經濟部中央標準局員工消費合作社印家 A7 B7 五、發明説明（丨） 【發明背景】 本發明係提供-種頭部手術㈣以統及方法，尤則 利用電腦及影像技術，配合立體枝系统，將腦部組織透 過即時類示的立體模型與三個正交剖面影像，使醫師對病 情的診斷與手㈣程中料錢容易與㈣確，且在手術 過程中對病人不會造成二次傷害者。 現今傳統的頭料術中，㈣由二維_層揀描圖 片來作病情的診斷，據以研判病灶位置與規 路径， 之後再進行手術，這種方式至少有以下三個缺點： ，、手術前的診斷與手術路徑規劃的猜確度不高。 雖然在二維的斷層掃描圏片中都有比例尺^醫師判斷病 灶位置及尺寸，但這只是二_資料，事實上頭及病灶 皆是三維物體’因此臀師在診斷時只能靠其對頭部解剖 組織的知識概略評估病灶與頭部各部位的㈣方位，如 果病灶或規H的路徑靠近重要組織，將因無法確認於手 術過程中可否避免傷害重要組織而被返放棄可執行的手 術0 、手術中的定位精確度不高。 由於前項所規劃的手術路徑都是在影像中進行的，而手 術時通常都是藉由解剖學上的標誌（如耳根）及預先规 劃的手術路徑來決定出開顱位置與範園，這種方式很難 在病人頭部定出正確的路徑，因此通常醫師開刀時都會 開一個較大的開口以容許可能的方位誤差，如此將造成 病人較大的傷害。又如採用框架式定位方式，則有存定 (請先閱讀背面之注意事項再填寫本頁) 訂 本紙張尺度適用中國國家標隼（CNS ) Α4規格（210Χ297公釐） A7 B7 嫂濟部中央標隼扃員工消費合作社印繁 發明说明（丄） — — 偉及重復定位等不便之處。 ,醫師需要豐富的臨床經驗。 前述的手術前的路徑規劃與手術中的定位，必須仰 師多年的臨床手術經驗才能有良好的手術品質，而典^科主治醫師是非常耗時的，加上醫師的經驗與:2| 識的不同，使得手術品質因人而異。 爲了改t傳統手術只依據二維影像進行手術比較不精確 的缺舜’目則使用的方法是有框式（frame )導引系統 遠移方式的原理是病人在做斷層掃描前先在頭骨上釘入四 櫚矣座，然後裝上定位用的框架，掃描完畢後，一樣進行 彩像糢娄重建，然後醫師在影像t進行手術路徑规劃，手 術時，將疋位框架換成導引框架，先前醫師在影像中規劃 的矛#硌徑經由軟體程式轉換成導引框架上每一個關節的 角度與位移，此時醫師只將定位框架上每一個關節的角度 與佟#調好即可，然後依照導引框架所導引的方向進行手 漱。然而’該框式導引系統卻有以下之缺失： 對病人的傷害較大。 該方法必須在做斷層掃描前先在頭上釘入四個基座，而 真要持續到手術完畢，這對病人來説是額外的傷害，且造 成f術後活動的不便。 二、手術時缺乏輿影像的互動。 該方法在調好定位框架上每一個關節的角度與位移之 後，费師就只能依照此方向進行手術，沒有對應的影像可 以參考，若手術位置充滿血水，將難以判别目前手術位置 Λ 本紙張尺度適用中國國家橾牟（CNS ) A4規格（21〇X 297公楚）Employees 'Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs, Consumers' Printing House A7 B7 V. Description of the Invention (丨) [Background of the Invention] The present invention provides a system and method of head surgery, especially using computer and imaging technology in conjunction with the three-dimensional branch system Through the real-time three-dimensional model and three orthogonal cross-section images of the brain tissue, it is easy and accurate for the doctor to diagnose the disease and the process of handwork, and it will not cause a second time to the patient during the operation. The injured. In today ’s traditional head surgery, two-dimensional layer-by-layer picking pictures are used to diagnose the disease. Based on this, the location and path of the lesion are determined, and then the operation is performed. This method has at least the following three shortcomings: The diagnosis and surgical path planning are not highly accurate. Although there are scales in two-dimensional tomographic scans, the doctor judges the location and size of the lesion, but this is only two data. In fact, both the head and the lesion are three-dimensional objects. The knowledge of the anatomy of the body is used to roughly evaluate the sacral orientation of the lesion and various parts of the head. If the path of the lesion or the rule H is close to important tissues, it will be impossible to confirm whether it is possible to avoid damage to important tissues during the operation and will be abandoned. 2. The positioning accuracy during surgery is not high. Because the surgical path planned in the previous paragraph is performed in the image, the cranium location and fan garden are usually determined by anatomical landmarks (such as the root of the ear) and the planned surgical path. It is difficult to determine the correct path on the patient's head, so usually a large opening is opened by the physician to allow a possible azimuth error, which will cause greater harm to the patient. If the frame positioning method is adopted, it is stored (please read the notes on the back before filling in this page). The size of the paper is applicable to China National Standards (CNS) A4 specifications (210 × 297 mm) A7 B7 Central Ministry of Economic Affairs Standards of the Employees' Consumer Cooperatives, India and India (说明)-Wei and repeated inconveniences such as positioning. Doctors need extensive clinical experience. The aforementioned path planning before surgery and positioning during surgery must be based on many years of clinical surgery experience of the teacher to have good surgical quality, and the attending physician of the department is very time-consuming, plus the experience of the physician and: 2 | The difference makes the quality of surgery different from person to person. In order to change the traditional surgery, which is only based on two-dimensional images, the surgery is relatively inaccurate. The method used is to move the frame guidance system far away. The principle is that the patient first places the skull on the skull before performing a tomographic scan. Nailed into the four palm pedestal, and then installed the positioning frame, after scanning, the color image model reconstruction is also performed, and then the doctor performs the surgical path planning in the image t. During the operation, the niche frame is replaced by the guide frame. The spear # 硌 diameter planned by the doctor in the image was converted into the angle and displacement of each joint on the guide frame by a software program. At this time, the doctor only adjusted the angle of each joint on the positioning frame with 佟 #, and then Hand