathological sections of the pathological number, it is recorded that all the pathological sections of the pathological number will be recorded; 2.3) Remaining pathological section individual warehouse adjustment step: all pathological sections that cannot be adjusted as a whole are collected together, and the corresponding wareDistance value is taken out; 2 times of Hungarian matching are implemented for individuals, and the matching principle is to preferentially select pathological sections with the smallest distance and match the corresponding warehouse according to wareDistance, which is recorded as the target warehouse wareID p , and the positionCoord coordinate value of the pathological section is recorded under adjustIDList p , and wareCapacity p = wareCapacity p -1 is recorded at the same time; thus, the process is repeated until the matching is completed; 2.4) Arrangement of target position information step: according to the finally obtained adjustIDList, two groups of queues inIDList and outIDList are constructed; wherein inIDList sequentially records the coordinate position source before archiving, which is changed from adjustIDlist dimension to one-dimensional vector; and outIDList sequentially records the coordinate position destination after archiving.
5. The control method for automatic filing of pathological slide storage according to claim 4, wherein, For different pathological numbers with the same number of pathological sections, the Hungarian method is used to realize the optimal matching of the positions.
Citation Information
Patent Citations
Control method for automatic archiving of pathological film storage
CN117104756A