TWM526875U - Image processing apparatus - Google Patents
Image processing apparatus Download PDFInfo
- Publication number
- TWM526875U TWM526875U TW105206876U TW105206876U TWM526875U TW M526875 U TWM526875 U TW M526875U TW 105206876 U TW105206876 U TW 105206876U TW 105206876 U TW105206876 U TW 105206876U TW M526875 U TWM526875 U TW M526875U
- Authority
- TW
- Taiwan
- Prior art keywords
- breast
- image
- training
- breast image
- interest
- Prior art date
Links
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
Description
本新型是有關於一種影像處理裝置,且特別是關於一種***影像的影像處理裝置。The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus for a breast image.
乳腺癌(mammary carcinoma)是女性常見的惡性腫瘤之一,其主要癥狀包括***腫瘤(tumor)、異常分泌物或形狀變異等。提早篩檢出***的異常癥狀,將有助於盡早針對腫瘤進行治療,以降低癌細胞惡化或擴散等問題。諸如臨床或自我***檢測、活體組織檢查、***攝影術(mammography)、超音波(ultrasound)顯像或磁共振(magnetic resonance)顯像等篩檢方式已廣泛在臨床上使用或成為學術研究的重要議題。Mammary carcinoma is one of the common malignant tumors in women. The main symptoms include breast tumor, abnormal secretion or shape variation. Screening for abnormal symptoms of the breast early will help treat the tumor as early as possible to reduce the deterioration or spread of cancer cells. Screening methods such as clinical or self-breast testing, biopsy, mammography, ultrasound imaging or magnetic resonance imaging have been widely used clinically or become important for academic research. issue.
一般而言,不論是***攝影術(mammography)、超音波(ultrasound)顯像或磁共振(magnetic resonance)顯像等顯像技術,為了確保後續的***檢查範圍可以囊括整個***,每個***都需要由多個方向進行拍攝或掃描以取得多張***影像,而組成相對的三維立體影像更需要數百張的切片。當存在大量的***影像,特別是前述***影像分別由多種顯像技術來取得時,不管是透過人眼或者是使用電腦輔助偵測(Computer Aided Detection;CADe)系統來逐一地對每張影像上是否存在腫瘤、腫塊或鈣化點進行檢查,都是相當耗時且浪費人力資源或硬體資源。In general, whether it is mammography, ultrasound imaging or magnetic resonance imaging, in order to ensure that the subsequent breast examination range can cover the entire breast, each breast It is necessary to shoot or scan in multiple directions to obtain multiple breast images, and the corresponding three-dimensional images require hundreds of slices. When there are a large number of breast images, especially when the aforementioned breast images are respectively obtained by various imaging techniques, each image is imaged one by one through a human eye or using a Computer Aided Detection (CADe) system. The presence of tumors, lumps, or calcifications is a time consuming and waste of human or hardware resources.
本新型實施例提供影像處理裝置,可以有效地降低對一系列***影像進行查驗的時間並提升查驗的效能。The novel embodiment provides an image processing apparatus, which can effectively reduce the time for checking a series of breast images and improve the performance of the inspection.
本新型實施例另提供一種影像處理裝置。影像處理裝置包括儲存單元與處理單元。儲存單元儲存以X光由不同的拍攝方向所拍攝的一組第一***影像、以超音波所掃描的第二***影像以及記錄多個模組。處理單元耦接儲存單元,且存取並執行儲存單元所記錄的模組。前述模組包括影像輸入模組、第一選取模組、第一計算模組與第一判斷模組。影像輸入模組取得該組的第一***影像與第二***影像。第一選取模組於該組的第一***影像上,選取共同的第一關注區域。第一計算模組計算第一關注區域在每一第一***影像的距離參數與方位參數。第一判斷模組透過跨模態對位模型,基於前述距離參數與前述方位參數,判斷第一關注區域在第二***影像中的位置。The embodiment of the present invention further provides an image processing apparatus. The image processing device includes a storage unit and a processing unit. The storage unit stores a set of first breast images captured by X-rays in different shooting directions, a second breast image scanned by ultrasonic waves, and a plurality of modules recorded. The processing unit is coupled to the storage unit and accesses and executes the module recorded by the storage unit. The module includes an image input module, a first selection module, a first calculation module and a first determination module. The image input module obtains the first breast image and the second breast image of the group. The first selection module selects a common first region of interest on the first breast image of the group. The first computing module calculates a distance parameter and an azimuth parameter of each first breast image in the first region of interest. The first determining module passes the trans-modal alignment model, and determines the position of the first region of interest in the second breast image based on the distance parameter and the azimuth parameter.
基於上述,本新型實施例所提供影像處理裝置,基於第一關注區域在第一***影像上的各項距離參數與方位參數,透過跨模態對位模型來判斷第一關注區域在第二***影像中的位置。第一***影像與第二***影像例如是經由不同攝像技術所取得的***影像。換言之,前述影像處理裝置,針對選定的關注區域,能自動地於其他相關的***影像上進行自動對位,藉以大幅度地降低對每張***影像進行查驗所需花費的時間並提升查驗效能。Based on the above, the image processing apparatus provided by the present invention determines the first attention area in the second breast through the cross-modal alignment model based on the distance parameters and the azimuth parameters of the first region of interest on the first breast image. The position in the image. The first breast image and the second breast image are, for example, breast images obtained through different imaging techniques. In other words, the image processing device can automatically perform automatic alignment on other related breast images for the selected region of interest, thereby greatly reducing the time required for each breast image to be inspected and improving the inspection performance.
為讓本新型的上述特徵和優點能更明顯易懂,下文特舉實施例,並配合所附圖式作詳細說明如下。The above described features and advantages of the present invention will be more apparent from the following description.
本新型的部份實施例接下來將會配合附圖來詳細描述,以下的描述所引用的元件符號,當不同附圖出現相同的元件符號將視為相同或相似的元件。這些實施例只是本新型的一部份,並未揭示所有本新型的可實施方式。更確切的說,這些實施例只是本新型的專利申請範圍中的裝置的範例。The components of the present invention will be described in detail in the following description in conjunction with the drawings. These embodiments are only a part of the present invention and do not disclose all of the embodiments of the present invention. Rather, these embodiments are merely examples of devices in the scope of the present patent application.