rinse in the direction guided by the guide frame. However, the frame guidance system has the following shortcomings: It is more harmful to the patient. This method must first nail the four bases on the head before tomography, and it must continue until the operation is completed, which is an additional injury to the patient and causes inconvenience of postoperative activities. Second, the lack of public image interaction during surgery. After adjusting the angle and displacement of each joint on the positioning frame in this method, the master can only perform surgery in this direction. There is no corresponding image to refer to. If the surgical position is filled with blood, it will be difficult to determine the current surgical position. Standards are applicable to China National Standards (CNS) A4 (21〇X 297)
I 之 注 意 I 再 填 寫 本 頁 -訂 經濟部中央標準扃員工消費合作社印製 A7 B7 五、發明说明（》） 所在的組織，同時無祛預測由目前的方位繼續深入會碰 什麼組織，因而危險性較高。 5 【發明概述】 本發明之主要目的在提供一種頭部手術用導今丨系统及 法，其在於使進行頭部手術時，可使醫師與影像及時及勤为 藉由影像而充份了解目前手術位置周園組織，進而能避 手術器械傷及重要之組織，以維護病人的健康^ ^ 本發明之次一目的在提供一種頭部手術用導3丨系统 法，乃藉由無框式的導引系統，使得病人在進;f亍手術、方 受到二次傷害。 '脅 本發明之再一目的在提供一種頭部手術用導幻系 法，其系統之體積較傳統者爲小，因此使用上非常蕾及方 不會坊礙醫師工作。 /舌’ 基於此，本發明所提供之頭部手術用導引系殊 係由影像重建、定位裝置及方位校準等三大單元方法’ 該影像重建單元是以行進方塊法將二維掃描影像^級成； 體影像’並求出手術器械端點所在的三個正交1 建出义 剖面影像。該定位裝置單元係利用磁場強度方面輿备意 的位置與手術器械的方位作量測。該方位校準L量》則檩記 標記爲媒介，再由基因法則找出影像座標與定3 =則是以 的轉換關係。藉由上述各單元所組成之系统，可供=== 手術前診断病灶並規劃手術路徑，也可在手術時協助醫^ 精確定出手術器械的方位及其所在的組織剖面；醫師於影 像中所規劃的手術路徑可經由此轉換關係以定位裝置精^ I . 訂 各 f請先閲讀背面之注意事項再填寫本頁〕 本紙張尺度適用中國國家標隼（CNS ) A4規格（210X297公釐 A7 B7 五、發明説明（十） 的定出，而手術進行中，手術器械的方位也可轉換成栢到 於影像座標的方位，益於影像中即時類示出手術器械所在 的組織剖面影像。 本發明之其他目的及功能經配合下列圖式作進一步諸| 明後將更爲瞭解。 ~I pay attention to this I fill in this page again-set the central standard of the Ministry of Economic Affairs 扃 printed by the consumer consumer cooperative A7 B7 V. The description of the invention (") At the same time, there is no need to predict which organization will continue to touch from the current position, so it is dangerous Sex is higher. 5 [Summary of the invention] The main purpose of the present invention is to provide a guidance system and method for head surgery, which is to enable doctors and images to understand the current situation in time and diligently when performing head surgery. The tissues around the surgical site can avoid injury to important tissues by the surgical instruments to protect the patient's health. ^ A second object of the present invention is to provide a guide method for head surgery. The guidance system allows the patient to progress; f 亍 surgery, the party suffered secondary injuries. Another object of the present invention is to provide a guided system for head surgery. The volume of the system is smaller than that of the conventional one. Therefore, it is very easy to use and will not hinder the doctor's work. / Tongue 'Based on this, the guidance system for head surgery provided by the present invention consists of three major unit methods: image reconstruction, positioning device, and orientation calibration. The image reconstruction unit is a two-dimensional scanning image based on a traveling square method ^ Level into a volume image 'and obtain three orthogonal 1s where the end points of the surgical instrument are located to create a sense profile image. This positioning device unit uses the publicly known position in magnetic field strength and the orientation of surgical instruments for measurement. The azimuth calibration L amount "is marked with the mark as the medium, and then the genetic rule is used to find the image coordinate and the setting 3 = is the conversion relationship between. The system composed of the above units can be used to === diagnose the lesion and plan the surgical path before surgery, and also assist the doctor during surgery to accurately determine the position of the surgical instrument and the tissue section where it is located; the physician in the image The planned surgical path can be used to locate the device through this conversion relationship. I. For each order, please read the notes on the back before filling in this page.] This paper size applies to China National Standard (CNS) A4 specifications (210X297 mm A7). B7 5. The description of the invention (ten), and during the operation, the orientation of the surgical instrument can also be converted to the orientation of the image coordinate, which is beneficial to the real-time image showing the tissue section image of the surgical instrument. Other objects and functions of the invention will be further understood in conjunction with the following drawings |
