Automatic brightness control of a region of interest based on medical images

Abstract

A method for performing automatic brightness control can include receiving an image of a target site from an imaging system of a medical device, and identifying a region of interest in the image. The region of interest can include a physical feature in the target site identified in the image. The method can further include determining a current image brightness value of the identified region of interest, and adjusting one or more operating parameters of the imaging system based on the current image brightness value and a target image brightness value.

Description

Cross-references to related applications

[0001] This application claims the benefit of priority to U.S. Provisional Patent Application No. 63 / 595,389, filed November 2, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This disclosure generally relates to systems and methods for automatic brightness control. More specifically, aspects of this disclosure relate to systems and methods for identifying one or more regions of interest in an image and performing automatic brightness control based on the identified regions of interest. Background Technology

[0003] Medical imaging systems may include imaging devices and a light source integrated with a medical device, such as an endoscope. During a medical procedure, an endoscope may be inserted into a patient's body cavity and navigate through that cavity to a target site. The light source may be configured to emit light onto the target site to illuminate objects and / or features within the target site, thereby facilitating their visualization in images captured by the imaging device. In some examples, the medical imaging system may be configured to perform automatic brightness control to evaluate and adjust (if necessary) one or more parameters of the imaging system, and / or perform additional image processing to optimize the brightness or illumination of the target site in subsequent images captured by the imaging device. Summary of the Invention

[0004] The techniques described herein relate to methods for performing automatic brightness control. Example methods may include: receiving an image of a target site from an imaging system of a medical device, and identifying a region of interest (ROI) in the image. The ROI may include physical features in the identified target site in the image. The method may further include: determining a current image brightness value for the identified ROI, and adjusting one or more operating parameters of the imaging system based on the current image brightness value and a target image brightness value.

[0005] In any of the example methods disclosed herein, the method may include determining an average pixel intensity value for a subset of pixels in the image that includes the region of interest. The average pixel intensity value may be the current image brightness value. In some examples, identifying the region of interest in an image includes detecting multiple edges in the image and identifying the subset of pixels with the highest edge density among the multiple pixel subsets in the image as the region of interest. The pixel subset may include physical features. The image may be converted to grayscale before detecting multiple edges.

[0006] In other examples, the image is in a first color space, and identifying regions of interest (ROIs) in the image may include: converting the image from the first color space to a second color space; generating multiple histograms for the image in the second color space; performing a color selection process based on analysis of one or more histograms in the multiple histograms; and identifying ROIs based on the color selection process. The second color space may include multiple channels, and generating the multiple histograms may include generating one histogram for each of the multiple channels, wherein the analyzed one or more histograms represent the color distribution in the image.

[0007] In some aspects, analysis of one or more histograms among the plurality of histograms may include identifying color shadow distinctions in the image, and determining suspicious color regions in the image based on deviations of the identified color shadow distinctions from color shadow distinction patterns for the anatomical structure type included in the image. A subset of pixels in the image that includes suspicious color regions may be identified as regions of interest, and the subset of pixels may include physical features. In some examples, determining suspicious color regions may include comparing one or more histograms among the plurality of histograms with one or more reference color shadow distinction patterns for the anatomical structure type to identify deviations. In other examples, determining suspicious color regions may include feeding the one or more histograms among the plurality of histograms as input to a machine learning model trained to identify deviations from one or more learned color shadow distinction patterns for the anatomical structure type.

[0008] In some aspects, performing a color selection process based on the analysis of one or more histograms among multiple histograms may include applying a mask to each pixel in the image that is not included in the suspected color region to generate a masked image. Alternatively, a binary image may be generated based on the masked image to facilitate the identification of regions of interest.

[0009] In another example, identifying a region of interest (ROI) in an image may include: receiving a feature type associated with the physical feature to be identified in the image, and detecting physical features in the image corresponding to that feature type based on the feature type. A subset of pixels in the image that includes the detected physical feature can be identified as a ROI. The feature type can be a shape or pattern associated with the physical feature.

[0010] In some aspects, the imaging system includes a light source configured to illuminate a target area, and adjusting one or more operating parameters of the imaging system may include adjusting the intensity of light emitted by the light source to illuminate the target area. The intensity of the light can be adjusted by controlling the amount of current supplied to the light source. In other aspects, the imaging system includes an imaging device configured to capture an image, and adjusting one or more operating parameters of the imaging system may include adjusting one or more of the gain or exposure time of the imaging device.

[0011] Furthermore, the technology described herein relates to a computing system for performing automatic brightness control, the computing system including at least one memory storing instructions and at least one processor coupled to the at least one memory and configured to execute the instructions to perform operations. These operations may include: receiving from a medical imaging system including an imaging device and a light source an image of a target region captured by the imaging device when the light source illuminates the target region. The image may include a plurality of pixels. These operations may further include identifying a subset of the plurality of pixels as a region of interest in the image. The subset of the plurality of pixels may include physical features detected in the target region in the image. These operations may further include: determining a current image brightness value based on an average pixel intensity value of the subset of the plurality of pixels, and adjusting one or more operating parameters of one or more of the light source or the imaging device based on the current image brightness value to achieve a target image brightness value for the subset of the plurality of pixels identified as the region of interest.

[0012] In any exemplary computing system disclosed herein, identifying a subset of the plurality of pixels as a region of interest in the image may include: detecting a plurality of edges in the image, and identifying a subset of the plurality of pixels with the highest edge density as the region of interest from the plurality of pixels in the image.

[0013] In other examples, the image may be in a first color space, and identifying a subset of multiple pixels as a region of interest in the image may include: converting the image from the first color space to a second color space; generating multiple histograms for the image in the second color space; identifying color shadow distinctions in the image based on analysis of one or more histograms; and determining suspicious color regions in the image based on the deviation of the identified color shadow distinctions from color shadow distinction patterns for the anatomical structure type at the target site. A subset of multiple pixels in the image that includes the suspicious color regions can be identified as regions of interest.

[0014] In another example, identifying a subset of pixels as a region of interest in an image may include: receiving a feature type associated with a physical feature to be detected in the image, and detecting physical features in the image corresponding to that feature type based on the feature type. A subset of pixels in the image that includes the detected physical features can be identified as a region of interest.

[0015] Other aspects of the technology described herein may relate to a medical imaging system. Example medical imaging systems may include a medical device comprising an imaging apparatus configured to capture an image of a target site and a light source configured to illuminate the target site during image capture. The medical imaging system may also include a computing device communicatively coupled to the medical device. The computing device may include at least one memory storing instructions and at least one processor coupled to the at least one memory and configured to execute the instructions to perform operations. These operations may include: receiving an image from the medical device and identifying a region of interest (ROI) in the image, the ROI including physical features detected in the target site in the image. The ROI may be identified using at least one of edge detection, color detection, or feature detection to detect physical features. These operations may further include: determining a current image brightness value for the identified ROI, and adjusting one or more operating parameters of one or more of the light source or imaging apparatus based on the current target image brightness value and the target brightness value of the identified ROI to optimize the brightness of the ROI in subsequent images of the target site captured by the imaging apparatus.