於本新型的實施例中所提出之影像處理裝置,可以針對不同攝像技術所取得的***影像進行影像對位。具體而言,針對一組第一***影像上的第一關注區域,藉由前述影像處理裝置,能對應地取得第一關注區域在第二***影像上的位置。在本新型的一實施例中,該組的第一***影像例如是以X光由不同的拍攝方向所拍攝的***影像,而第二***影像例如是以超音波所掃描的***影像。The image processing device proposed in the embodiment of the present invention can perform image alignment on the breast images obtained by different imaging technologies. Specifically, for the first region of interest on the first set of first breast images, the position of the first region of interest on the second breast image can be correspondingly obtained by the image processing device. In an embodiment of the present invention, the first breast image of the group is, for example, a breast image taken by X-rays in different shooting directions, and the second breast image is, for example, a breast image scanned by ultrasonic waves.
更有甚者,於本新型的實施例中所提出之影像處理裝置,還可以針對同一攝像技術所取得的***影像進行影像對位。具體而言,針對第二***影像上的第二關注區域,藉由前述影像處理裝置,能對應地取得第二關注區域在目標***影像上的對應位置。在本新型的一實施例中,第二***影像與目標***影像例如為超音波在相同的掃描時間但不同的掃描方向進行掃描而取得,或者是例如為超音波在不同的掃描時間但相同的掃描方向進行掃描而取得。What is more, the image processing device proposed in the embodiment of the present invention can perform image alignment on the breast image obtained by the same imaging technology. Specifically, with respect to the second region of interest on the second breast image, the image processing device can correspondingly acquire the corresponding position of the second region of interest on the target breast image. In an embodiment of the present invention, the second breast image and the target breast image are obtained, for example, by scanning the ultrasonic waves at the same scanning time but different scanning directions, or for example, the ultrasonic waves are the same at different scanning times. Scanning direction is obtained by scanning.
圖1A為依據本新型一實施例所繪示的影像處理裝置的方塊示意圖。參照圖1A,影像處理裝置100至少包括處理單元120以及儲存單元140,並且處理單元120耦接至儲存單元140,但本新型不限於此。影像處理裝置100可以是伺服器、桌上型電腦、筆記型電腦、工作站、個人數位助理(Personal digital assistant, PDA)、平板個人電腦(Personal computer, PC)、電腦輔助偵測(CADe)系統等電子裝置,但不以此為限。影像處理裝置100例如是連接***X光攝影機與自動***超聲波系統(Automated breast ultrasound system, ABUS)。FIG. 1A is a block diagram of an image processing apparatus according to an embodiment of the invention. Referring to FIG. 1A, the image processing apparatus 100 includes at least a processing unit 120 and a storage unit 140, and the processing unit 120 is coupled to the storage unit 140, but the present invention is not limited thereto. The image processing apparatus 100 may be a server, a desktop computer, a notebook computer, a workstation, a personal digital assistant (PDA), a personal computer (PC), a computer aided detection (CADe) system, or the like. Electronic device, but not limited to this. The image processing apparatus 100 is connected, for example, to a mammography camera and an automated breast ultrasound system (ABUS).
於本新型的一實施例中,處理單元120例如是以中央處理單元(Central processing unit, CPU)、數位信號處理(Digital signal processing, DSP)晶片、場可程式化邏輯閘陣列(Field programmable gate array, FPGA)、微處理器、微控制器等可程式化單元來實施,但本新型不限於此。處理單元120亦可以獨立電子裝置或積體電路(Integrated circuit, IC)來實施。In an embodiment of the present invention, the processing unit 120 is, for example, a central processing unit (CPU), a digital signal processing (DSP) chip, or a field programmable gate array. , FPGA, microprocessor, microcontroller, etc. can be implemented by a programmable unit, but the present invention is not limited thereto. The processing unit 120 can also be implemented as an independent electronic device or an integrated circuit (IC).
於本新型的一實施例中,儲存單元140可以是任何型態的固定或可移動隨機存取記憶體(Random access memory, RAM)、唯讀記憶體(Read-only memory, ROM)、快閃記憶體(Flash memory)或類似元件或上述元件的組合。在本實施例中,儲存單元140儲存以不同攝像技術取得的***影像,例如是以X光所拍攝的第一***影像、第一訓練***影像以及以超音波所掃描的第二***影像、目標***影像、第二訓練***影像。另一方面,儲存單元140還儲存各項參數與影像特徵。In an embodiment of the present invention, the storage unit 140 can be any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash. Flash memory or similar component or a combination of the above components. In this embodiment, the storage unit 140 stores breast images acquired by different imaging technologies, such as a first breast image taken by X-rays, a first training breast image, and a second breast image scanned by ultrasound. Breast image, second training breast image. On the other hand, the storage unit 140 also stores various parameters and image features.
參照圖1A所示的影像處理裝置100,於本實施例中,儲存單元140記錄影像輸入模組142、第一選取模組144、第一計算模組146、第一判斷模組148與第一訓練模組150等模組,而前述模組可運作以針對不同攝像技術所取得的***影像進行影像對位。前述模組的詳細運作內容待稍後實施例詳細說明。然而,儲存單元140所記錄的模組並不限於此。Referring to the image processing apparatus 100 shown in FIG. 1A, in the embodiment, the storage unit 140 records the image input module 142, the first selection module 144, the first calculation module 146, the first determination module 148, and the first The modules are trained such as modules 150, and the modules are operable to perform image alignment for breast images obtained by different camera technologies. The detailed operation of the foregoing modules will be described in detail later in the embodiments. However, the module recorded by the storage unit 140 is not limited thereto.
圖1B為依據本新型另一實施例所繪示的影像處理裝置的方塊示意圖。參照圖1B所示的影像處理裝置100,於本實施例中,儲存單元140還記錄第二選取模組152、第二計算模組154、目標選取模組156、目標計算模組158、差異計算模組160、第二判斷模組162與第二訓練模組164等模組,而前述模組可運作以針對同一攝像技術所取得的***影像進行影像對位。前述模組的詳細運作內容待稍後實施例詳細說明。FIG. 1B is a block diagram of an image processing apparatus according to another embodiment of the present invention. Referring to the image processing apparatus 100 shown in FIG. 1B, in the embodiment, the storage unit 140 also records the second selection module 152, the second calculation module 154, the target selection module 156, the target calculation module 158, and the difference calculation. The modules 160, the second determining module 162 and the second training module 164 are operable, and the modules are operable to perform image alignment on the breast images obtained by the same camera technology. The detailed operation of the foregoing modules will be described in detail later in the embodiments.
