This disclosure relates to imaging tools, and more particularly to a system and method for mapping internal passages to maintain spatial orientation and direction during navigation.
Endoscopy is a minimally invasive real-time imaging modality in which a camera is inserted into the body for visual inspection of internal structures such as the lung airways or the gastrointestinal system. Typically, the endoscope is a long flexible fiber-optic system connected to a light source at a proximal end outside of a patient's body and a lens at a distal end inside the patient's body. In addition, some endoscopes include a working channel through which the operator can perform suction or pass instruments such as brushes, biopsy needles or forceps. Video feedback gives a physician or technician cues to maneuver the scope to a targeted region.
Referring to
Image guided endoscopy, as compared to conventional endoscopy, enjoys the advantage of its real-time connection to a three-dimensional (3D) roadmap of the lung by fusing pre-operative computed tomography (CT) images with video data. While the interventional procedure is performed, physicians can determine where the scope is located with respect to the 3D CT space. In the research of bronchoscope localization, there are three types of ways to track the tip of the endoscope. Type (a) tracks based on a position sensor mounted to the tip of the endoscope; Type (b) tracks based on live image registration, and Type (c) is a combination of types (a) and (b) two.
Electro-magnetic (EM) guided endoscopy (type (a) system) has been recognized as a valuable tool for many lung applications, but it requires employing a supplemental guidance device. Image-registration based endoscopy (type (b) system), requires constant real-time frame-by-frame registration which can be time consuming, and prone to errors when fluids inside the airway obscure the video images. All of these systems, however, despite utilizing EM tracking or image-registration based tracking, demand a fast and powerful computer workstation (equipped with fine-resolution CT data) that is enabled to execute a multitude of non-trivial tasks, such as bronchus segmentation, image registration, path planning and real-time navigation. This technological integration, particularly with the fine resolution pre-operative CT images, poses an enormous challenge to many remote, less resourceful regions (particularly in developing countries) where hospitals have limited access to advanced technology while lung cancer occurrence in these regions may be extraordinarily high.
In accordance with the present principles, given that an obstacle in most bronchoscopy procedures resides in that the physicians lose spatial orientation in highly convoluted airways, a novel solution incorporates a video-based navigation method to a bronchoscopy suite. Instead of tracking the entire course of scope trajectory, directions are provided when the scope reaches branching intersections by analyzing video sequences. In this way, cues can be provided in the video images as to which way to go to reach a target or to indicate the current position of the tip of the scope. By analyzing motion fields of the video sequences, the system is able to label the branches of the airways or other branched cavities. The present solution is very cost-effective and does not need pre-operative CT images to be reconstructed as the roadmap, nor additional position tracking facilities (such as electro-magnetic (EM) tracking). Thus, this versatile solution can be applied to almost all pulmonology clinics, especially where access to advanced technology is limited. This guidance technology is particularly useful to pulmonology physicians, and more particularly to physicians in less-developed areas or countries. The present embodiments reduce or eliminate the need to purchase additional guidance devices or computer workstations to perform the navigation tasks.
A system and method for locating a position of an imaging device includes a guided imaging device configured to return images of internal passageways to a display. A processing module is configured to recognize patterns from the images and employ image changes to determine motion undergone by the imaging device such that a position of the imaging device is determined solely from information received from images obtained internally in the passageways and general knowledge of the passageways.
Another system for locating a distal end of an endoscope includes an illuminated endoscope tip mounted on a cable and configured to receive reflected light signals. A display is configured to render images received from the tip. A processing module is configured to recognize patterns from the images and employ image changes to determine direction choices and motion undergone by the tip. A general anatomical reference cross-references recognized patterns and image changes to the anatomical reference, wherein the position of the tip is determined relative to features deciphered from recognized patterns and image changes and the anatomical reference.
A method for locating a distal end of an endoscope includes illuminating an area around an endoscope tip, receiving reflected light through the tip, rendering images received from the tip, recognizing patterns from the images and employing image changes to determine motion undergone by the tip, and cross-referencing recognized patterns and image changes against a general anatomical reference, wherein the position of the tip is determined relative to features deciphered from the images and the anatomical reference.
These and other objects, features and advantages of the present disclosure will become apparent from the following detailed description of illustrative embodiments thereof, which is to be read in connection with the accompanying drawings.