I 寫 本 % 【圖式簡要説明】 囷一爲顯示本發明之硬禮架構之示意阈； 圖二爲顯示本發明之磁場式定位裝置之示意圖； 圖三爲顯示本發明之使用系統流程圖； 圈四爲顯示以本發明影像系統所重建出之頭部皮膚外觀影| 像圖； ^ 囷五爲顯示本發明影像系統所重建出之剖面影像； 圖六爲顯示本發明影像系統所重建出之紐合剖面影像； 囷七爲顯示本發明以組合剖面原理做出之立體圖形； 圖八爲顯示本發明計算標記的影像座標流程圖； 囷九爲類示本發明之系統使用示意圏； 經濟部中央榡隼局員工消費合作社印來 圖十爲顯示本發明之系統配合機械手臂之使用示意圖； 圖十一爲顯示本發明之影像重建單元的流程圖； 圖十二爲類示本發明在進行電腦斷層影像及處理時的 電腦螢幕畫面； 圏十三面爲本發_行手術關料料的電腦勞幕畫 【阐號説明】 本紙張尺度㈣國家轉（CNS) A4規格（21GX297公釐 (5 )標記 (6)機械手臂 (7 )頭架 (8 )手術台 A7 B7 五、發明説明（[) (1)發射器 (2 )電腦 (3)接收器 (4 )探針 【實施例説明】 本發明所提供之頭部手術用導引系統，係由影像重建、 定位裝置及方位校準等三大單元所組成，而其硬體架構係 如囷一所示，電腦斷層掃描（CT)影像單元經由網路傳到 電版中，定位裝置單元則將手術器械的方位傳給電腦，電 腦將這兩個資料經過計算處理以後，於顯示幕中顯示手術 器械所在的頭部組織影像。 —其中，該影像重建單元的作業流程係如圖十一所示； 貼著標記的病人經斷層所得的影像其灰階约有“Μ種，爲 了，化特定組織的對比，以調整level輿wind〇w値方式來 決定類示的灰階範固。由於掃描影像可能會有雜訊與不必 要的區域’例如U形的頭墊，必須將各種雜訊去除，之後 "T依照繫師的需要求出任意剖面影像。又，手術常需要知 道特定组織的大小、表面積、…等的參考資訊，因此有必 要從掃插影像中描出特定組織的輪廓，並依照所畫的组織 ，廓計算組織的體積、表面積、方位等資訊。繼之，利用 亍進方塊法（Marching Cube)重建頭顱表面或顱骨立體影 像’再將立體影像與各種任意剖面同時類示在螢幕中。有 關襟記在影像中的座標則可由標誌影像的超高灰度値，經 ’氏張尺度剌巾陶家縣（CNS) A4規格（2丨Q χ 297公董） — 丨--r-----------Iΐτ------.^ (請先閱讀背面之注意事項再填寫本頁) 經濟部中央梂準局員工消費合作.社印製 A7 A7 經濟部中央榡率局員工消費合作社印掣 B7 五、發明説明（6 ) 閥値分類，即可在斷層掃描影像中找到標誌的位置及座標。 如囷八所示本發明計算標記的影像座流程圖，由於屬 於同一個標記的像點彼此之間必定兩兩相連，即使由於澎 像品質的關係而不相連，也必定在某個距離（標記的大小 内，因此可以經由此種關係將屬於附加標記的像點依所屬 的附加標記而分群。接著要把標記的中心位置找出，由於 我們已有標記所有組成像點的位置資料，因此可用重心法 求出其中心位置。求得位置後，還必須轉換成以爲單位 的空間座標，假設每一張掃描影像的尺寸爲2〇〇X 200mm， 掃描解析度爲512Χδ12個像點，因此每個像點的長寬分别 爲200 + 512 = 〇.39〇625随，而掃描間隔爲lmm，因此像點的 高爲1mm。則上面所求出的附加標記中心位置（χ，γ，z ) 中，X與Y必須乘以〇·39〇625 , Z乘以i才是空間座標値， 如此才算求出方位校準中所需要的標記相對於三維模型影 像座標系統的座標値，然後將此座標値記綠下來。 有了標記的影像座標後，還要有標記的裝置座標。標記 的裝置座標是病人的頭部在手術台上囡定之後，先去除掉 有明顯位移的標記，然後以定位装置量測出每一個要做方 位校準的標記的座標値，然後將此座標値記錄下來。 本發明之定位裝置係採用磁場式或光學式定位裝置，如 固九所示，其係以發射器1發射磁場或光訊號，接收器3 接收其所在位置的磁場訊號或光訊號，傳給定位裝置電子 單元轉換成其所在位置相對於發射器i的位置與方向後再 傳給電腦2。本系統採用一個發射器〗對兩個接收器3的系 本紙張尺度適用中國國家榇準（CNS ) A4規格（210X297公釐） (請先閱讀背面之注意事項再填寫本頁) 丁 A7 B7 _、發明説明（7 ) (請先閱讀背面之注意事項再填寫本頁) 乞，一個接收器裝在手術器械上以量測器械的方位，另 g固定在病人頭上以防萬一病人與發射器相對關係改變時 色夠進行補償。 該方位校準單元則是以標記爲媒介，再由基因法則找出 ，像座標與定位裝置座標的轉換關係者。其大概的方法是 亂數產生很多組初始（第一代）參數値，然後將這些參 U直帶入差異函數中求得函數値，函數値較小的參數値有 t高的機會與其他參數値交換其參數中的某些位元（cross ver)以求得第二代的參數値，重複這個動作直到有某組參 ί的函數値達到目標値（由於差異函數無法趨近零，所以 i*們依所能容忍的定位誤差設定差異函數的目標値）。其特 ί是對整個面做搜尋，如此容易找到全域極値，而且收斂 i快’不須做微分運算。找到六個空間參數後，代回轉換 ί陣就可知道標記的影像座標與裝置座標的轉換關係。找 丨轉換矩陣以後，就可開始進行手術，手術進行時，將接 :器裝在手術器械上以隨時量測手術器械的方位，再將這 :方位經由轉換矩陣轉成影像座標，電腦即可將器械與立 :影像同時顯示，酱師也就能知道目前手術位置是否爲預 t規劃的路徑，以及目前手術位置周圍組織的情形。 本發明之系統使用流程係如圖三所示，其分爲手術前與 術進行中兩個部份。手術前的部份係先在病人頭上貼上 .顆約2mm 直徑的金屬球標記，然後進行頭部電腦斷層掃 ’接^將二維斷廣择描影像做影像處理，並將處理過的 _______Ϊ--;----- 本紙張纽關㈣Ϊ家辟⑽s)A4規格(21QX 297公麓） 經濟部中央標準局員工消費合作社印製‘ __；___if__ 本纸張尺度適用中國國家標隼（CNS ) A4規格（21 OX297公釐） A7 B7 五、發明説明（F ) 影像進行艾_像$建，以錢醫㈣行錄㈣與规畫 手術路徑。同時由處理過的影像中，找出標記中心點的意 像座標。 手H行時先用定位装置定出所有標記的裝置座標（即 相對於座標系的座標），再將此組座標與先前找出 me行方位校準以找出兩座標系的轉換關 後，醫師進行手術時’手術刀的位置就 了經由此轉換_、轉換成祕座標系的 =腦模型可輿立麟同時類示。《下依序 一、 在病人頭部貼著標記： 方位校準是以標記爲媒介，找出影像座 轉換關係，因此首先必須在病人頭上㈣記5、(裝== 示），本發明以能於斷層掃描影像中清楚分辨的金尸 ::早:ΐ數目最少要三個，貼的越多，單-檩記的誤差 會被平均掉，g此越猜確，但考量使用方便性 個。貼的仫置以不易移動的地方爲原則，例如 確 等等，由於方位較準的結果在榡記禁= 地方較爲精確，因此最好能夠貼在病灶的周固。 二、 電腦斷層掃描： 常用的斯層掃描技術有兩種：一是以χ光爲光源的 (Computed T〇mography>以及用磁場和無線電波 磁振造影 MRI (Magnetic Res〇nance Image)。 遇仃的 由於本系統要做精確的定位，因此選擇精確的ct爲掃描浐 ί «Λ (請先閱讀背面之注意事項再填寫本頁}I copy% [Schematic description] Figure 1 is a schematic threshold showing the hard gift structure of the present invention; Figure 2 is a schematic diagram showing the magnetic field positioning device of the present invention; Figure 3 is a flowchart showing the use system of the present invention; The fourth is to display the skin image of the head reconstructed by the imaging system of the present invention | image map; the fifth is to display the section image reconstructed by the imaging system of the present invention; the sixth is to display the reconstructed button of the imaging system of the present invention Combined section images; VII shows the three-dimensional graphics of the present invention based on the principle of combined sections; Figure VIII shows the image coordinate flow chart of the calculation mark of the present invention; 囷 Nine is a schematic diagram showing the use of the system of the present invention; 中央 Central of the Ministry of Economic Affairs Printed by the Bureau ’s Consumer Cooperatives Figure 10 is a schematic diagram showing the use of the system of the present invention with a robotic arm; Figure 11 is a flowchart showing the image reconstruction unit of the present invention; and Figure 12 is a view showing the computer tomography of the present invention Image and computer screen during processing; 圏 Thirteen sides of this computer _ screen drawing of surgical materials [instruction of explanation] This paper size 纸张National Transfer (CNS) A4 specification (21GX297 mm (5) marking (6) robot arm (7) head holder (8) operating table A7 B7 5. Description of the invention (() (1) transmitter (2) computer (3) ) Receiver (4) Probe [Explanation of the embodiment] The guidance system for head surgery provided by the present invention is composed of three major units such as image reconstruction, positioning device and orientation calibration, and its hardware architecture is as follows As shown in the first example, the computerized tomography (CT) image unit is transmitted to the electronic version via the network, and the positioning device unit transmits the position of the surgical instrument to the computer. The computer calculates these two data and displays them on the display screen. Display the head tissue image where the surgical instrument is located.