[0016] It is understood that the foregoing general description and the following detailed description are exemplary and illustrative only, and do not limit the claimed invention. As used herein, the terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article of manufacture, or apparatus that comprises a list of elements may include not only those elements but also other elements not expressly listed or inherent to such process, method, article of manufacture, or apparatus. The term “exemplary” is used in the sense of “example” rather than “ideal.” The term “distal” refers to a direction away from the operator / towards the treatment site, and the term “proximal” refers to a direction towards the operator. The term “about” or similar terms (e.g., “substantially”) include values ​​+ / - 10% of the stated value. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate examples of this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0018] Figure 1 An example environment in which an automatic brightness control process can be performed is described.

[0019] Figure 2 An example process for performing automatic brightness control based on the identified region of interest is described.

[0020] Figure 3A and Figure 3B An example process for using edge detection to identify regions of interest in an image is described.

[0021] Figure 4A and Figure 4B An example process for using color detection to identify regions of interest in an image is described.

[0022] Figure 5 An example process for using feature detection to identify regions of interest in an image is described.

[0023] Figure 6 An example computing device is depicted. Detailed Implementation

[0024] As briefly mentioned above, a medical imaging system may include a light source configured to emit light onto a target site during a medical procedure to illuminate physical objects and / or features within the target site, thereby facilitating visualization of the physical objects and / or features in images of the target site captured by the system's imaging apparatus. In some examples, the medical imaging system may be configured to perform automatic brightness control to evaluate and adjust (if necessary) one or more operating parameters of the medical imaging system, and / or perform additional image processing to optimize the brightness or illumination of subsequent images captured by the imaging apparatus.

[0025] Some conventional systems and methods for performing automatic brightness control assess the brightness across the entire image and, based on this assessment, adjust one or more operating parameters of the medical imaging system to optimize the overall brightness or illumination of the image. For example, an image may comprise multiple pixels. As part of the assessment, the pixel intensity values ​​of multiple pixels may be averaged to determine the target image brightness value for adjustment. However, the region of interest (ROI), which includes one or more objects and / or features to be visualized, typically lies only in a subset or portion of the image. Therefore, adjustments made to optimize the overall brightness or illumination of the image may not necessarily produce optimal brightness or illumination for the ROI.

[0026] Other conventional systems and methods for performing automatic brightness control may assess the brightness in the central region of an image and adjust one or more operating parameters of the medical imaging system based on this assessment to optimize the brightness or illumination of the central region. For example, as part of the assessment, pixel intensity values ​​of a subset of pixels forming the central region of the image may be averaged, and adjustments may be made based on the average pixel intensity value of the central region. The central region may be assessed based on the assumption that it most likely contains the object of interest and / or feature to be visualized. However, the object of interest and / or feature may not typically be located in the central region of the image or may extend beyond it. Furthermore, based on anatomical configuration, an endoscopic operator may not be able to navigate the medical device in a manner that would allow the object of interest and / or feature to be located in the central region.

[0027] Other conventional systems and methods for performing automatic brightness control may adjust one or more operating parameters of a medical imaging system to reduce or eliminate hotspots. Hotspots may consist of saturated pixels (e.g., white pixels) that cause details of any object and / or feature at the location of the saturated pixel to be washed away or become unvisible. To reduce or eliminate hotspots, an exemplary automatic brightness control process may determine the percentage or ratio of saturated pixels and adjust one or more operating parameters of the medical imaging system until the percentage or ratio of saturated pixels falls below a threshold. While such adjustment may reduce or eliminate hotspots, it may also darken the remainder of the image, potentially affecting the operator's ability to visualize objects and / or features of interest when they are now located in darker parts of the image.

[0028] Therefore, aspects of this disclosure relate to systems and methods for performing automatic brightness control based on or specific to regions of interest (ROIs) identified in images captured by a medical imaging system. The disclosed aspects improve upon the field of medical image processing, including, for example, the field of automatic image brightening control. The identified ROI can be a region of an image comprising a physical object and / or feature of interest to be visualized within a target area. Various techniques, including but not limited to edge detection, color detection, and / or feature detection techniques, can be applied to identify ROIs in an image. Automatic brightness control can then be performed to achieve a target image brightness value that optimizes (or otherwise improves) visualization within the identified ROI. For example, the current image brightness value of the ROI can be determined based on the average pixel intensity values ​​of a subset of pixels in the image that includes the identified ROI. One or more operating parameters of the medical imaging system can then be adjusted based on the current image brightness value and the target brightness value. For example, the difference between the current image brightness value and the target brightness value can be determined, and this difference can be used to adjust one or more operating parameters to achieve the target brightness value for the ROI. This adjustment can, for example, optimize the brightness or illumination of physical objects and / or features within the region of interest in subsequent images of the target site by optimizing the raw image data captured by the imaging device. In some examples, additional image processing techniques can be performed on the received raw image data to further enhance the brightness or illumination beyond what can be achieved by adjusting the operating parameters of the imaging system. Compared to conventional techniques, the disclosed aspects improve the visualization of the region of interest within the target site.

[0029] Figure 1 An exemplary environment 100 is depicted in which an automatic brightness control process can be implemented. Environment 100 may include one or more of a medical device 102, a computing system 104, one or more display devices 106, one or more optional server-side systems 130 and / or a network 140.

[0030] Medical device 102 can be used to perform diagnostic and / or interventional medical procedures on a patient, which, for the sake of brevity, will be referred to as medical procedures below. Medical device 102 may be an endoscope or other type of endoscope, such as a bronchoscope, ureteroscope, duodenoscope, gastroscope, endoscopic ultrasound (“EUS”), colonoscope, laparoscope, arthroscope, cystoscope, aspiration endoscope, sheath or catheter, etc.

[0031] Medical device 102 may include imaging system 108. Imaging system 108 may include at least one imaging device 110 and at least one light source 112. Imaging device 110 may be located at the distal end of medical device 102 (e.g., at the distal tip of medical device 102). Imaging device 110 may be configured to continuously capture image signals during medical procedures as the distal end of medical device 102 is inserted into a patient's body cavity and navigates through the body cavity to a target site. Imaging device 110 may include one or more cameras, one or more image sensors, one or more endoscopic viewing elements, or one or more optical components including one or more image sensors and one or more lenses, and other similar devices. In some examples, light source 112 may be located at the distal end of medical device 102 (e.g., at the distal tip of medical device 102) together with imaging device 110. In other examples (not shown), the light source 112 may be a standalone device or may be integrated with the computing system 104, wherein light from the light source 112 is transmitted via an optical fiber extending the length of the medical device 102 (e.g., from the proximal end of the medical device 102 connected to the computing system 104 to the distal end of the medical device 102). The light source 112 may be configured to illuminate an area of ​​the patient's body (e.g., a target site) during a medical procedure to facilitate imaging of the target site by the imaging device 110. The light source 112 may include one or more LEDs, incandescent light sources, optical fibers, and / or other illuminators.