This disclosure will present in detail the following description of preferred embodiments with reference to the following figures wherein:
The present disclosure describes an apparatus and method for scope navigation and imaging. The present principles analyze motion fields of scope video sequences to identify and label branches. In particularly useful embodiments, the scope may include a bronchoscope or any scope for pulmonary, digestive system, or other minimally invasive surgical viewing. In other embodiments, an endoscope or the like is employed for other medical procedures as well. These procedures may include minimally invasive endoscopic pituitary surgery, endoscopic skull base tumor surgery, intraventricular neurosurgery, arthroscopic surgery, laparoscopic surgery, etc. In other embodiments, the scope may be configured for viewing internal plumbing, pipe systems or for scoping animal or insect burrows. Other scoping applications are also contemplated. The present principles include components which (1) recognize patterns to identify bifurcations (or trifurcations, etc.) in video images, (2) use video motion detection to detect motion of the scope and the direction(s) of each turn, (3) using a rule-based technique to trigger a pre-defined knowledge base that can be derived from the anatomical imaging data and (4) using the 3D topology of known anatomy of the examined structures to determine where the scope is located in three dimensions after the scope makes a sequence of turns. Branches may be labeled dynamically on the display screen of the scope. The present embodiments are cost-effective for a plurality of reasons, e.g., pre-operative CT images are not needed to be reconstructed as a roadmap and position tracking facilities (such as EM tracking) are not needed.
Radial motion field vectors are employed to designate camera movement decisions (e.g., the viewing camera moves away from the scene—the vectors converge, and the viewing camera moves toward the scene—the vectors diverge). The motion fields (2D vector fields of velocities of the image feature points) are preferably employed to show the viewing camera is making different movements. When a turning translation (parallel translation) motion is discovered, a corresponding branch can be labeled accordingly on a display. The methods described herein can be built into a video-processor of an endoscope without the need for a powerful computer workstation (to perform air-way extraction, volume rendering and registration, etc.). This tracking technology would then be available where the cost of the workstation cannot be justified (e.g., at a rural pulmonology clinic). The methods described herein may also be implemented on a computer or in a custom designed apparatus.
It should be understood that the present invention will be described in terms of a bronchoscope; however, the teachings of the present invention are much broader and are applicable to any optical scope that can be employed in internal viewing of branching, curved, coiled or other shaped systems (e.g., digestive systems, circulatory systems, piping systems, animal or insect passages, mines, caverns, etc.). Embodiments described herein are preferably displayed for viewing on a display monitor. Such monitors may include any suitable display device including but not limited to handheld displays (e.g., on personal digital assistants, telephone devices, etc.), computer displays, televisions, designated monitors, etc. Depending of the scope, the display may be provided as part of the system or may be a separate unit or device.
It should also be understood that the optical scopes may include a plurality of different devices connected to or associated with the scope. Such devices may include a light, a cutting device, a brush, a vacuum, a camera, etc. These components may be formed integrally with a head on a distal end portion of the scope. The optical scopes may include a camera disposed at a tip of the scope or a camera may be disposed at the end of an optical cable opposite the tip. Embodiments may include hardware elements, software elements or both hardware and software elements. In a preferred embodiment, the present invention is implemented with software, which includes but is not limited to firmware, resident software, microcode, etc.
Furthermore, the present principles can take the form of a computer program product accessible from a computer-usable or computer-readable medium providing program code for use by or in connection with a computer or any instruction execution system. A computer-usable or computer readable medium can be any apparatus that may include, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The medium can be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system (or apparatus or device). Examples of a computer-readable medium include a semiconductor or solid state memory, magnetic tape, a removable computer diskette, a random access memory (RAM), a read-only memory (ROM), a rigid magnetic disk and an optical disk. Current examples of optical disks include compact disk—read only memory (CD-ROM), compact disk—read/write (CD-R/W) and DVD.
A data processing system suitable for storing and/or executing program code may include at least one processor coupled directly or indirectly to memory elements through a system bus. The processor or processing system may be provided with the scope system or provided independently of the scope system. The memory elements can include local memory employed during actual execution of the program code, bulk storage, and cache memories which provide temporary storage of at least some program code to reduce the number of times code is retrieved from bulk storage during execution. Input/output or I/O devices (including but not limited to keyboards, displays, pointing devices, etc.) may be coupled to the system either directly or through intervening I/O controllers.