-Among them, the operation process of the image reconstruction unit is shown in Figure 11; the image obtained by tomography of the labeled patient has about "M kinds of gray scales. For the comparison of specific organizations, the level of gray scale can be determined by adjusting the level of the window. Since the scanned image may have noise and unnecessary areas, such as a U-shaped head pad, various noises must be included. After removal, " T according to the needs of the teacher Obtain arbitrary section images. Also, surgery often needs to know the reference information of the size, surface area, etc. of specific tissues, so it is necessary to trace the contours of specific tissues from the scanning images and calculate them according to the drawn tissues and contours. Tissue volume, surface area, orientation and other information. Then, using the Marching Cube method to reconstruct the skull surface or skull stereoscopic image, the stereoscopic image and various arbitrary sections are displayed on the screen at the same time. The coordinates in the middle can be represented by the ultra-high gray scale of the logo image, which is in accordance with the A-size specification of Taojia County (CNS) A4 (2 丨 Q χ 297) — 丨 --r ------- ---- Iΐτ ------. ^ (Please read the precautions on the back before filling out this page) Employee Cooperative Cooperative of the Central Government Procurement Bureau of the Ministry of Economic Affairs. Printed by the agency A7 A7 Employee Cooperative of the Central Government Procurement Bureau of the Ministry of Economic Affairs Stamp B7 V. Description of the invention (6) Classification of the valve 即可 can find the position and coordinates of the sign in the tomographic image. As shown in Figure 28, the flowchart of calculating the image base of the marker of the present invention, because the image points belonging to the same marker must be connected with each other, even if they are not connected due to the quality of the image, they must be at a certain distance (marker Within the size, the image points belonging to the additional mark can be grouped according to the attached additional mark through this relationship. Next, the center position of the mark must be found. Since we have the position data of all the image points of the mark, we can use it. The center of gravity method is used to find the center position. After the position is obtained, it must be converted into spatial coordinates in units. Assume that the size of each scanned image is 200 × 200mm and the scanning resolution is 512 × δ12 pixels. The length and width of the image point are 200 + 512 = 0.39 and 625, respectively, and the scanning interval is 1 mm, so the height of the image point is 1 mm. Then the center position (χ, γ, z) of the additional mark obtained above , X and Y must be multiplied by 0.339625, Z is multiplied by i to be the space coordinate 値, so as to calculate the coordinates needed for the orientation calibration relative to the coordinates of the three-dimensional model image coordinate system, then Write down this coordinate in green. After the marked image coordinates, there must be marked device coordinates. The marked device coordinates are after the patient's head is set on the operating table, the mark with obvious displacement is removed first, Then use the positioning device to measure the coordinate 値 of each mark to be azimuth calibrated, and then record this coordinate 値. The positioning device of the present invention uses a magnetic field or optical positioning device, as shown by Gu Jiu. The transmitter 1 transmits a magnetic field or optical signal, and the receiver 3 receives the magnetic field signal or optical signal at its location, transmits it to the positioning unit electronic unit, converts it to the position and direction of its location relative to the transmitter i, and then transmits it to the computer 2 . This system uses one transmitter. For the paper size of two receivers 3, the Chinese National Standard (CNS) A4 specification (210X297 mm) is applicable. (Please read the precautions on the back before filling this page.) D7 B7 _, Description of the invention (7) (Please read the precautions on the back before filling out this page). One receiver is mounted on the surgical instrument to measure the orientation of the instrument, and the other is fixed to the patient. Just in case the patient ’s relative relationship with the transmitter changes when the color is sufficient to compensate. The orientation calibration unit is based on the marker and then found by the genetic law, such as the conversion relationship between coordinates and positioning device coordinates. Its approximate The method is to generate random sets of initial (first-generation) parameters 乱, and then bring these parameters U directly into