[0032] One or more components of the medical device 102 (including the imaging system 108 and its components) may be communicatively coupled to the computing system 104 via wired and / or wireless connections (e.g., via network 140) to enable communication of various signals between the medical device 102 and the computing system 104. For example, image signals (e.g., raw image data) captured by the imaging device 110 may be received by the computing system 104. Additionally, the computing system 104 may provide one or more signals to the imaging device 110 and / or the light source 112 to induce separate adjustments to one or more parameters of the imaging device 110 and / or the light source 112, as described in detail below.

[0033] In some examples, the computing system 104 is a controller, control unit, computing device, or other similar independent processing unit separate from the medical device 102. In other examples, the computing system 104 may be integrated with the medical device 102. For example, the computing system 104 may be located in the handle of the medical device 102. In other examples, the computing system 104 may be located at the distal end of the medical device 102.

[0034] The computing system 104 may include a memory 114 and one or more processors 116. The memory 114 may store instructions to be executed by the processors 116 to cause the computing system 104 to perform corresponding operations. At least a portion of the instructions stored in the memory 114 may include an automatic brightness control process. The memory 114 may also include one or more data stores. Alternatively or additionally, the computing system 104 may include one or more data stores separate from the memory 114. The processor 116 may include at least one image processor 118. The image processor 118 may be configured to process one or more image signals (e.g., raw image data) captured by the imaging device 110 and received by the computing system 104 to generate an image. Additionally, the image processor 118 may be configured to apply an automatic brightness control process to the image. As described in more detail below, the automatic brightness control process may include: identifying a region of interest in the image, and determining the current brightness value of the identified region of interest, such that one or more operating parameters of the imaging system 108 can be adjusted to achieve a target brightness value for the identified region of interest. In some examples, the image processor 118 may be or include a field-programmable gate array (FPGA), a digital signal processing (DSP) processor, a graphics processing unit (GPU), etc.

[0035] The computing system 104 may further include an optional communication interface 120 for providing connectivity to the network 140. The optional communication interface 120 may also provide connectivity to the medical device 102 and / or the display device 106. In some examples, communication connections between the computing system 104 and the medical device 102 (or components thereof) and / or between the computing system 104 and the display device 106 may be at least partially supported via the network 140.

[0036] Display device 106 can be configured to display image data, including at least an image generated by computing system 104. In some examples, the image data may also include an image with a visual indicator of an identified region of interest as part of an automatic brightness control process. For example, the visual indicator may be superimposed on the image at the location of the region of interest. Display device 106 may include one or more combinations of a monitor, computing device screen, touchscreen display device, etc. In some examples, one or more of display devices 106 may be devices separate from computing system 104, which may be communicatively coupled to computing system 104 via a wired and / or wireless connection. In other examples, at least one of display devices 106 may be a display of computing system 104 itself.

[0037] In some examples, computing system 104 may generate or cause the generation of one or more graphical user interfaces (GUIs) based on instructions or information stored in memory 114, instructions or information received from one or more optional server-side systems 130, and may cause these GUIs to be displayed via display device 106. The GUI may be, for example, an application interface or a browser user interface, and may include text, selection controls, etc., in addition to the displayed image data. Display device 106 may include a touchscreen or a display with other input systems (e.g., mouse, keyboard, voice, etc.) for an operator of computing system 104 to control the functions of computing system 104, medical device 102 (or components thereof), and / or display device 106 via computing system 104. As an example, an operator may select one or more control elements displayed on the GUI of display device 106 to manually adjust one or more operating parameters of imaging system 108 (e.g., based on operator preference). This selection may be received by computing system 104, and a corresponding signal may be transmitted from computing system 104 to imaging system 108 and / or its specific components.

[0038] One or more components of environment 100 (such as medical device 102, computing system 104, and / or display device 106) may be network-connected and may communicate with each other via wired or wireless networks (such as network 140). Network 140 may be an electronic network. Network 140 may include one or more wired and / or wireless networks, such as a wide area network (“WAN”), a local area network (“LAN”), a personal area network (“PAN”), a cellular network (e.g., 3G network, 4G network, 5G network, etc.). In other examples, components of environment 100 may communicate and / or connect to network 140 via a universal serial bus (USB) or other similar local low-latency connection or direct wireless protocol. Components of environment 100 may connect via network 140 using one or more standard communication protocols, enabling components to transmit and receive communications with each other across network 140.

[0039] In some examples, environment 100 may also include one or more optional server-side systems 130 when one or more components of environment 100 are connected to network 140. Optional server-side systems 130 may include one or more remote image processing systems configured to perform at least a portion of image processing, including but not limited to more resource-intensive processes such as machine learning processes (e.g., to conserve local resources of computing system 104 when network connectivity is available). Alternatively or additionally, optional server-side systems 130 may include data storage systems for storing images generated by computing system 104 (e.g., in response to action input from an operator to record or otherwise save an image). In some examples, at least one of the data storage systems may include a Picture Archiving and Communication System (PACS) that stores the image, as well as other types of imaging data associated with the patient from various imaging modalities (e.g., ultrasound, MRI, nuclear medicine imaging, positron emission tomography, computed tomography, mammography, digital radiography, histopathology, etc.).

[0040] Despite the various components in Environment 100 Figure 1 While depicted as separate components, it should be understood that in some embodiments, a component or a portion thereof in environment 100 may be integrated with or incorporated into one or more other components. For example, one of the display devices 106 may be integrated with computing system 104, and / or computing system 104 may be integrated with medical device 102. In some embodiments, the operation or aspects of one or more of the components discussed above may be distributed across one or more other components. Any suitable arrangement and / or integration of various systems and devices in environment 100 may be used.

[0041] Specific examples included throughout this disclosure implement endoscopic imaging systems configured to perform automatic brightness control processes in real-time or near real-time during medical procedures based on identified regions of interest, thereby optimizing the brightness or illumination of the identified regions of interest. However, it should be understood that the techniques according to this disclosure can be adapted to other medical imaging systems with different types of imaging devices and light sources. It should also be understood that the examples above are merely illustrative. The techniques and methods of this disclosure can be adapted to any suitable activity.

[0042] Figure 2An example process 200 for performing automatic brightness control based on an identified region of interest is depicted. In some examples, one or more steps of process 200 may be performed by computing system 104. At step 202, process 200 may include receiving an image from a medical imaging system (such as imaging system 108). For example, image processor 118 may receive and process image signals including raw image data captured by imaging device 110 when medical device 102 is positioned at a target site to generate an image. The image may be provided to one or more display devices 106 for display.