Network adapters may also be coupled to the system to enable the data processing system to become coupled to other data processing systems or remote printers or storage devices through intervening private or public networks. Modems, cable modem and Ethernet cards are just a few of the currently available types of network adapters.
Referring now to the drawings in which like numerals represent the same or similar elements and initially to
Light reflected 111 from walls of internal tissues 112 is detected and propagated down the cable 104 as optical (or electrical) signals. The signals are interpreted preferably using a processing device 114, such as a computer or other platform configured with a photosensing device 116 in the case of a distally disposed camera. Photosensing device 116 may be mounted on a printed circuit board, be included in a camera device (e.g., a CCD camera) or be integrated in an integrated circuit chip. Many configurations and implementations may be employed to decipher and interpret the optical signals. If the camera is included in the tip 106, the signals are converted to electrical signals and interpreted by the processing device without photosensing device 116.
Processing device 114 may include a computer device, processor or controller configured to implement a program or programs 120. The program 120 includes instructions for interpreting and executing functions in accordance with the present principles. The program 120 may dynamically label branches, such as bronchial branches 122, where the scope tip 106 is currently located. The labeling process is an inexpensive alternative to perform navigation guidance for procedures such as a bronchoscopy procedure.
The processing device 114 provides dynamic labeling of airway branches 122 into an existing screen or display 124 of scope 102. No additional external monitor or work station is needed. By analyzing the video streams' motion patterns, the processing device 114 determines where the tip 106 of scope 102 is located, e.g., in the left primary bronchus or the right tertiary bronchus. No external tracking instruments are needed. The registration to high resolution pre-op CT images can also be omitted.
Features of the program 120 include a pattern recognition program 123 to identify bifurcations in video images. A motion detection program 125 is also used to detect if the scope is making a turn, and if so, which direction the scope takes. A general reference (e.g., an anatomical reference) 126 is also stored in memory 130. The general anatomical reference 126 stores prior knowledge about airway anatomy (as generic information, as opposed to CT scans or other imaging scans). This airway anatomy can be presented in the form of a set of rules or a 3D topology map. According to different designs, a rule-based technique or a model-based geographic matching algorithm can be used to determine where the scope is located after the scope makes a sequence of turns. It should be noted that a prior understanding on the particular patient is not needed and the rule or model may be used for all patients, hence generic information.
The rule-based technique uses features identified through pattern recognition to provide a connected path of previously traversed portions of the passageway. In other words, the present principles employ milestones or identify features in the passageway to help determine where the scope is located. For example, each bifurcation is pattern recognized followed by a determination of which bifurcation was selected to go down. This information will determine the current location. This process continues so that the location of the endoscope is known throughout the process. Rules such as a sequence of directions (e.g., left, right, left) may be employed to identify a present position of the tip 106.
Another approach may employ topology mapping and comparison to an atlas of lung airway anatomy. Based on the real-time motion analysis, it is possible to establish the topology (the qualitative shape) of the airways traversed by the endoscope using the camera's internal parameters. Until the tertiary bronchi, the topology is largely conserved across subjects, such that a standard topology can be described, with each segment of the topology named according to the typical conventions of pulmonologists. Based on the standard topology from the atlas and the observed topology of the airways traversed by the endoscope, the current location of the endoscope can be described relative to the atlas, and then the atlas naming convention is used to identify the current airway segment.
The scope 102 may include its own video-processor or the video-processor may be part of the processing device 114. The components built into the video-processor of the endoscope employ the signals to detect patterns in the images and then use the patterns to identify a position in the system or body. The endoscope monitor 124 will display not only the current video feedback, but also, preferably, the labeling information of each branch where the scope is located. Pattern recognition 123 identifies the bifurcation of the passage. Due to the nature of illumination in the endoscope system 100, the further (deeper) objects are located, the less they are illuminated. Thus, in the lungs, two bronchial sub-branches present less illuminated images in the video than the main branch from which they originated.