值得注意的是,前述實施例中所述的儲存單元140並未限制是單一記憶體元件,上述之各模組亦可以分開儲存在兩個或兩個以上相同或不同型態之記憶體元件中。在本實施例中,前述模組以軟體形式存放於儲存單元140,並且由處理單元120存取並執行,但本新型不限於此。在本新型的其他實施例中,前述模組還例如是分別以特定的電路結構而實現。It should be noted that the storage unit 140 described in the foregoing embodiments is not limited to a single memory component, and the foregoing modules may be separately stored in two or more memory components of the same or different types. . In the embodiment, the foregoing module is stored in the storage unit 140 in the form of software, and is accessed and executed by the processing unit 120, but the present invention is not limited thereto. In other embodiments of the present invention, the aforementioned modules are also implemented, for example, in a specific circuit configuration.
為了方便理解本新型實施例的操作流程,以下將舉諸多實施例詳細說明本新型實施例中影像處理裝置100對***影像進行影像對位的流程。圖2為依據一實施例所繪示的***影像對位方法的流程圖。參照圖2,本實施例的方法適用於圖1A與圖1B中的影像處理裝置100,下文中,將搭配影像處理裝置100中的各項元件及模組說明本實施例所述之方法。In order to facilitate the understanding of the operation flow of the embodiment of the present invention, the flow of image alignment of the breast image by the image processing apparatus 100 in the present embodiment will be described in detail below. 2 is a flow chart of a breast image alignment method according to an embodiment. Referring to FIG. 2, the method of the present embodiment is applied to the image processing apparatus 100 of FIG. 1A and FIG. 1B. Hereinafter, the method described in this embodiment will be described in conjunction with various components and modules in the image processing apparatus 100.
於本實施例中,影像輸入模組142取得以X光由不同的拍攝方向所拍攝的一組第一***影像(步驟S110),並且取得以超音波所掃描的第二***影像(步驟S115)。In the embodiment, the image input module 142 acquires a set of first breast images captured by X-rays in different shooting directions (step S110), and acquires a second breast image scanned by ultrasonic waves (step S115). .
於本實施例中,第一***影像例如是由***X光攝影機對應***部位而在不同的拍攝方向所拍攝的***影像,而第二***影像例如是由自動***超聲波系統(ABUS)對應***部位所掃描的***影像。影像輸入模組142例如是透過有線通訊或無線通訊而從***X光攝影機與自動***超聲波系統(ABUS)取得第一***影像與第二***影像,或者影像輸入模組142由儲存單元140取得第一***影像與第二***影像。In the embodiment, the first breast image is, for example, a breast image captured by a mammography camera corresponding to a breast portion in different shooting directions, and the second breast image is, for example, an automatic breast ultrasound system (ABUS) corresponding to the breast portion. The scanned breast image. The image input module 142 obtains the first breast image and the second breast image from the mammography camera and the automatic breast ultrasound system (ABUS) through wired communication or wireless communication, for example, or the image input module 142 is obtained by the storage unit 140. A breast image and a second breast image.
具體而言,在本實施例中,一組第一***影像通常包括斜位向***影像以及頭腳向***影像。圖3為依據本新型一實施例所繪示的第一***影像的拍攝示意圖。參照圖3,斜位向***影像以及頭腳向***影像為***X光攝影機在不同拍攝方向(斜位向、頭腳向)且相同或相近的拍攝時間所分別取得的***影像。Specifically, in the present embodiment, a set of first breast images typically includes an oblique position to the breast image and a head and foot to the breast image. FIG. 3 is a schematic diagram of photographing a first breast image according to an embodiment of the invention. Referring to Fig. 3, the oblique image to the breast image and the head and foot to the breast image are respectively obtained by the mammography camera in different shooting directions (oblique direction, head and foot) and the same or similar shooting time.
參照圖2,於本實施例中,取得第一***影像後,第一選取模組144於該組的第一***影像上,選取共同的第一關注區域(步驟S120)。具體而言,第一選取模組144例如是基於電腦輔助偵測(CADe)系統對第一***影像的偵測結果而選定第一關注區域,並且第一關注區域可能是***部位中疑似存在腫瘤、腫塊或鈣化點的區域或位置,但本新型不限於此。在另一實施例中,第一選取模組144還例如是基於使用者以輸入介面在第一***影像上所指定的區位或位置而選定第一關注區域。值得注意的是,第一關注區域在斜位向***影像以及頭腳向***影像上會因兩者的拍攝方位而有呈現上的偏差。Referring to FIG. 2, in the embodiment, after the first breast image is acquired, the first selection module 144 selects a common first region of interest on the first breast image of the group (step S120). Specifically, the first selection module 144 selects a first region of interest based on, for example, a detection result of the first breast image by a computer-aided detection (CADe) system, and the first region of interest may be a suspected tumor in the breast region. The area or location of the lumps or calcifications, but the present invention is not limited thereto. In another embodiment, the first selection module 144 further selects the first region of interest based on, for example, a location or location designated by the user on the first breast image by the input interface. It is worth noting that the first region of interest has a bias in the orientation of the breast image and the head and foot to the breast image due to the orientation of the two.
參照圖2,於本實施例中,選取第一關注區域後,第一計算模組146計算第一關注區域在每一張第一***影像的距離參數與方位參數(步驟S125)。詳細而言,前述距離參數與方位參數是以第一***影像中的乳頭特徵、***皮膚作為參考基準而計算取得,並且可用於判斷第一關注區域在***部位的位置。Referring to FIG. 2, in the embodiment, after the first region of interest is selected, the first computing module 146 calculates a distance parameter and an orientation parameter of the first region of interest in each of the first breast images (step S125). In detail, the distance parameter and the azimuth parameter are calculated by using the nipple feature in the first breast image and the breast skin as a reference, and can be used to determine the position of the first region of interest at the breast site.