the difference function to find the function 値. The function 値 smaller parameters 値 have a high chance of being exchanged with other parameters 其Some bits in the parameters (cross ver) to obtain the second-generation parameters 値, repeat this action until a function 组 of a certain set of parameters 値 reaches the target 値 (because the difference function cannot approach zero, i * men according to The tolerable positioning error sets the target of the difference function 値). Its special feature is to search the entire surface, so that it is easy to find the global limit 而且, and the convergence i is fast, and no differentiation is required. After the six spatial parameters are found, the transformation relationship can be known on behalf of the array. The conversion relationship between the marked image coordinates and the device coordinates can be known. After finding the transformation matrix, you can start the operation. When the surgery is in progress, attach the connector to the surgical instrument to measure the orientation of the surgical instrument at any time, and then convert this: orientation to the image coordinates through the transformation matrix, and the computer can The instrument and the stand: the image is displayed at the same time, and the sauce chef can also know whether the current surgical location is a pre-planned path, and the situation of the tissue around the current surgical location. The process of using the system of the present invention is shown in Fig. 3, which is divided into two parts before and during operation. Before the operation, the patient's head was first affixed with a metal ball with a diameter of about 2mm. Then a computerized tomography scan of the head was performed. The two-dimensional tomographic images were image processed, and the processed _ ______ Ϊ-; -------- This paper is a new standard of A4 size (21QX 297 feet) Printed by the Consumer Cooperatives of the Central Standards Bureau of the Ministry of Economic Affairs' __; ___if__ This paper size applies to Chinese national standards ( CNS) A4 specification (21 OX297 mm) A7 B7 V. Description of the invention (F) The image is made of Ai_Like $, which is recorded and planned by Qianyi. At the same time, from the processed image, find the image coordinates of the center point. When the hand is in line H, first use the positioning device to determine all the marked device coordinates (that is, the coordinates relative to the coordinate system), and then align the set of coordinates with the previously found me line to find the conversion between the two coordinate systems. The physician During the operation, the position of the scalpel was converted through this conversion, and the = brain model converted into the mysterious coordinate system can be shown at the same time. "Next, first, put a mark on the patient's head: The orientation calibration is based on the mark as the medium to find the image base conversion relationship. Therefore, you must first mark 5 on the patient's head (install == show). The present invention can The golden corpses that are clearly distinguished in the tomographic image: at least three: the number of maggots must be at least three. The more they are posted, the errors of the single-memory will be averaged out, and the more accurate this is, but considering the convenience of use. The placement of the stick is based on the difficult-to-move place, such as true, etc., because the result of the more accurate orientation is in the forbidden place = the place is more accurate, so it is best to stick to the surrounding area of the lesion. Computer tomography: There are two commonly used tomography techniques: one is using χ light as the light source (Computed Tomography) and magnetic resonance imaging (MRI) using magnetic fields and radio waves. Because the system needs to perform precise positioning, select the precise ct for scanning 浐 ί «Λ (Please read the precautions on the back before filling this page}
、1T A7 B7 五、發明説明（)） 像來源，未來也將使用MRI，並以CT (fusion )，如此獲得的影像就能兼顧CT的準確與MRI的 影像品質，也由於本系統以CT爲掃描影像，因此所選擇的 標記必須能於CT中顯像。 三、 影像處理： 斷層掃描完畢後，在進行影像重建之前，爲了使重建的 效果更好，以及便於醫師的診斷，通常必須對掃描影像進 行影像處理。本系統在二維影像處理方面具備點、線以及 自動去除不必要區域的功能，經由設定點、線的寬度及灰 階値便可去除影像中的雜訊與不需要的部份，以及晝出特 定組織的邊界，以方便計算體積與重建（如圖十二所示）。 四、 三維重建與二維剖面顯示： 影像經過處理以後，接下來要進行影像重建與剖面的求 取，然後組合成立體影像。本系統採用行進方塊法進行影 像重建，該法的主旨在產生一個特定灰階値的等値面 (iso-surface)，這個方法常被應用在連續影像的三維重 建。行進方塊法針對斷層掃描影像堆疊的體積資料的等値 面做表面著色的工作，此一等値面的灰階値由使用者依所 要重建的表面灰階値來指定，在體積資料中，相鄰兩張掃 描影像八個像素組成一個方塊或稱體素（voxel )，這個演 算法便是要決定所要著色的等値面如何和這個方塊相交， 然後依序對體積資料中每一個方塊做處理，再將每一個相 交的面組合成三角網格，再進行著色。 _I_ (請先閱讀背面之注意事項再填寫本頁) 經濟部中央標準局員工消費合作社印聚 本紙張尺度適用中國國家標準（CNS ) Α4規格（210X 297公釐） A7 B7 五、發明説明（p) 五、 計算標記的影像座標： 請 閱 讀 背 之 注 意 事 項 再 填 寫 本 頁 影像重建只是方便使用者觀察，使使用者不須再像以往 傳統手術般用想像的，而在方位校準中，需要有標記的澎 像座標，也就是影像中標記的位置，因此接下來必須將標 記從影像中辨識出來並求出其中心位置，其流程係如圖八 所示，首先分析影像資料中，所有像點的灰階分布，並找 出附加標記的“門檻値“，亦即灰階値在此門檻値以上的 像點爲屬於附加標記的像點，然後將這些像點的“位置“記 錄下來，這裡所謂的位置爲（X像點，Y像點，掃描間隔）， 前兩者爲像點在其所屬的二維影像上的位置，第三者爲此 二維影像在所有二維影像中的位置。 -訂 六、 量測標記相對於定位裝置的座標（裝置座標）： 有了標記的影像座標後，還要有標記的裝置座標◊標記 的裝置座標是病人的頭部在在手術台上固定之後，先去除 掉有明類位移的標記，然後以定位裝置量測出每-個要做 方位校準的標㈣座標値，減將此座财下 七、 方位校準： 經濟部中央標準局員工消費合作社印掣 料基料㈣㈣像座標與定位裝 八、 手術進行： 找到轉換矩陣錢，就可開始進行手術，手術進行時， 將，收器裝在手術器械上以隨時量測 的方位，存 ㈣轉換矩陣轉成料㈣，減將器械與立 •同時類示，醫師就能知道目前手術位置是否爲預先 本紙(CNS) 二 297疋 經濟部中央標隼局貝工消費合作'杜印製 A7 B7 五、發明説明（丨丨） ---- 規劃的路徑，以及目前手術位置周園組織的情形。 九、病灶診斷與手術路徑規劃： ” 做出前述的立體影像之後，醫師就可以從立體 調整任意剖面的位置來進行㈣與手術路 丄 由於影像是立嫌的’醫師不必像從二 :::描影像來診斷與規劃手術路程’因此其锖確度將: 本系統是對病人的CT 進行重建，㈣能重建 廣外觀，也就是皮廣輿空氣的交界面，這個面在= 影像的灰階値约爲5〇Q ,設定好灰階値以後，再指定每一 體素在空間中實際所佔體積的長寬高，就可進=維= (如圖四所示之重建後的皮膚模型囷）。 本系統的立體影像除了行進方塊法所重建的三維模型 外，由於醫師主要是要看病人體内的資訊，因此本系統還 需要能夠同時顯示通過任意點的任意方向剖面影像，這樣 對醫師的診斷才有幫助，因此可以説剖面影像才是眞正重 要的本系统的剖面分爲兩種’ 一是純粹剖面（pure cross section)，例如正交剖面，這種剖面是以一個平面切過斷 層掃描的髏積資料，然後將切過的影像資料顯示出來（如 圖五所示〉’另一種是組合剖面（combined cross section)，將純粹剖面與行進方塊法的立體模型同時顯示， 然後將超出立體模型剖面範園的剖面影像去除，同時將被 剖掉的立體模型去除，如此看起來就像是病人的頭被削去 Ο 本紙張尺度適用中國國家標準（CNS ) A4規格（210X 297公慶） (請先閱讀背面之注4^事項再填寫本頁) -訂- 五、發明説明（丨> ) A7 B7 塊’同時看到制去部份的剖面影像（如_六所示）。匿 五所不的純㈣面係於斷層掃插的體積資料中，求出剖适 位置的平面輿體積資料蚊的灰階値，然_示出來者。 = = = = : =輿行，塊法的以 1玄除ej面以外的部份。如將各剖面 組口則可做出如圖七般更爲複雜的立禮囷形。 本系統的方位校準必須在頭部與發射器的相對方位固定 HU:他們的相對方位不固定，則頭部任意-點的 裝，座標也將不能確定，會導致錯誤的結果，因 ==:與發射器的相對方位固定，通常是將發: Ϊ 上（如固九所示），則手術台8與頭 ^定Π 定’加上病人頭部與頭架7的相對方位 二頭部與發射器1的相對方位便可固定；而 :標二針4附有-接收器3,且該探針4係用於接 生改f 發射器的讀方位意外發 夺像中丰=有 碰到發射器導致其位置改變，則 二，械與病人頭部的相對方位關係將會發生錯 :二沒i現=錯誤，就必須重新進行方位校準以更正錯誤， 爲了避’則醫師可能做出錯誤判斯導致手術失敗。 接收器30 (如H的f生，乃在病人頭部貼上另外-個 器械上作^ 5 )，此接收器3G與原本固定在手術 _ ’用的接收器3接收同—個發射器的訊號。若 II - - - - I- -1 I --- - - - J 卜1 I__ - - ...... 丁 f · 、-口 (請先閱讀背面之注意事項再填寫本頁) 經濟部中央標準局員工消費合作社印繁 -—- itc 本紙張尺度適用中國----- 知隼（CNS ) A4規格（210X297公釐） 經濟部中央榡芈局員工消費合作衽印製 A7 B7 五、發明説明（丨）） 將裝置座標（接收器3相對於發射器的方位）再乘以接收 器3〇相對於發射器方位的逆矩陣，則裝置座標變成接收器 3相對於接收器的方位，如此一來，只要接收器3〇與病 人頭部的相對方位固定，即使發射器因故移動，也不會對 裝置座標造成影響。 本系統在實際手術中使用時，餐師操作的手術器械輿病 人頭部的相對方位能夠在影像中及時且正確的顯示，而在 影像中也同時顯示著醫師在手術前規劃的手術路徑，爲了 使手術此夠依照先則所規劃的手術路徑進行，醫師必須在 手術中邊看著影像邊移動手術器械，直到影像中的手術器 械與預定路徑重合。 如圖十所示，本系統爲了增加酱師使用本系統的方便 性，其另一實施例係將定位裝置與機械手臂6結合，在機 械手臂6末端裝上一個附有定位裝置接收器3的探針4，並 求出機械手臂6的座標與裝置座標的轉換關係。之後在方 位校準元畢後，程式自動將影像上的路徑轉換成相對於裝 置座標，再轉成相對於機械手臂的座標，如此機械手臂即 可移動至預定的手術路徑方位，並固定在這個方位上，手 術時，醫師只需要依照機械手臂所定的方向進 由於本料餘框式（frameless)WU統，故具有 下的優點： 、對病人沒有傷害性：本發明以標記黏貼於病人頭上， 以標記代替定位框架，因此對病人沒有傷害性。 本紙張尺度適财關家縣（CNS) Α4規格（21()><297公 (請先閲讀背面之注意事項再填寫本頁) λ. 訂 A7 ----—---B7__：_ 五、發明説明（丨V ) 一手術時可與影像及時互動：在動手術時，醫師可透過 ，像知道^手術位置周®有什麼組織，同時影像中也 了看到目則手街的位置再繼續下去會碰到什麼組織，會 ^會有危險’以及該如何避免，因而得以提高手術之安 全性。 -使用方便·由於趙積小，沒有框架，因此使用上非常 靈活，不會妗礙贅師工作。 以上所述者僅爲用以解釋本發明之較佳實施例，並非企 圖據以對本發明作任何形式上之限制，是以，凡有在相同 ，發月精神下所作有關本發明之任何修飾或變更 ，皆仍應 包括在本發明意固保護之範疇。 (請先閱讀背面之注意事項再填寫本頁) τν 經濟部中央標準局貝工消費合作社印繁 _______—___ 本紙張尺度適用中國國家標準（CNS )八4祕（210x297公瘦）1T A7 B7 V. Explanation of the invention ()) In the future, MRI will be used as the source of the image, and CT (fusion) will be used in this way. The image obtained in this way can take into account both the accuracy of the CT and the image quality of the MRI. The image is scanned, so the selected marker must be able to be imaged in CT. 3. Image processing: After the tomographic scan is completed, before the image reconstruction, in order to make the reconstruction effect better and to facilitate the diagnosis of the doctor, the scanned image must usually be image processed. The system has the functions of points, lines, and automatic removal of unnecessary areas in two-dimensional image processing. By setting the point, line width, and gray scale, the noise and unwanted parts in the image can be removed, as well as day-to-day output. Boundaries of specific tissues to facilitate calculation of volume and reconstruction (see Figure 12). 4. Three-dimensional reconstruction and two-dimensional section display: After the image is processed, the next step is to reconstruct the image and obtain the section, and then combine them to form a volume image. This system uses the traveling block method for image reconstruction. The main purpose of this method is to generate a specific gray-scale chirped iso-surface. This method is often used in the three-dimensional reconstruction of continuous images. The marching square method performs surface coloring on the isosurfaces of the volume data of the tomographic image stack. The grayscale of this isosurface is specified by the user according to the surface grayscale to be reconstructed. In the volume data, the phase Eight pixels from two adjacent scanned images form a block or voxel. This algorithm is to determine how the isosurface to be colored intersects this block, and then process each block in the volume data in order. , And then combine each intersecting surface into a triangle mesh, and then color. _I_ (Please read the notes on the back before filling out this page) Printed by the Consumer Standards Cooperative of the Central Bureau of Standards of the Ministry of Economic Affairs The paper size is applicable to Chinese National Standards (CNS) A4 specifications (210X 297 mm) A7 B7 V. Description of the invention (p ) 5. Calculate the image coordinates