[0043] The image can be a color image of the target area, consisting of multiple pixels, such as a red-green-blue (RGB) image. The target area can include one or more objects or features of interest to be visualized, such as polyps, lesions and / or other objects, structures, or features indicating abnormal tissue or foreign bodies.

[0044] At step 204, process 200 may include identifying a region of interest (ROI) in the image. The ROI may include a subset of pixels in the image comprising a physical object or feature of interest within the target region to be visualized in the image. One or more detection methods or techniques may be used to identify the ROI, including but not limited to, reference... Figure 3A and Figure 3B The described edge detection, as in the reference Figure 4A and Figure 4B The described color detection, or as referenced Figure 5 The described feature detection. In some examples, visual indicators of the identified regions of interest can be overlaid on the image and provided to one or more display devices 106 for display.

[0045] At step 206, process 200 may include determining a current image brightness value for the identified region of interest. For example, the current image brightness value may be determined by averaging pixel values ​​representing the intensity or brightness of a subset of pixels included in the region of interest (e.g., pixel intensity values ​​of the subset of pixels). For example, the average pixel intensity value of the subset of pixels may be the current image brightness value.

[0046] At step 208, process 200 may include adjusting one or more parameters of imaging system 108 based on the current image brightness value and the target image brightness value for the identified region of interest. For example, the difference between the current image brightness value and the target brightness value may be determined. This difference may be used to adjust one or more operating parameters to achieve the target brightness value for the identified region of interest. The target brightness value may be a predefined value for achieving optimal (or otherwise improved) visualization. Specifically, the target brightness value may be a predefined percentage brightness value on a scale from 0% (e.g., a completely black image) to 100% (e.g., a completely white image). As a non-limiting example, the target brightness value may be 40% + / - 5% brightness. The target brightness value may be stored in memory 114 of computing system 104 and / or communicatively coupled to other data storage of computing system 104 (e.g., the data storage system of optional server-side system 130) to enable retrieval of the target brightness value for use in process 200.

[0047] In some examples, the target brightness value can be adjusted based on operator preference. Continuing the example above, the target brightness value can be increased or decreased from a predefined percentage brightness value to a higher or lower percentage brightness value (e.g., it can be increased to 50% + / - 5%). The operator can interact with computing system 104 and / or display device 106 (e.g., by providing input via one or more associated input systems or devices) to adjust the brightness value. In some examples, the adjusted brightness value can be associated with the user and stored in memory 114 and / or other data sources for later retrieval and use.

[0048] One example of an operating parameter being adjusted is the intensity of light emitted by light source 112. To adjust the intensity, based on the current image brightness value, the calculation system 104 can control (e.g., increase or decrease) the amount of current supplied to light source 112 so that the intensity of light emitted by light source 112 meets the target image brightness value. For example, based on information provided by the manufacturer of light source 112 and / or based on calibration performed prior to the distribution and / or use of medical device 102, the correlation between the value of the current supplied to light source 112 and the intensity of light emitted from that light source 112 can be known. Using the known correlation, the calculation system 104 can adjust the current value supplied to light source 112 that causes the current image brightness value to a current value corresponding to the light intensity value that meets the target image brightness value. The adjustment can further depend on the type of anatomical structure at the target site. In some examples, the calculation system 104 can implement a proportional-integral-derivative (PID) loop to control the intensity adjustment. Alternatively, depending on the type of light source 112, the computing system 104 may adjust one or more filters located between the light source 112 and one or more optical fibers to adjust the intensity of the emitted light or reduce the brightness of the light source 112.

[0049] Another example of the operating parameter being adjusted in step 208 could be the gain of the imaging device 110. Gain adjustment can be a means of adjusting the apparent sensitivity of the imaging device 110 to light. For example, gain can represent the relationship between the number of electrons acquired on the image sensor of the imaging device 110 and the analog-digital unit (ADU) representing the generated image signal. Increasing the gain amplifies the signal by increasing the ratio of the ADU to the electrons acquired on the image sensor. Therefore, increasing the gain can improve the apparent brightness of the image at a given exposure. Conversely, decreasing the gain can reduce the apparent brightness. Based on the current image brightness value, the calculation system 104 can send a signal to the imaging device 110 to control (e.g., increase or decrease) the gain to achieve an apparent image brightness that meets the target image brightness value.

[0050] Another example of an operational parameter that can be adjusted is the exposure time of the imaging device 110. The exposure time (also known as shutter speed) of the imaging device 110 can be the duration for which the image sensor is exposed to light. Increasing the duration can cause the sensor to receive more light, resulting in increased pixel intensity and brightness of the image. Conversely, decreasing the duration can cause the sensor to receive less light, resulting in decreased pixel intensity and brightness of the image. Based on the current image brightness value, the calculation system 104 can send a signal to the imaging device 110 to control (e.g., increase or decrease) the exposure time to achieve an image brightness that meets the target image brightness value.

[0051] The adjustments performed at step 208 can, for example, optimize the brightness or illumination of the region of interest in subsequent images of the target area by optimizing the raw image data captured by imaging device 110. In some examples, additional image processing techniques can be performed on the received raw image data to further enhance the brightness or illumination beyond what can be achieved by adjusting the operating parameters of imaging system 108.

[0052] Therefore, some aspects may include performing automatic brightness control based on the identified region of interest. The process 200 described above is provided only as an example and may include... Figure 2 The steps described in the text are compared to additional, fewer, different, or arranged steps. Use edge detection technology to identify regions of interest.

[0053] In an image of an anatomical target site, one or more regions that contain more detail than other areas of the image are candidate regions of interest (ROIs) because greater detail can indicate abnormalities (e.g., abnormal tissue, foreign bodies, etc.). Examples of regions containing more detail can include areas that expose the blood vessel distribution of the underlying tissue, areas that include different textures within the tissue, areas that include different structures within the tissue (e.g., polyps, lesions, etc.), or areas that include any other different structures or appearances within the tissue that indicate the presence of an abnormality. Surrounding tissue (e.g., the healthy walls of a body cavity) may be less detailed (e.g., smoother and / or lacking features) compared to the RIO. Edge detection techniques can be used to identify such regions with more detail as RIOs.

[0054] Figure 3A An example process 300 for identifying regions of interest in an image using edge detection is described. In some examples, one or more steps of process 300 may be performed by a computing system 104. Process 300 may be used to perform a reference... Figure 2 At least a portion of step 204 of the described process 200 is used to identify the region of interest. Figure 3B A conceptual diagram depicting an image generated as part of process 300.

[0055] Also refer to Figure 3A and Figure 3BAt optional step 302, process 300 may optionally include converting image 310 captured by the medical imaging device into a grayscale image (e.g., to generate a grayscale image (not shown)). Image 310 may be an image captured by imaging device 110 and received at step 202 of process 200. For example, image 310 may be a color image consisting of multiple pixels, such as a red-green-blue (RGB) image. Any known or future algorithm for converting a color image to a grayscale image may be used to convert image 310. In some examples, raw image data received as part of an image signal captured by imaging device 110 may be directly converted into a grayscale image (e.g., compared to generating image 310 from raw image data and then converting image 310 into a grayscale image). Image 310 may be converted into a grayscale image when using a grayscale image (as opposed to a color image) as input to an edge detection process to facilitate or otherwise enhance a type of edge detection process applied at step 304 as described below.