圖4為依據本新型一實施例所繪示的距離參數與方位參數的拍攝示意圖。具體而言,圖4以一張第一***影像作為範例來說明各項距離參數與方位參數。參照圖4,第一關注區域R與***皮膚在第一***影像上具有相距最近的一個鄰近點S。另一方面,第一關注區域R在第一***影像上的水平延伸線H與***皮膚相交於一個水平點I。在本實施例中,距離參數包括第一關注區域R與鄰近點S的距離RS、鄰近點S與乳頭特徵N沿***皮膚的弧線距離SN以及水平點I與乳頭特徵N沿***皮膚的弧線距離IN。相對來說,方位參數則為第一關注區域R在第一***影像相對於乳頭特徵N的方位角度。需要注意的是,計算前述距離參數以及方位參數時,例如是以第一關注區域R內的中心點r作為基準點來進行計算,但本新型不限於此。FIG. 4 is a schematic diagram of photographing distance parameters and azimuth parameters according to an embodiment of the invention. Specifically, FIG. 4 illustrates a distance parameter and an azimuth parameter by taking a first breast image as an example. Referring to FIG. 4, the first region of interest R and the breast skin have a closest neighboring point S on the first breast image. On the other hand, the horizontal extension line H of the first region of interest R on the first breast image intersects the breast skin at a horizontal point I. In this embodiment, the distance parameter includes a distance RS between the first region of interest R and the neighboring point S, an arc distance SN between the neighboring point S and the nipple feature N along the breast skin, and an arc distance between the horizontal point I and the nipple feature N along the breast skin. IN. Relatively speaking, the orientation parameter is the azimuthal angle of the first region of interest R relative to the nipple feature N of the first region of interest R. It is to be noted that, when the distance parameter and the azimuth parameter are calculated, for example, the center point r in the first region of interest R is used as a reference point for calculation, but the present invention is not limited thereto.
對於一組第一***影像中的斜位向***影像以及頭腳向***影像來說,第一計算模組146會分別計算距離參數以及方位參數。換言之,第一計算模組146會計算斜位向***影像的距離參數與方位參數MLO ori,而斜位向***影像的距離參數包括距離RS MLO、距離SN MLO以及距離IN MLO。另一方面,第一計算模組146會計算頭腳向***影像的距離參數與方位參數CC ori,而頭腳向***影像的距離參數包括距離RS CC、距離SN CC以及距離IN CC。 For a set of oblique images of the first breast image to the breast image and the head and foot to the breast image, the first calculation module 146 calculates the distance parameter and the orientation parameter, respectively. In other words, the first calculation module 146 calculates the distance parameter and the orientation parameter MLO ori of the oblique position to the breast image, and the distance parameter of the oblique position to the breast image includes the distance RS MLO , the distance SN MLO , and the distance IN MLO . On the other hand, the first calculation module 146 calculates the distance parameter of the head and foot to the breast image and the orientation parameter CC ori , and the distance parameter of the head and foot to the breast image includes the distance RS CC , the distance SN CC and the distance IN CC .
參照圖2,於本實施例中,取得距離參數與方位參數後,第一判斷模組148透過跨模態對位模型,基於距離參數與方位參數,判斷第一關注區域R在第二***影像中的位置(步驟S130)。詳細而言,在本實施例中,第一判斷模組148將一組第一***影像的距離參數與方位參數分別代入跨模態對位模型而取得第一關注區域R在第二***影像中的位置。Referring to FIG. 2, in the embodiment, after obtaining the distance parameter and the azimuth parameter, the first determining module 148 transmits the trans-modal alignment model, and determines the first region of interest R in the second breast image based on the distance parameter and the azimuth parameter. The position in (step S130). In detail, in the embodiment, the first determining module 148 substitutes the distance parameter and the azimuth parameter of the first breast image into the trans-modal alignment model to obtain the first region of interest R in the second breast image. s position.
在本新型的一實施例中,跨模態對位模型包括距離預測模型以及方位預測模型。針對前述的第一***影像,第一判斷模組148依據該組的第一***影像的距離參數,並且利用跨模態對位模型的距離預測模型,藉以計算第一關注區域R在第二***影像上相對於乳頭特徵N的距離M dist。詳細而言,距離預測模型包括一組第一權重α MLO、β MLO、γ MLO以及一組第二權重α CC、β CC、γ CC,並且以距離預測模型計算距離M dist的方式如下。 …(1) …(2) …(3) In an embodiment of the present invention, the cross-modal alignment model includes a distance prediction model and an azimuth prediction model. For the first breast image, the first determining module 148 calculates the first region of interest R in the second breast according to the distance parameter of the first breast image of the group and the distance prediction model of the cross-modal alignment model. The distance M dist from the nipple feature N on the image. In detail, the distance prediction model includes a set of first weights α MLO , β MLO , γ MLO , and a set of second weights α CC , β CC , γ CC , and the distance M dist is calculated by the distance prediction model as follows. …(1) …(2) ...(3)
另一方面,針對前述的第一***影像,第一判斷模組148依據該組的第一***影像的方位參數,並且利用跨模態對位模型的方位預測模型,藉以計算第一關注區域R在第二***影像上相對於乳頭特徵N的方位角度。方位預測模型包括一組第三權重μ MLO、ν CC,並且以方位預測模型計算方位角度M ori的方式如下。 …(4) On the other hand, for the first breast image, the first determining module 148 calculates the first region of interest R according to the orientation parameter of the first breast image of the group and the azimuth prediction model of the cross-modal alignment model. The azimuthal angle relative to the nipple feature N on the second breast image. The azimuth prediction model includes a set of third weights μ MLO , ν CC , and the azimuth prediction model calculates the azimuth angle M ori as follows. ...(4)
第一權重α MLO、β MLO、γ MLO、第二權重α CC、β CC、γ CC與第三權重μ MLO、ν CC的各個權重值是分別介於0與1之間,並且各組的權重值相加為1。藉由距離M dist以及方位角度M ori,影像處理裝置100可以進一步繪示第一關注區域R在第二***影像上相對於乳頭特徵N的位置。 The respective weights of the first weights α MLO , β MLO , γ MLO , the second weight α CC , β CC , γ CC and the third weight μ MLO , ν CC are between 0 and 1, respectively, and The weight values are added to 1. The image processing apparatus 100 can further illustrate the position of the first region of interest R on the second breast image relative to the nipple feature N by the distance M dist and the azimuth angle M ori .
圖5為依據本新型一實施例所繪示的第一關注區域在第二***影像上的示意圖。參照圖5,藉由距離M dist以及方位角度M ori,第一關注區域R在第二***影像上相對於乳頭特徵N的位置PR可以被取得。在圖5中,第二***影像為冠狀面的***影像。 FIG. 5 is a schematic diagram of a first region of interest on a second breast image according to an embodiment of the invention. Referring to FIG. 5, by the distance M dist and the azimuth angle M ori , the position PR of the first region of interest R relative to the nipple feature N on the second breast image can be obtained. In Figure 5, the second breast image is a coronal breast image.