of the mark: Please read the notes on the back and fill in this page. The image reconstruction is only for the convenience of the user, so that the user no longer needs to use the imagination as in traditional surgery. In the orientation calibration, it is necessary to have The coordinate of the marked image is the position of the mark in the image. Therefore, the mark must be identified from the image and its center position must be obtained. The process is shown in Figure 8. First, all the image points in the image data are analyzed. Distribution of gray levels, and find the "thresholds" with additional marks, that is, the image points above the threshold of the gray levels are those that belong to the additional marks, and then record the "position" of these image points, here The so-called position is (X image point, Y image point, scanning interval), the first two are the position of the image point on the two-dimensional image to which it belongs, and the third one is the second one. Position of the image in all 2D image. -Order six. Coordinates of the measuring mark relative to the positioning device (device coordinates): After the marked image coordinates, the marked device coordinates are required. The marked device coordinates are the patient's head after it is fixed on the operating table. First, remove the mark with clear displacement, and then use the positioning device to measure each of the coordinates of the bearing to be calibrated. Reduce the bearing. 7. Bearing calibration: Printed by the Consumers' Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs Material coordinates and positioning of the base material. 8. Surgery: Find the transformation matrix and start the surgery. When the surgery is in progress, install the receiver on the surgical instrument to store the transformation matrix at any time. Transformed into materials, and reduce the number of instruments to be displayed at the same time, the physician can know whether the current surgical location is in advance of the paper (CNS) 2297 (Central Bureau of Ministry of Economic Affairs, Shellfish Consumer Cooperation) Du printed A7 B7 5. Description of the invention (丨 丨) ---- The planned path and the situation of the current weekly organization of the surgical site. Nine, lesion diagnosis and surgical path planning: ”After making the aforementioned three-dimensional image, the doctor can adjust the position of any section from the three-dimensional to carry out the operation and the operation path. Because the image is suspected, the doctor does not have to look like from two ::: Tracing the image to diagnose and plan the surgical journey 'therefore its accuracy will be: This system is to reconstruct the CT of the patient, and can not reconstruct the wide appearance, that is, the interface of the Pi Guangyu air. This surface is at the gray level of the image. It is about 50Q. After setting the gray scale 値, then specify the length, width, and height of the volume that each voxel actually occupies in space, then you can enter = dimensional = (as shown in the reconstructed skin model in Figure 4) In addition to the three-dimensional model reconstructed by the traveling cube method of the stereo image of the system, since the physician mainly depends on the information in the patient's body, the system also needs to be able to simultaneously display cross-sectional images in any direction through any point, so that the diagnosis of the physician It is helpful, so it can be said that the cross-section image is the most important. The cross section of this system is divided into two types. One is the pure cross section, such as the orthogonal section. The cross section is a cross section of cross-section data scanned through a tomographic scan, and then the cut image data is displayed (as shown in Figure 5). 'The other is a combined cross section, which combines the pure cross section with the traveling block method. The three-dimensional model is displayed at the same time, and then the section image beyond the three-dimensional model section Fan Yuan is removed. At the same time, the three-dimensional model that has been cut off is removed, so it looks like the patient's head has been cut off. 0 This paper size applies Chinese National Standards (CNS ) A4 size (210X 297 public holidays) (Please read Note 4 ^ on the back before filling out this page)-Order-V. Description of the invention (丨 >) A7 B7 block 'Simultaneous section image (As shown in _ 六). The pure ㈣ planes of the Five Unknowns are tied to the volume data of the fault sweep, and the gray scale 蚊 of the mosquito is obtained from the plane volume data of the appropriate position. =: = Yuxing, the block method takes 1 part except the ej plane. If the sections are grouped, a more complicated ritual shape like Figure VII can be made. The orientation calibration of this system must be in The relative orientation