[0056] At step 304, process 300 may include detecting multiple edges in image 310 (or, if the transformation at optional step 302 was performed, detecting a grayscale image). Any known or future edge detection process or technique can be used to identify edges. In some examples, if, for example, the appearance of an edge is driven by one of the color channels, edge detection can be performed on a specific color channel (e.g., one of the red, green, or blue channels).

[0057] One example edge detection process that can be implemented is Sobel edge detection using the Sobel operator or filter. When implementing Sobel edge detection, image 310 can be converted to a grayscale image at optional step 302. Sobel edge detection techniques may include measuring pixel intensity values ​​on the grayscale image to identify edges based on changes in pixel intensity values. For example, a 3 × 3 matrix (also called a kernel) can be run on each pixel in the grayscale image. At each iteration, the change in gradient intensity values ​​of pixels falling within the kernel can be measured in all directions. A larger change indicates a more significant edge at that pixel location. If the gradient intensity value measured in any direction exceeds a predefined threshold, the corresponding pixel can be set to white. The remaining pixels whose gradient intensity values ​​do not exceed the predefined threshold can be set to black to generate a binary image, such as binary image 312. Other example edge detection processes can similarly generate binary images including white and black pixels, where white pixels represent detected edges, but different techniques can be used to do this depending on the type of operator or filter utilized.

[0058] In some examples, a predefined threshold used as a benchmark for setting pixel values ​​can be adjusted based on one or more factors. For example, an operator can manually adjust the threshold based on the type of anatomical structure and / or disease state observed at the target site. Alternatively, the edge detection process can further include automatic threshold adjustment. For example, the threshold can be automatically adjusted based on the percentage or ratio of white pixels to black pixels identified in the binary image 312. For example, if the binary image 312 has a large number of edges (e.g., mainly consisting of white pixels), the threshold can be automatically increased. Conversely, if the binary image 312 has a small number of edges (e.g., mainly consisting of black pixels), the threshold can be automatically decreased.

[0059] At step 306, process 300 may include identifying a subset of pixels in image 310 with the highest edge density as a region of interest 316. To identify the subset of pixels with the highest edge density, a sliding window of a predetermined size may be moved across binary image 312 to determine the number of white pixels representing edges within each instance of the sliding window as the sliding window moves. The sliding window may have a variable geometry. For example, the sliding window may be circular, elliptical, square, and / or rectangular, among other geometries. As the sliding window moves across binary image 312, the white pixel count of each subset of pixels forming a given instance of the sliding window may be determined. The subset of pixels forming the instance of the sliding window with the highest white pixel count (e.g., highest edge density) may be identified as region of interest 316.

[0060] In some examples, the predetermined size of the sliding window may be based on one or more physical objects and / or features in the target region that are being visualized and / or detected in the image of the target region. Alternatively, the predetermined size of the sliding window may be based on characteristics of the imaged target region, such as whether the target region includes flat walls or at least a portion of an internal cavity. Typically, the predetermined size of the sliding window needs to be large enough that brightness optimization based on a subset of pixels forming a given instance of the sliding window can provide sufficient brightness or illumination across the region of interest as a whole, enabling the operator to achieve proper visualization. On the other hand, the predetermined size of the sliding window should not be so large that the subset of pixels forming a given instance of the sliding window covers most of the image.

[0061] Once the region of interest 316 is identified, its current image brightness value can be determined. The current image brightness value of the region of interest 316 and the target image brightness value can be used to adjust one or more operating parameters of the imaging system 108, as described in detail in steps 206 and 208 of reference process 200. In some examples, a visual representation of the identified region of interest 316 (e.g., visually represented by an instance of a sliding window with the highest edge density) can be displayed or overlaid on image 310 to generate an enhanced image 314. The enhanced image 314 can be provided by the computing system 104 to one or more of the display devices 106 for display. Although the region of interest 316 is in Figure 3B The region of interest 316 is shown as a square in the enhanced image 314, but it can be any other shape depending on the geometry of the sliding window. Additionally, in some examples, the size of the visual representation of the identified region of interest 316 can be at least slightly larger than the sliding window to prevent occlusion of any pixels within it.

[0062] Therefore, some aspects may include performing an edge detection process for region of interest identification. The process 300 described above is provided only as an example and may include... Figure 3A The steps described in the text are compared to additional, fewer, different, or arranged steps. Use color detection technology to identify regions of interest.

[0063] Typically, healthy tissue will appear visually as similar (e.g., substantially uniform) colors or tones in an image of the tissue. Polyps, precancerous lesions, or necrotic areas within the tissue can be different colors from the surrounding healthy tissue. Therefore, the detection of color or tone differences within an image of tissue (e.g., at a target site) can be used to identify regions of interest.

[0064] Figure 4A An example process 400 for identifying regions of interest in an image using color detection is depicted. In some examples, one or more steps of process 400 may be performed by a computing system 104. Process 400 may be used to perform a reference... Figure 2 At least a portion of step 204 of the described process 200 is used to identify the region of interest. Figure 4B Conceptual diagrams depicting images and / or diagrams generated as part of process 400.

[0065] Also refer to Figure 4A and Figure 4BAt step 402, process 400 may include converting image 410 from a first color space to a second color space. Image 410 may be an image captured by imaging device 110 and received at step 202 of process 200. For example, image 410 may be a color image composed of multiple pixels, such as a red-green-blue (RGB) image. In such an example, the first color space may be an RGB color space with red, green, and blue channels, and image 410 may be converted to a second color space different from the RGB color space. The second color space may be a color space in which color characteristics of image 410 can be quantified to determine differences in color shadows (e.g., color shadow distinction). For example, the second color space may include L a and b The CIELAB color space of the channel. L Channels can represent brightness (e.g., luminance), while a Channel and b Channels represent chromaticity. Chromaticity can be an objective specification of color quality, regardless of brightness. Specifically, a A channel can represent the chromaticity at a color location between red and green, and b A channel can represent the chromaticity of a color position between yellow and blue.

[0066] At step 404, process 400 may include generating a plurality of histograms 412 for the image in the second color space. One histogram may be generated for each of the plurality of channels in the second color space. Continuing with the example of the second color space being the CIELAB color space, histogram 412 may include histograms for L... The first histogram of the channel 414, for a The second histogram of the channel 416 and for b The third histogram of the channel is 418. For example, for L... The first histogram 414 of the channel can represent the brightness distribution of image 410, indicating the number of pixels (e.g., pixel count) at each brightness intensity value on a scale of 0 to 100. For a The second histogram 416 of the channel can represent the color distribution of image 410, indicating the pixel count at each color location between red and green on a given scale. For b The third histogram 418 of the channel can represent the color distribution of the image 410, which indicates the pixel count at each color position between yellow and blue on a given scale.