Due to the nature of design, after multiple trips within the airway tunnels, the present approach may disorientate the endoscope if initialization parameters are not correctly chosen. Thus, we propose using, e.g.: a) a local initialization method to start tracking when the bifurcations are seen in the video image, and/or b) a global initialization method where the length of endoscope that is inside the human body is taken into consideration. In the latter case, this depth information is recorded as a geographic parameter to constrain the possible location (or location range) of the tip of the endoscope. Thus, by knowing if the scope has reached the peripheral region or is still in the central airway, one can obtain better initialization parameters.
In
A real time motion analysis method 125 is stored in memory 130 and is employed to analyze images to determine a position or change in position. The method 125 can compare a current image map to a previous image map to determine direction, velocity, rotation, translation and other parameters. The motion analysis method 125 can use features in the image to track these parameters. Two sub-problems of motion analysis include 1) correspondence of elements: that is which elements of a frame correspond to which elements of a next frame of the sequence; and 2) reconstruction of motion: that is given a number of corresponding elements, what can be understood about the 3-D motion of the observed world.
In one embodiment, a Scale Invariant Feature Transform (SIFT) is employed to identify image features for scene recognition and tracking. Using SIFT, image features are invariant to image scaling and rotation, and partially invariant to change in illumination and 3D camera viewpoint. Other motion detection methods may also be employed such as optical flow methods, etc.
Based on 2D motion fields of sparse image features computed over time, a motion of the camera can be determined by tracking changes to the image based on one or more reference points (e.g., a predefined point with known absolute coordinates in 3D space). According to one or more reference points which show absolute location and orientation in 3D space, a program will be able to determine if the scope is making a left turn or right turn, up or down and thus label the branch-to-be-entered correspondingly.
In
Referring again to
In the illustrative embodiment, the scope preferably uses the knowledge of lung anatomy to name the branch where the scope is currently located. This may include a coordinate map 140 of anatomical data 126. The data in the map 140 may include ranges of dimensions for internal organs or features, include adjustments for individuals based on e.g., age, gender, surgical history, ethnicity, etc. The map 140 provides a reference against which images may be compared or features deciphered to be capable of identifying milestones, targets, abnormalities, etc. Since no pre-op CT roadmap is used for guidance, a set of rules, or an atlas based approach may be employed to determine the spatial location of the scope based on the sequence of turns it makes and gross anatomy of lung airways. For example, a rule specifies that after the scope makes a left turn followed by another right turn, it is now located in a left secondary bronchus.
In one embodiment, depending on the circumstances, a patient's internal configuration may be mapped out in a preliminary procedure by inserting the scope of the present system into the patient and recording and cataloging the images as the scope moves through the patient. In this way, a record of the condition and features can be collected and stored. This method provides the most accurate location detection since the actual images are employed in the mapping and labeling. This is particularly useful when a particular patient undergoes or will undergo multiple procedures. For example, if a technician finds a lesion in a lung during a first procedure, stored data may be employed to assist in guiding the technician back to that location. In this way, instead of labeling a current position, the technician is provided with internal directions on how to achieve a particular position. It should be understood that video images of entire procedures may be stored to provide a motion video of the procedure.
The present principles can be applied in pulmonology procedures, digestive procedures, or any other procedure where an endoscope or other camera device needs to be tracked. The present principles are particularly useful where access to advanced technology (such as powerful computers, position tracking devices, external monitors) is limited. The system is very cost-effective and does not require high-resolution pre-operative CT images to be reconstructed as the roadmap.
Referring to
In block 310, recognized patterns and image changes are cross-referenced against a general anatomical reference. The position of the tip is determined relative to features deciphered from the images and the anatomical reference in block 312. In block 314, features in the images on a display are labeled to identify a position of the endoscope tip. This is preferably performed in real-time to give clues as to which passage to select or to maintain spatial orientation of the technician/user during the procedure.
In interpreting the appended claims, it should be understood that:
Having described preferred embodiments for systems and methods for real-time scope tracking and branch labeling without electro-magnetic tracking and pre-operative scan roadmaps (which are intended to be illustrative and not limiting), it is noted that modifications and variations can be made by persons skilled in the art in light of the above teachings. It is therefore to be understood that changes may be made in the particular embodiments of the disclosure disclosed which are within the scope of the embodiments disclosed herein as outlined by the appended claims.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/IB2010/051452 | 4/2/2010 | WO | 00 | 11/7/2011 |
Number | Date | Country | |
---|---|---|---|
61176539 | May 2009 | US |