of the head and the transmitter is fixed HU: Their relative orientation is not fixed, then the head-arbitrary-point installation, the coordinates will not be determined, which will lead to wrong results, because ==: the relative orientation with the transmitter is fixed, usually it will be sent: Ϊ 上 （如 固9), then the operating table 8 and head ^ 定 Π 定 'plus the relative orientation of the patient's head and headrest 7 and the relative orientation of the head and the transmitter 1 can be fixed; and: two needles 4 are attached -Receiver 3, and the probe 4 is used to read the position of the transmitter. The transmitter accidentally captures the image. Zhongfeng = there is a change in the position of the transmitter when it touches the transmitter. An error will occur: if there is no error = error, you must re-align the position to correct the error. In order to avoid this, the doctor may make a wrong judgment and cause the operation to fail. Receiver 30 (such as H's health, it is affixed to another device attached to the patient's head for 5), this receiver 3G receives the same transmitter as the receiver 3 used in surgery Signal. If II----I- -1 I ------J bu 1 I__--...... ding f ·,--(Please read the precautions on the back before filling this page) Ministry of Economy Central Standards Bureau employee consumer cooperatives printed and printed --- itc This paper size is applicable to China -----Knowing (CNS) A4 size (210X297 mm) Printed by the Central Bureau of Consumer Affairs, Ministry of Economic Affairs A7 B7 5. DESCRIPTION OF THE INVENTION (丨)) Multiplying the device coordinates (the position of the receiver 3 with respect to the transmitter) by the inverse matrix of the receiver 30 with respect to the transmitter position, the device coordinates become the position of the receiver 3 with respect to the receiver. In this way, as long as the relative orientation of the receiver 30 and the patient's head is fixed, even if the transmitter moves for any reason, it will not affect the device coordinates. When this system is used in actual surgery, the relative orientation of the patient's head and the surgical instruments operated by the restaurant can be displayed in the image in a timely and accurate manner, and the surgical path planned by the physician before the operation is also displayed in the image. To make the surgery follow the planned surgical path, the physician must move the surgical instrument while looking at the image during the operation until the surgical instrument in the image coincides with the predetermined path. As shown in FIG. 10, in order to increase the convenience of the sauce chef to use the system, another embodiment of the system is to combine the positioning device with the robot arm 6, and attach a positioning device receiver 3 to the end of the robot arm 6. The probe 4 calculates the conversion relationship between the coordinates of the robot arm 6 and the coordinates of the device. After the azimuth calibration element is completed, the program automatically converts the path on the image into coordinates relative to the device, and then into coordinates relative to the robotic arm, so that the robotic arm can move to the predetermined surgical path orientation and be fixed at this orientation. In the past, during surgery, the physician only needs to follow the direction set by the robotic arm. Because of the frameless WU system of this material, it has the following advantages: No harm to the patient: The present invention uses a mark to stick to the patient's head. Markers replace the positioning frame and are therefore not harmful to the patient. This paper is suitable for Guancai County (CNS) A4 specification (21 () > < 297) (Please read the precautions on the back before filling this page) λ. Order A7 ------------ B7__: _ V. Description of the invention (丨 V) Able to interact with the image in real time during the operation: During the operation, the doctor can see through it, like knowing what organization ^ Surgery Week® has, and at the same time, you can see the eyes of the hand in the street. What kind of organization will be encountered when the position is continued, and how to avoid it, and how to avoid it, so that the safety of the operation can be improved. -Easy to use · Because Zhao Ji is small and there is no frame, it is very flexible to use and will not be stingy The above is only used to explain the preferred embodiments of the present invention, and is not intended to limit the present invention in any form, so that, if there is any relevant Any modification or change of the invention should still be included in the scope of the present invention. (Please read the precautions on the back before filling out this page.) Τν Printed by the Bayer Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs Paper size applies to China Home Standard (CNS) Eighty Four Secrets (210x297 Male Thin)