然而,於本新型的一實施例中,在利用跨模態對位模型進行第一***影像與第二***影像間的影像對位前,第一訓練模組150更藉由多組第一訓練***影像來訓練跨模態對位模型內的距離預測模型與方位預測模型。以X光由不同的拍攝方向所拍攝的多組第一訓練***影像存放於儲存單元140內,並且由影像輸入模組142所取得。However, in an embodiment of the present invention, before the image alignment between the first breast image and the second breast image is performed by using the cross-modal alignment model, the first training module 150 further adopts multiple sets of first training. The breast image is used to train the distance prediction model and the azimuth prediction model in the cross-modal alignment model. The plurality of sets of first training breast images captured by X-rays in different shooting directions are stored in the storage unit 140 and acquired by the image input module 142.
每一組的第一訓練***影像分別包括斜位向訓練***影像與頭腳向訓練***影像,並且每一組的第一訓練***影像還分別對應一張對位訓練影像。對位訓練影像為以超音波所掃描的訓練***影像。每一組第一訓練***影像以及所對應的對位訓練影像具有共同的關注區域,並且前述關注區域在第一訓練***影像以及所對應的對位訓練影像上的位置皆為已知。The first training breast images of each group respectively include an oblique position training breast image and a head and foot training breast image, and each group of first training breast images respectively corresponding to one alignment training image. The alignment training image is a training breast image scanned by ultrasound. Each of the first training breast images and the corresponding alignment training images have a common region of interest, and the positions of the aforementioned regions of interest on the first training breast image and the corresponding alignment training images are known.
在本實施例中,第一訓練模組150以斜位向訓練***影像,訓練距離預測模型的第一權重α MLO、β MLO、γ MLO,以頭腳向訓練***影像,訓練距離預測模型的第二權重α CC、β CC、γ CC,並且以斜位向訓練***影像與頭腳向訓練***影像,訓練方位預測模型的第三權重μ MLO、ν CC。詳細而言,第一訓練模組150例如是邏輯迴歸(Logistic Regression)、支持向量機器(Support Vector Machine;SVM)、類神經網路(Neural network;NN)等方式訓練跨模態對位模型,但本新型不限於此。 In this embodiment, the first training module 150 trains the breast image in an oblique position, and the first weights α MLO , β MLO , and γ MLO of the distance prediction model are trained to train the breast image with the head and foot, and the distance prediction model is trained. The second weights α CC , β CC , γ CC , and training the breast image with the oblique position to the training breast image and the head and foot, training the third weights M M M , ν CC of the azimuth prediction model. In detail, the first training module 150 trains a cross-modal alignment model by, for example, Logistic Regression, Support Vector Machine (SVM), and Neural Network (NN). However, the present invention is not limited to this.
圖2以及前述相關的實施例,主要是針對不同攝像技術所取得的***影像進行影像對位,但本新型不限於此。圖6為依據另一實施例所繪示的***影像對位方法的流程圖。具體而言,圖6繪示了針對同一攝像技術所取得的***影像進行影像對位。參照圖6,本實施例的方法適用於圖1B中的影像處理裝置100,下文中,將搭配影像處理裝置100中的各項元件及模組說明本實施例所述之方法。2 and the related embodiments described above mainly perform image alignment for breast images obtained by different imaging technologies, but the present invention is not limited thereto. FIG. 6 is a flow chart of a breast image alignment method according to another embodiment. Specifically, FIG. 6 illustrates image alignment for a breast image acquired by the same imaging technique. Referring to FIG. 6, the method of the present embodiment is applied to the image processing apparatus 100 of FIG. 1B. Hereinafter, the method described in this embodiment will be described in conjunction with various components and modules in the image processing apparatus 100.
於本實施例中,影像輸入模組142取得以超音波所掃描的第二***影像(步驟S115),並且取得以超音波所掃描的目標***影像(步驟S135)。具體而言,第二***影像與目標***影像為超音波在相同或相近的掃描時間但不同的掃描方向進行掃描而取得。一般而言,由自動***超聲波系統(ABUS)對應***部位所掃描的***影像,依據掃描方向的不同而可以分別為冠狀面、矢狀面與橫狀面的***影像,但本新型不限於此。於其他實施例中,第二***影像與目標***影像為超音波在不同或不相近的掃描時間但相同的掃描方向進行掃描而取得。In the present embodiment, the image input module 142 acquires the second breast image scanned by the ultrasonic wave (step S115), and acquires the target breast image scanned by the ultrasonic wave (step S135). Specifically, the second breast image and the target breast image are acquired by scanning the ultrasonic waves in the same or similar scanning time but in different scanning directions. Generally, the breast images scanned by the breast area corresponding to the automatic breast ultrasound system (ABUS) may be coronal, sagittal, and transverse breast images depending on the scanning direction, but the present invention is not limited thereto. . In other embodiments, the second breast image and the target breast image are acquired by scanning the ultrasonic waves at different or different scanning times but in the same scanning direction.
參照圖6,於本實施例中,取得第二***影像後,第二選取模組152在第二***影像上選取第二關注區域(步驟S140)。類似於選取第一關注區域,第二選取模組152例如是基於電腦輔助偵測(CADe)系統對第二***影像的偵測結果或者是基於使用者以輸入介面在第二***影像上所指定的區位或位置而選定第二關注區域。第二關注區域可能是***部位中疑似存在腫瘤、腫塊或鈣化點的區域或位置,但本新型不限於此。Referring to FIG. 6, in the embodiment, after the second breast image is acquired, the second selection module 152 selects the second region of interest on the second breast image (step S140). Similar to the selection of the first region of interest, the second selection module 152 is, for example, based on a computer aided detection (CADe) system to detect the second breast image or based on the user's input interface on the second breast image. The second region of interest is selected for the location or location. The second region of interest may be a region or location in the breast site where a tumor, mass or calcification is suspected to exist, but the present invention is not limited thereto.