[0067] At step 406, process 400 may include performing a color selection process based on analysis of one or more histograms in histogram 412. For example, a portion of histogram 412 representing color distribution (such as second histogram 416 and third histogram 418) may be analyzed to identify color shadow distinctions on image 410, while excluding those caused by brightness (as determined by L). The differences are caused by channel representation. Further, one or more suspicious color regions in image 410 can be identified based on analysis. For example, suspicious color regions can be identified based on one or more detected deviations between the identified color shadow differences and typical color shadow difference patterns for the types of anatomical structures included in the image. Example deviations between the identified color shadow differences and typical patterns are visually highlighted by boxes 420, 422, as shown in the second histogram 416 and the third histogram 418. A subset of pixels contributing to the pixel count at the color locations in boxes 420, 422 can be identified as suspicious color regions.

[0068] In some examples, analyzing histogram 412 may include comparing histogram 412, specifically comparing the color distribution representation depicted in the second histogram 416 and the third histogram 418 with one or more reference color tone distinction patterns for that anatomical structure type, to identify biases. This can be based on comparing multiple images converted to a second color space for a... and b Reference patterns are generated by analyzing the color distribution representation in multiple histograms generated from the channels. The analyzed images can depict healthy tissue of the anatomical structures at the target site, enabling the identification of typical patterns or ranges of color distribution on healthy tissue. This can be targeted at a The channel generates a first reference pattern for comparison with the second histogram 416. This can be targeted at b. The channel generates a second reference pattern for comparison with the third histogram 418.

[0069] In other examples, the analysis may include executing a machine learning model trained to identify deviations from one or more learned color dark-tone differentiation patterns for that anatomical structure type. For example, and as described in more detail below, one or more histograms in histogram 412 may be provided as input to the machine learning model to receive the deviations as output. The machine learning model may be executed by computing system 104 (or by one of optional server-side systems 130 to conserve local computing resources).

[0070] In some examples, computing system 104 may perform one or more of the following: generating, storing, training, or using a machine learning model. Computing system 104 may include a machine learning model and / or instructions associated with the machine learning model, such as instructions for generating the machine learning model, training the machine learning model, using the machine learning model, etc. In other embodiments, systems or devices other than computing system 104 may be used to generate and / or train the machine learning model. For example, such a system may include instructions for generating the machine learning model and training data, and / or instructions for training the machine learning model. The trained machine learning model can then be provided to computing system 104 for use.

[0071] To train a machine learning model, multiple training datasets can be acquired and processed to generate (e.g., build) the machine learning model. An exemplary training dataset may include multiple histograms generated from multiple images of a specific anatomical structure type, captured using the same imaging modality as imaging device 110 (e.g., endoscopic images rather than X-ray images) and converted from a first color space to a second color space. The generated histograms may resemble histogram 412 generated at step 404. In some examples, when utilizing supervised or semi-supervised machine learning techniques, the exemplary training dataset may also include corresponding labels indicating actual subsets of pixels in the images, representing one or more deviations from a typical color distribution pattern for that anatomical structure type. In some examples, the model may be trained based on color differences rather than absolute color. For example, a ΔE measurement in the CIELAB color space can be used to determine color differences (e.g., color darkness distinctions) from which deviations are identified. The model can be trained to identify subsets of pixels in an image whose color differences (e.g., ΔE) are anomalous or deviate from typical patterns. The training dataset may be generated, received, or otherwise obtained from internal and / or external resources.

[0072] Typically, a model comprises a set of variables, such as nodes, neurons, filters, etc., which are tuned (e.g., weighted, biased) to different values ​​by applying a training dataset. In some examples, the training process can employ supervised, unsupervised, semi-supervised, and / or reinforcement learning procedures to train the model. In some examples, a portion of the training dataset can be retained during training and / or used to validate the trained machine learning model.

[0073] When employing a supervised learning process, labels corresponding to images in the training dataset can facilitate the learning process by providing real data. Training can be performed by feeding one or more histograms of the training dataset (e.g., samples) into a model with variables set with initial values ​​(e.g., randomly), based on Gaussian noise, a pre-trained model, etc. The model can output a subset of pixels from the samples that are predicted to deviate from a typical color distribution pattern. The output can be compared with corresponding labels (e.g., real data) indicating the deviation from the typical color distribution pattern to determine the error, which can then be backpropagated through the model to adjust the values ​​of the variables. This process can be repeated for multiple samples, at least until the determined loss or error is below a predefined threshold.

[0074] For unsupervised learning processes, the training dataset may not include pre-assigned labels used to assist the learning process. Instead, unsupervised learning processes can include clustering, classification, etc., to identify patterns naturally present in the training dataset. K-means clustering or K-nearest neighbors can also be used, and these can be supervised or unsupervised. Combinations of K-nearest neighbors and unsupervised clustering techniques can also be used. For semi-supervised learning, the model can be trained using a combination of a training dataset with pre-assigned labels and a training dataset without pre-assigned labels.

[0075] When reinforcement learning is employed, an agent (e.g., an algorithm) can be trained to make decisions about predicted subsets of pixels from samples in a training dataset through trial and error. For example, upon making a decision, the agent can then receive feedback (e.g., a positive reward if the predicted subset of pixels does indeed represent a deviation from a typical color distribution pattern), adjust its next decision to maximize the reward, and repeat until the loss function is optimized.

[0076] In some examples, trained machine learning models can be generated and / or trained such that the model is general or applicable across images of different anatomical structures and / or across anatomical structures that include variable objects of interest or features. In other examples, separately trained machine learning models specific to the anatomical structure type and / or specific to the object or feature type can be generated. For example, a first trained machine learning model can be generated for images of the esophagus, a second trained machine learning model can be generated for images of the stomach, and so on. Alternatively or additionally, a first set of trained machine learning models can be generated for images of the esophagus, wherein one model in the first set is applied to images of the esophagus including polyps, another model in the first set is applied to images of the esophagus including lesions, and so on.

[0077] Once trained, the machine learning model can be stored (e.g., in memory 114 on computing system 104, and / or in one or more data storage systems associated with optional server-side system 130) and subsequently applied during the deployment phase. When multiple anatomical structure type-specific and / or object or feature type-specific machine learning models are generated and stored, the machine learning models can be stored in association with identifiers indicating the specific anatomical structure type and / or object or feature type for which the machine learning model was trained, to facilitate subsequent retrieval and application.

[0078] During the deployment phase, the trained machine learning model can receive input data. The input data may include one or more histograms from histogram 412 generated at step 404, such as at least the second histogram 416 and the third histogram 418. The trained machine learning model can provide a subset of pixels identified as deviating from typical color distribution patterns (e.g., suspicious color regions) as output data.