參照圖6,於本實施例中,在第二***影像選定第二關注區域後,第二計算模組154取得第二關注區域在第二***影像上的特徵參數(步驟S145)。在本新型的一實施例中,第二關注區域在第二***影像的特徵參數包括位置特徵(Location feature)、亮度特徵(Intensity feature)、型態學特徵(Morphology feature)以及紋理特徵(Texture feature)。Referring to FIG. 6, in the embodiment, after the second region of interest is selected in the second breast image, the second computing module 154 obtains the feature parameter of the second region of interest on the second breast image (step S145). In an embodiment of the present invention, the feature parameters of the second region of interest in the second breast image include a location feature, an intensity feature, a Morphology feature, and a texture feature. ).
於本新型的一實施例中,位置特徵(Location feature)包括第二關注區域在第二***影像上相對於乳頭特徵的距離、第二關注區域在第二***影像上相對於乳頭特徵的方位角度、複數個第二關注區域在第二***影像上的距離等,但本新型不限於此。亮度特徵(Intensity feature)包括第二關注區域內的亮度標準差、第二關注區域與鄰近區域的亮度差異等,但本新型不限於此。型態學特徵(Morphology feature)包括第二關注區域的本徵值向量、主要軸長、次要軸長等,但本新型不限於此。紋理特徵(Texture feature)包括以不同灰度共生矩陣(gray-level co-occurrence matrix;GLCM)對第二關注區域進行計算而得到的平均及標準差、能量、熵、相關度等,但本新型不限於此。In an embodiment of the present invention, the location feature includes a distance of the second region of interest on the second breast image relative to the nipple feature, and an orientation of the second region of interest on the second breast image relative to the nipple feature. The distance between the plurality of second attention areas on the second breast image, etc., but the present invention is not limited thereto. The intensity feature includes a luminance standard deviation in the second region of interest, a luminance difference between the second region of interest and the neighboring region, and the like, but the present invention is not limited thereto. The Morphology feature includes an eigenvalue vector of the second region of interest, a major axial length, a minor axial length, and the like, but the present invention is not limited thereto. The texture feature includes an average and standard deviation, energy, entropy, correlation, etc. obtained by calculating a second region of interest by using a gray-level co-occurrence matrix (GLCM), but the novel Not limited to this.
參照圖6,於本實施例中,取得前述的特徵參數後,目標選取模組156基於第二關注區域在第二***影像上相對於乳頭特徵的位置資訊,於目標***影像選定目標範圍(步驟S150)。一般而言,由自動***超聲波系統(ABUS)對應***部位所掃描的***影像都具有乳頭特徵。因此,在取得第二關注區域在第二***影像上的特徵參數後,目標選取模組156可基於第二關注區域在第二***影像上相對於乳頭特徵的位置資訊來於目標***影像選定目標範圍,並且前述的目標範圍內可能是第二關注區域在目標***影像上的對應位置。Referring to FIG. 6 , in the embodiment, after obtaining the foregoing characteristic parameters, the target selection module 156 selects a target range in the target breast image based on the position information of the second attention area on the second breast image relative to the nipple feature (steps). S150). In general, breast images scanned by the automatic breast ultrasound system (ABUS) corresponding to the breast site have nipple characteristics. Therefore, after obtaining the feature parameter of the second region of interest on the second breast image, the target selection module 156 can select the target for the target breast image based on the position information of the second region of interest on the second breast image relative to the nipple feature. The range, and the aforementioned target range may be the corresponding position of the second region of interest on the target breast image.
參照圖6,於本實施例中,目標計算模組158更取得目標範圍在目標***影像上的目標特徵參數(步驟S155)。目標特徵參數的種類與形式類似於前述實施例提及的第二關注區域在第二***影像的特徵參數,在此不再贅述。接著,差異計算模組160計算前述特徵參數與目標特徵參數的特徵差值(步驟S160)。具體而言,差異計算模組160計算第二關注區域在第二***影像的特徵參數與目標範圍在目標***影像的目標特徵參數間的特徵差值。值得注意的是,差異計算模組160是計算相同形式的特徵參數與目標特徵參數之間的特徵差值。換言之,基於特徵參數與目標特徵參數在形式上的數量,特徵差值的數量也會有所變化。Referring to FIG. 6, in the embodiment, the target calculation module 158 further obtains the target feature parameter of the target range on the target breast image (step S155). The type and form of the target feature parameter are similar to the feature parameters of the second region of interest mentioned in the foregoing embodiment, and are not described herein again. Next, the difference calculation module 160 calculates a feature difference value between the aforementioned feature parameter and the target feature parameter (step S160). Specifically, the difference calculation module 160 calculates a feature difference between the feature parameter of the second breast region and the target feature of the second breast region in the target image of the target breast image. It should be noted that the difference calculation module 160 calculates the feature difference between the feature parameters of the same form and the target feature parameters. In other words, based on the number of formal parameters and target feature parameters, the number of feature differences will also vary.
參照圖6,於本實施例中,取得特徵差值後,第二判斷模組162透過同模態對位模型,基於前述特徵差值,判斷目標範圍是否為第二關注區域在目標***影像上的對應位置(步驟S165)。具體而言,同模態對位模型是基於第二關注區域與目標範圍在各項特徵參數與目標特徵參數間的特徵差值來判斷目標範圍是否為第二關注區域在目標***影像上的對應位置。因此,在使用同模態對位模型進行判斷前,第二訓練模組164更藉由多張的第二訓練***影像來訓練同模態對位模型。Referring to FIG. 6, in the embodiment, after obtaining the feature difference value, the second determining module 162 transmits the same modality alignment model, and determines whether the target range is the second attention area on the target breast image based on the feature difference value. Corresponding position (step S165). Specifically, the same modality alignment model is based on the difference between the feature parameters of the second region of interest and the target range between the feature parameters and the target feature parameter to determine whether the target range is the correspondence of the second region of interest on the target breast image. position. Therefore, before using the same modality alignment model to judge, the second training module 164 further trains the same modality alignment model by using a plurality of second training breast images.
以超音波所掃描的多張第二訓練***影像例如是存放於儲存單元140,並且由影像輸入模組142所取得。對於多張第二訓練***影像,第二訓練模組164以第二訓練***影像中的多個匹配影像對,計算多組匹配特徵差值,然而再以該些匹配特徵差值,訓練同模態對位模型。在本新型的一實施例中,匹配影像對為已知且對位成功的多張訓練***影像,或者是訓練***影像上已知且對位成功的關注區域。需要注意的是,匹配影像對內部的多張訓練***影像為超音波在相同或相近的掃描時間但不同的掃描方向進行掃描而取得的訓練***影像,或者是超音波在相同的掃描方向但不同或不相近的掃描時間進行掃描而取得的訓練***影像。The plurality of second training breast images scanned by the ultrasonic waves are stored in the storage unit 140, for example, and are acquired by the image input module 142. For a plurality of second training breast images, the second training module 164 calculates a plurality of sets of matching feature differences by using a plurality of matching image pairs in the second training breast image, and then training the same model with the matching feature differences. State alignment model. In an embodiment of the present invention, the matching image pair is a plurality of training breast images that are known and successfully aligned, or an area of interest that is known and successfully aligned on the breast image. It should be noted that the plurality of training breast images in the matching image are the training breast images obtained by scanning the ultrasonic waves in the same or similar scanning time but different scanning directions, or the ultrasonic waves are different in the same scanning direction. Training breast images obtained by scanning with different scanning times.
基於前述匹配特徵值,第二訓練模組164例如是邏輯迴歸(Logistic Regression)、支持向量機器(Support Vector Machine;SVM)、類神經網路(Neural network;NN)等方式訓練同模態對位模型,但本新型不限於此。Based on the foregoing matching feature values, the second training module 164 trains the same modal alignment by, for example, Logistic Regression, Support Vector Machine (SVM), and Neural Network (NN). Model, but the present invention is not limited to this.
第二判斷模組164對同模態對位模型輸入所取得的特徵差值後,即可依據同模態對位模型的輸出結果判斷目標範圍是否為第二關注區域在目標***影像上的對應位置。After the second judgment module 164 inputs the obtained feature difference value to the same modality alignment model, it can determine whether the target range is the correspondence of the second attention area on the target breast image according to the output result of the same modality aligning model. position.
需要注意的是,當第二***影像與目標***影像為超音波在相同或相近的掃描時間但不同的掃描方向進行掃描而取得時,若目標範圍為第二關注區域在目標***影像上的對應位置,則代表目標範圍與第二關注區域為重覆掃描的部分。相對而言,當第二***影像與目標***影像為超音波在不同或不相近的掃描時間但相同的掃描方向進行掃描而取得時,若目標範圍經由同模態對位模型判斷為第二關注區域在目標***影像上的對應位置,可代表***部位的該區域並無顯著變化。It should be noted that when the second breast image and the target breast image are obtained by scanning the ultrasonic waves at the same or similar scanning time but different scanning directions, if the target range is the correspondence of the second attention area on the target breast image. The position represents the target range and the second area of interest is the portion of the repeated scan. In contrast, when the second breast image and the target breast image are obtained by scanning the ultrasonic waves at different or different scanning times but in the same scanning direction, if the target range is determined as the second attention via the same modal registration model. The corresponding position of the area on the target breast image may represent no significant change in the area of the breast area.
圖7為依據本新型一實施例所繪示的第二關注區域在目標***影像上的對應位置的示意圖。藉由圖6所示的***影像對位方法,對於在不同掃描方向上取得的***影像,例如是冠狀面的***影像、橫狀面的***影像、矢狀面的***影像,皆可取得第二關注區域的對應位置PR。值得注意的是,圖2與圖6所分別繪示的***影像對位方法可以合併使用,藉以分別在不同攝像技術取得的多張***影像以及相同攝像技術取得的多張***影像中,同時或連續地進行影像對位。此時,第一關注區域與第二關注區域可能是相同的關注區域。FIG. 7 is a schematic diagram of a corresponding position of a second region of interest on a target breast image according to an embodiment of the invention. According to the breast image aligning method shown in FIG. 6, for the breast images obtained in different scanning directions, for example, a coronal breast image, a horizontal breast image, and a sagittal breast image, Second, the corresponding position PR of the area of interest. It should be noted that the breast image alignment methods illustrated in FIG. 2 and FIG. 6 can be used in combination, so that multiple breast images obtained by different camera technologies and multiple breast images obtained by the same camera technology can be simultaneously or simultaneously. Image alignment is performed continuously. At this time, the first attention area and the second attention area may be the same attention area.
綜上所述,本新型實施例所提供的影像處理裝置,基於第一關注區域在第一***影像上的各項距離參數與方位參數,透過跨模態對位模型來判斷第一關注區域在第二***影像中的位置。第一***影像與第二***影像例如是經由不同攝像技術所取得的***影像。另一方面,本新型實施例還提供影像處理裝置,用以在相同攝像技術所取得的***影像上進行影像對位。換言之,前述影像處理裝置,針對選定的關注區域,能自動地於其他相關的***影像上進行自動對位,藉以大幅度地降低對每張***影像進行查驗所需花費的時間並提升查驗效能。In summary, the image processing apparatus provided by the new embodiment determines the first region of interest through the cross-modal alignment model based on the distance parameters and the azimuth parameters of the first region of interest on the first breast image. The position in the second breast image. The first breast image and the second breast image are, for example, breast images obtained through different imaging techniques. In another aspect, the present invention further provides an image processing apparatus for performing image alignment on a breast image obtained by the same imaging technique. In other words, the image processing device can automatically perform automatic alignment on other related breast images for the selected region of interest, thereby greatly reducing the time required for each breast image to be inspected and improving the inspection performance.
雖然本新型已以實施例揭露如上,然其並非用以限定本新型,任何所屬技術領域中具有通常知識者,在不脫離本新型的精神和範圍內,當可作些許的更動與潤飾,故本新型的保護範圍當視後附的申請專利範圍所界定者為準。Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention, and any person skilled in the art can make some changes and refinements without departing from the spirit and scope of the present invention. The scope of the present invention is defined by the scope of the appended claims.