[0079] In some examples, a trained machine learning model can be retrained or updated based on received feedback. For example, the values ​​or weights of one or more variables in a trained machine learning model can be adjusted to improve its accuracy. Feedback can include indications from the operator whether a subset of pixels identified as biased is indeed a subset representing a region of interest. In some examples, feedback can include the correct subset of pixels when the identified subset is inaccurate. Feedback can be used as labels to create new training datasets for retraining the trained machine learning model. In some examples, the trained machine learning model can be retrained after a predefined number of new training datasets have been received.

[0080] Returning to the color selection process in step 406, once one or more histograms in histogram 412 have been analyzed to identify color shadow differences in image 410, and suspicious color regions in the image have been determined based on the deviation of the identified color shadow differences from typical patterns, the color selection process can then be applied to image 410 based on this analysis. A masked image 424 can be generated. For example, to generate masked image 424, a mask can be applied to each pixel in image 410 that is not included in the identified suspicious color regions. Applying a mask to a given pixel sets the pixel to black. Each pixel to which a mask is applied can be referred to as a masked pixel. The remaining pixels of image 410 included in the identified suspicious color regions can be referred to as unmasked pixels.

[0081] At step 408, process 400 may include identifying the region of interest based on a color selection process. In some examples, and such as... Figure 4BAs shown, a binary image 426 can be generated based on the mask image 424 to facilitate the identification of regions of interest. For example, the pixel value of any non-masked pixel in the mask image 424 (e.g., pixels included in identified suspicious color regions) can be set to white, and white pixels can be identified as regions of interest. In other examples, non-masked pixels in the mask image 424 can be directly identified as regions of interest.

[0082] Once the region of interest (ROI) is identified, the current image brightness value of the ROI can be determined. The current image brightness value of the ROI and the target image brightness value can be used to adjust one or more operating parameters of the imaging system 108, as described in detail in steps 206 and 208 of reference process 200. In some examples, an enhanced image can be generated using a mask image 424 and / or a binary image 426, which includes a visual indicator of the identified ROI superimposed on image 410. The visual indicator can be the outline, shadow, highlight, and / or other similar visual emphasis of non-masked pixels in the mask image 424. In some examples, the visual indicator may also include a portion of the mask pixels surrounding or adjacent to non-masked pixels to prevent the visual indicator from obscuring any non-masked pixels. The enhanced image can be provided by the computing system 104 to one or more display devices 106 for display.

[0083] Therefore, some aspects may include performing a color detection process for region of interest identification. The process 400 described above is provided only as an example and may include... Figure 4A The steps described in the text are compared to additional, fewer, different, or arranged steps. Use feature detection technology to identify regions of interest.

[0084] Prior to a patient undergoing a medical procedure using medical device 102, imaging in various modalities (e.g., X-ray, computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), etc.) is typically acquired and evaluated. Based on this preoperative imaging and / or patient-reported symptoms, features associated with an object or structure of interest at the target site of the medical procedure can be identified or at least inferred. For example, the target site may include polyps or lesions, and the overall shape of the polyps or lesions can be determined or inferred. As another example, the target site may include the distribution of blood vessels in the underlying tissue, and a grid pattern of the blood vessel distribution can be determined or inferred. Therefore, detection of these features in images of the target site captured during the medical procedure can be used to identify regions of interest.

[0085] Additionally, feature detection can be used as a default detection technique for identifying regions of interest based on the type of medical procedure being performed (e.g., compared to using edge or color detection techniques as described above). For example, polyp and lesion feature detection could be the default detection technique applied to identify regions of interest when performing a cancer screening procedure.

[0086] Figure 5 An example process 500 for identifying regions of interest in an image using feature detection is described. In some examples, one or more steps of process 500 may be performed by a computational system 104. Process 500 may be used to perform a reference... Figure 2 At least a portion of step 204 of the described process 200 is used to identify the region of interest.

[0087] At step 502, process 500 may include receiving a feature type to be detected in an image. The image in which the feature type is to be detected may be an image of the target site captured by imaging device 110 and received at step 202 of process 200. Example feature types received in step 502 may include shapes, patterns, or other similar visual or appearance-based features expected to be associated with an object or structure of interest at the target site. As an example, if the medical procedure is for a biopsy or polyp removal, the feature type may be a circular or elliptical shape. As another example, if a medical procedure is being performed to observe the distribution of blood vessels in the underlying tissue at the target site, the feature type may be a grid pattern.

[0088] Feature types can be received as input into computing system 104. For example, feature types can be received from an input system associated with one or more display devices in computing system 104 and / or display device 106. For example, an operator can manually input a feature type using the input system by entering the feature type into a displayed text box, selecting a feature type from a menu of displayed feature type options, and / or engaging in other similar interactions with one or more display devices in computing system 104 and / or display device 106. Alternatively, the operator can manually input an object or structure of interest, and computing system 104 can automatically select a feature type based on the object or structure of interest. For example, if the operator indicates a polyp as an object or structure of interest, a circular or elliptical shape can be automatically selected as the feature type.

[0089] In some examples, feature types can be received prior to the medical procedure. For instance, the received feature types may be based on information obtained from preoperative imaging and / or patient-reported symptoms. In other examples, feature types can be received during the medical procedure once the target site has been visualized. In yet another example, feature types can be received prior to the medical procedure, and if necessary, adjusted during the procedure once the target site has been visualized.

[0090] At step 504, process 500 may include detecting features corresponding to the received feature types in the image. A subset of pixels in the image that includes the detected features may be identified as regions of interest. Any known or future feature detection process or technique may be used to detect the features. For example, the image and feature types may be provided as input to the feature detection process, and each pixel in the image may be analyzed to determine whether a feature of that feature type is present at that pixel. Each pixel where a feature of that feature type is determined to be present may be included within the subset of pixels identified as regions of interest. In some examples, when multiple feature types are received, step 504 may be performed iteratively for each feature type to identify one or more regions of interest corresponding to each feature type.

[0091] Once the region of interest (ROI) is identified, the current image brightness value of the ROI can be determined. The current image brightness value of the ROI and the target image brightness value can be used to adjust one or more operating parameters of the imaging system 108, as described in detail in steps 206 and 208 of reference process 200. In some examples, an enhanced image can be generated, which includes a visual indicator superimposed on the identified ROI. The visual indicator can be the outline, shadow, highlight, and / or other similar visual emphasis of a subset of pixels that identify the presence of a feature type. In some examples, the visual indicator may also include a subset of pixels surrounding or adjacent to a subset of pixels that identify the presence of that feature type, to prevent the visual indicator from occluding any pixels in the subset. The enhanced image can be provided by the computing system 104 to one or more of the display devices 106 for display.

[0092] Therefore, some aspects may include performing a feature detection process for region of interest identification. The process 500 described above is provided only as an example and may include... Figure 5 The steps described in the text are compared to additional, fewer, different, or arranged steps.