100‧‧‧影像處理裝置
120‧‧‧處理單元
140‧‧‧儲存單元
142‧‧‧影像輸入模組
144‧‧‧第一選取模組
146‧‧‧第一計算模組
148‧‧‧第一判斷模組
150‧‧‧第一訓練模組
152‧‧‧第二選取模組
154‧‧‧第二計算模組
156‧‧‧目標選取模組
158‧‧‧目標計算模組
160‧‧‧差異計算模組
162‧‧‧第二判斷模組
164‧‧‧第二訓練模組
R‧‧‧第一關注區域
H‧‧‧水平延伸線
r‧‧‧中心點
S‧‧‧鄰近點
N‧‧‧乳頭特徵
I‧‧‧水平點
Mdist‧‧‧距離
Mori‧‧‧方位角度
PR‧‧‧關注區域的對應位置
S110、S115、S120、S125、S130‧‧‧***影像對位方法的步驟
S115、S135、S140、S145、S150、S155、S160、S165‧‧‧***影像對位方法的步驟100‧‧‧Image processing device
120‧‧‧Processing unit
140‧‧‧ storage unit
142‧‧‧Image input module
144‧‧‧First selection module
146‧‧‧First Computing Module
148‧‧‧First Judgment Module
150‧‧‧First training module
152‧‧‧Second selection module
154‧‧‧Second calculation module
156‧‧‧Target selection module
158‧‧‧Target Computing Module
160‧‧‧Differential Computing Module
162‧‧‧Second judgment module
164‧‧‧Second training module
R‧‧‧First Area of Interest
H‧‧‧ horizontal extension line
R‧‧‧ center point
S‧‧‧ neighboring points
N‧‧‧ nipple features
I‧‧‧ horizontal point
M dist ‧‧‧distance
M ori ‧‧‧ azimuth angle
PR‧‧‧Record of the area of interest
Steps for S110, S115, S120, S125, S130‧‧‧ breast image alignment methods
Steps of S115, S135, S140, S145, S150, S155, S160, S165‧‧‧ breast image alignment method
圖1A為依據本新型一實施例所繪示的影像處理裝置的方塊示意圖。 圖1B為依據本新型另一實施例所繪示的影像處理裝置的方塊示意圖。 圖2為依據一實施例所繪示的***影像對位方法的流程圖。 圖3為依據本新型一實施例所繪示的第一***影像的拍攝示意圖。 圖4為依據本新型一實施例所繪示的距離參數與方位參數的拍攝示意圖。 圖5為依據本新型一實施例所繪示的第一關注區域在第二***影像上的示意圖。 圖6為依據另一實施例所繪示的***影像對位方法的流程圖。 圖7為依據本新型一實施例所繪示的第二關注區域在目標***影像上的對應位置的示意圖。FIG. 1A is a block diagram of an image processing apparatus according to an embodiment of the invention. FIG. 1B is a block diagram of an image processing apparatus according to another embodiment of the present invention. 2 is a flow chart of a breast image alignment method according to an embodiment. FIG. 3 is a schematic diagram of photographing a first breast image according to an embodiment of the invention. FIG. 4 is a schematic diagram of photographing distance parameters and azimuth parameters according to an embodiment of the invention. FIG. 5 is a schematic diagram of a first region of interest on a second breast image according to an embodiment of the invention. FIG. 6 is a flow chart of a breast image alignment method according to another embodiment. FIG. 7 is a schematic diagram of a corresponding position of a second region of interest on a target breast image according to an embodiment of the invention.
100‧‧‧影像處理裝置 100‧‧‧Image processing device
120‧‧‧處理單元 120‧‧‧Processing unit
140‧‧‧儲存單元 140‧‧‧ storage unit
142‧‧‧影像輸入模組 142‧‧‧Image input module
144‧‧‧第一選取模組 144‧‧‧First selection module
146‧‧‧第一計算模組 146‧‧‧First Computing Module
148‧‧‧第一判斷模組 148‧‧‧First Judgment Module
150‧‧‧第一訓練模組 150‧‧‧First training module
Claims (7)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW105206876U TWM526875U (en) | 2016-05-12 | 2016-05-12 | Image processing apparatus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW105206876U TWM526875U (en) | 2016-05-12 | 2016-05-12 | Image processing apparatus |
Publications (1)
Publication Number | Publication Date |
---|---|
TWM526875U true TWM526875U (en) | 2016-08-11 |
Family
ID=57182878
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW105206876U TWM526875U (en) | 2016-05-12 | 2016-05-12 | Image processing apparatus |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWM526875U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI609674B (en) * | 2016-05-12 | 2018-01-01 | 太豪生醫股份有限公司 | Breast image matching method and image processing apparatus |
-
2016
- 2016-05-12 TW TW105206876U patent/TWM526875U/en not_active IP Right Cessation
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI609674B (en) * | 2016-05-12 | 2018-01-01 | 太豪生醫股份有限公司 | Breast image matching method and image processing apparatus |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107440730B (en) | Breast image alignment method and image processing device | |
US11551361B2 (en) | Method and system of computer-aided detection using multiple images from different views of a region of interest to improve detection accuracy | |
CN107480677B (en) | Method and device for identifying interest region in three-dimensional CT image | |
US8355552B2 (en) | Automated determination of lymph nodes in scanned images | |
EP3021753B1 (en) | Systems and methods for determining hepatic function from liver scans | |
WO2021021329A1 (en) | System and method for interpretation of multiple medical images using deep learning | |
KR102202617B1 (en) | Method and apparatus for analyzing abdominal disease based on medical image | |
WO2024067527A1 (en) | Hip joint angle measurement system and method | |
TWI839758B (en) | Processing method of medical image and computing apparatus for processing medical image | |
US20200253580A1 (en) | Tissue lesion detection and determination using quantitative transmission ultrasound | |
CN103455821B (en) | Image analysis apparatus and method based on BI-RADS | |
CN112568933B (en) | Ultrasonic imaging method, apparatus and storage medium | |
TWI574671B (en) | Analysis method for breast image and electronic apparatus thereof | |
CN111481233B (en) | Thickness measuring method for transparent layer of fetal cervical item | |
CN112529900A (en) | Method, device, terminal and storage medium for matching ROI in mammary gland image | |
TWM526875U (en) | Image processing apparatus | |
CN111275617A (en) | Automatic splicing method and system for ABUS breast ultrasound panorama and storage medium | |
CN112515705B (en) | Method and system for projection profile enabled computer-aided detection | |
WO2023133929A1 (en) | Ultrasound-based human tissue symmetry detection and analysis method | |
KR102332472B1 (en) | Tumor automatic segmentation using deep learning based on dual window setting in a medical image | |
Michahial et al. | A novel algorithm to select a seed point automatically in breast ultrasound image | |
Mahmood et al. | Ovary ultrasound image edge detection analysis: a tutorial using MATLAB | |
RU2812866C1 (en) | Method for processing computer tomography images (ct images) | |
Rafael-Palou et al. | Pulmonary nodule malignancy classification using its temporal evolution with two-stream 3D convolutional neural networks | |
TWI845365B (en) | Method for establishing model for identifying hormone receptor status, and method and system for identifying hormone receptor status |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
MM4K | Annulment or lapse of a utility model due to non-payment of fees |