[0093] Figure 6 An example of a computer 600 is depicted. Figure 6This is a simplified functional block diagram of a computer 600 according to an exemplary embodiment of the present disclosure, the computer being configured to perform... Figures 2 to 5 The device depicted or described with respect to these figures represents a process, step, or operation. For example, according to exemplary embodiments of this disclosure, computer 600 may be configured as one or more of a medical device 102, a computing system 104, a display device 106, an optional server-side system 130, and / or another device or component. In various embodiments, any system herein may be or include computer 600, which includes, for example, a data communication interface 620 for packet data communication. Computer 600 may communicate with one or more other computers, for example, using an electronic network 626 (e.g., via data communication interface 620). Electronic network 626 may include wired or wireless networks, such as those similar to... Figure 1 The network 140 is depicted in the text.

[0094] Computer 600 may also include a central processing unit (“CPU”) in the form of one or more processors 602 for executing program instructions 624. Program instructions 624 may include at least instructions for performing image processing, including automatic brightness control based on regions of interest (e.g., if computer 600 is computing system 104).

[0095] Computer 600 may include an internal communication bus 608. Computer 600 may also include a drive unit 606 (such as read-only memory (ROM), hard disk drive (HDD), solid-state drive (SDD), etc.) that can store data on a computer-readable medium 622 (e.g., a non-transitory computer-readable medium), although computer 600 may receive programming and data via network communication. Computer 600 may also have a memory 604 (such as random access memory (RAM)) storing instructions 624 for performing the techniques presented herein. However, it should be noted that in some aspects, instructions 624 may be temporarily or permanently stored within other modules of computer 600 (e.g., processor 602 and / or computer-readable medium 622). Computer 600 may also include user input and output devices 612 and / or a display 610 for connection to input and / or output devices such as a keyboard, mouse, touchscreen, monitor, display, etc. Various system functions may be implemented in a distributed manner on multiple similar platforms to distribute the processing load. Alternatively, these systems may be implemented by appropriately programming a single computer hardware platform.

[0096] The program aspect of this technology can be considered a "product" or "artifact" typically in the form of executable code and / or associated data carried or embodied on a type of machine-readable medium. "Storage" media includes any or all of the tangible memory of computers, processors, etc., or their associated modules (such as various semiconductor memories, tape drives, disk drives, etc.), which can provide non-transitory storage for software programming at any time. All or part of the software can sometimes be communicated via the Internet or various other telecommunications networks. Such communication, for example, enables the loading of software from one computer or processor into another. Therefore, another type of medium that can carry software elements includes light waves, radio waves, and electromagnetic waves, such as physical interfaces between local devices, used via wired and optical terrestrial networks, and via various air links. Physical elements carrying such waves (such as wired or wireless links, optical links, etc.) can also be considered as media carrying software. As used herein, unless limited to non-transitory tangible "storage" media, terms such as "computer or machine-readable medium" refer to any medium involved in providing instructions to a processor for execution.

[0097] While the principles of this disclosure have been described herein with reference to illustrative examples of specific applications, it should be understood that this disclosure is not limited thereto. Those skilled in the art and those who have received the teachings provided herein will recognize that additional modifications, applications, and substitutions of equivalents fall within the scope of the examples described herein. Therefore, the invention should not be considered limited to the foregoing description.

Claims

1. A method for performing automatic brightness control, the method comprising: Receive images of the target area from the imaging system of a medical device; Identify a region of interest in the image, the region of interest including physical features of the target region identified in the image; Determine the current image brightness value of the identified region of interest; as well as One or more operating parameters of the imaging system are adjusted based on the current image brightness value and the target image brightness value.

2. The method as described in claim 1, wherein, Determining the current image brightness value based on the identified region of interest includes: Determine the average pixel intensity value of a subset of pixels in the image that includes the region of interest, wherein the average pixel intensity value is the brightness value of the current image.

3. The method as claimed in claim 1 or claim 2, wherein, Identifying the region of interest in the image includes: Detect multiple edges in the image; and The region of interest is identified as the subset of pixels with the highest edge density among multiple pixel subsets in the image, wherein the pixel subset includes the physical features.

4. The method of claim 3, further comprising: The image is converted to grayscale before detecting the plurality of edges.

5. The method as claimed in claim 1 or claim 2, wherein, The image is in a first color space, and identifying the region of interest in the image includes: Convert the image from the first color space to the second color space; Generate multiple histograms for the image in the second color space; The color selection process is performed based on the analysis of one or more histograms among the plurality of histograms; and The region of interest is identified based on the color selection process.

6. The method of claim 5, wherein, The second color space includes multiple channels, and generating the multiple histograms includes: A histogram is generated for each of the plurality of channels, wherein the analyzed one or more histograms in the plurality of histograms represent the color distribution in the image.

7. The method of claim 5, wherein, Analysis of one or more histograms among the plurality of histograms includes: Identify color tones in the image; and Suspicious color regions in the image are determined based on the deviation between the identified color shadow differences and the color shadow difference patterns for the types of anatomical structures included in the image, wherein a subset of pixels in the image including the suspicious color regions is identified as the region of interest, and wherein the subset of pixels includes the physical features.

8. The method of claim 7, wherein, Identifying the suspicious color areas includes: One or more histograms in the plurality of histograms are compared with one or more reference color dark tone differentiation patterns for the anatomical structure type to identify the deviation.

9. The method of claim 7, wherein, Identifying the suspicious color areas includes: One or more of the histograms are provided as input to a machine learning model, which is trained to identify deviations from one or more learned color dark tone differentiation patterns for the anatomical structure type.

10. The method of claim 7, wherein, Performing the color selection process based on analysis of one or more histograms among the plurality of histograms includes: A mask is applied to each pixel in the image that is not included in the suspected color region to generate a masked image.

11. The method of claim 10, further comprising: A binary image is generated based on the mask image to facilitate the identification of the region of interest.

12. The method as claimed in claim 1 or claim 2, wherein, Identifying the region of interest in the image includes: Receive the feature type associated with the physical feature to be identified in the image; and Based on the feature type, the physical features corresponding to the feature type in the image are detected, wherein a subset of pixels in the image that includes the detected physical features are identified as the region of interest.

13. The method of claim 12, wherein, The feature type is a shape or pattern associated with the physical feature.

14. The method as described in any of the preceding claims, wherein, The imaging system includes a light source configured to illuminate the target region, and adjusting the one or more operating parameters of the imaging system includes: The intensity of the light emitted by the light source is adjusted to illuminate the target area, wherein the intensity of the light is adjusted by controlling the amount of current supplied to the light source.

15. The method as described in any of the preceding claims, wherein, The imaging system includes an imaging device configured to capture the image, and adjusting the one or more operating parameters of the imaging system includes: Adjust one or more of the gain or exposure time of the imaging device.

Images