Infrared Thermal Endoscopy

Abstract

Disclosed herein is the use of infrared (e.g., far infrared) and temperature detection of abnormalities within organs. Provided are systems, devices and methods for utilizing the output of far infrared detectors, particularly multiple far infrared detectors, to enhance abnormality detection during endoscopic procedures, where these components and a camera component are contained within a swallowable pill housing.

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to and the benefit of U.S. Patent Application No. 63 / 273,821, filed October 29, 2021, the contents of which are incorporated herein by reference in their entirety.

[0002] FIELD OF THE DISCLOSURE The present disclosure relates to infrared detection, and more particularly to a medical device and method that utilizes infrared imaging to measure the infrared emissions of thermally induced abnormalities during endoscopic procedures. [Background technology]

[0003] The term "endoscopy" broadly refers to the visualization of hollow organs or spaces with a scope. For example, a colonoscopy typically involves the examination of the lining of the rectum, all or part of the colon (large intestine), and even the very bottom of the small intestine (known as the ileum) for the detection of any abnormalities. During a colonoscopy, an endoscope is inserted through the anus and then slowly advanced into the rectum, colon, and sometimes the ileum. An endoscope typically includes a flexible tube for insertion (also known as an "insertion tube") and a camera with a light source at the distal end configured to capture visible light images of the GI tract as the tube is inserted and removed from the GI tract. The insertion tube can be rigid, flexible, or a combination of both, and the distal end of the insertion tube can be steerable by the user. Most endoscopes have a channel running along or through the insertion tube that allows for the injection of air, fluids, or even the insertion of instruments to aid in the visualization or manipulation of tissue. Specially adapted endoscopes exist that are shaped and sized to facilitate visualization of many organs and spaces within the human body, including, but not limited to, the small intestine, brain, paranasal sinuses, bronchi, upper digestive tract, peritoneal cavity, and joint cavities.

[0004] However, endoscopes often have a limited field of view (FOV), typically directed forward through the tube, and due to a variety of challenging conditions within the digestive tract, such as the large intestine's sharp bends, folds, twists and turns, and the constant presence of fecal matter, it is difficult to maneuver the insertion tube to image the digestive tract in a complete coverage manner.

[0005] There is also a special class of endoscopic devices that, instead of having an insertion tube, are designed to be swallowed or placed in the desired area that is difficult to reach with an insertion tube. These devices perform the same function, primarily abnormality detection. These devices are commonly referred to as "capsule endoscopes" and their use is referred to as "capsule endoscopy."

[0006] Visible light and endoscope-based visual inspection of tissues and organs has limitations. As mentioned above, many areas of the body can be difficult to access with a typical endoscope. This difficulty can arise due to the nature of the anatomy or the distances involved in traversing long hollow organs such as the small intestine. Direct visualization with an endoscope can also be limited by visible light blockage between the organ lining and the endoscopic camera, limitations of the endoscopic camera and related technology, and even subjectivity associated with the endoscopist's interpretation of images collected during the procedure. In particular, an endoscope operator may miss areas of an organ or space that contain abnormalities due to poor image resolution, the presence of tissue folds, obstructions on the tissue surface (e.g., due to unwashed waste material deposited on the tissue surface), or the fast travel speed of the distal tip.

[0007] Some endoscopic procedures have also been developed that use radiation other than visible light to detect abnormalities. For example, U.S. Patent Nos. 8,774,902 and 10,791,916, which are incorporated herein by reference in their entirety, detail the use of infrared sensors on endoscopes that detect infrared radiation passively emitted from abnormalities that have a temperature difference when compared to the surrounding tissue. However, these detection methods are difficult to implement in actual endoscopic procedures, and non-visible light alone often does not provide enough information about the abnormality to identify it. The concept of "angiogenesis" broadly associates increased heat emission (in the form of infrared radiation) with abnormalities such as tumors, and the presence of increased IR in an area can provide invaluable information to clinicians seeking to evaluate abnormalities at a stage when visible light is insufficient. Ongoing research indicates that IR may help identify cancers before they can be identified using traditional means. It has been found that many colon cancers discovered between standard endoscopies at typical 5-year intervals are due to lesions or polyps that were missed during endoscopies. Nevertheless, the implementation of IR measurements in endoscopic procedures has proven difficult and has not led to widespread adoption. There are a variety of possible reasons for this. For example, complexities associated with adequately measuring IR throughout an organ or space, whether IR parameter sets can accurately identify abnormalities during endoscopy, and linking IR information to endoscopic images make the use of infrared analysis during endoscopy challenging. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] U.S. Pat. No. 8,774,902 [Patent Document 2] U.S. Pat. No. 10,791,916 [Patent Document 3] U.S. Pat. No. 9,936,151 [Non-patent literature]

[0009] [Non-Patent Document 1] Stefanadis, C., Journal of Clinical Gastroentereology 36.3 (2003), pp. 215-218 [Non-Patent Document 2] Banic, M., Periodicum biologorum 113.4 (2011), pp. 439-444 Summary of the Invention [Problem to be solved by the invention]

[0010] The purpose of this disclosure is to describe imaging and analysis techniques for endoscopes and other visualization devices. This disclosure includes devices, software, and systems for augmenting endoscopes to increase data acquisition and enhance interpretation of acquired data compared to previously available endoscopic procedures. These techniques, when combined with endoscopic visible light image data, can enable more accurate detection of anatomy anomalies during use. [Means for solving the problem]

[0011] In accordance with the above and other objects, the present disclosure describes an endoscope, or a device attached to an endoscope, that includes a set of sensors configured to detect heat. These elements can then be coupled with additional hardware and software to capture, store, analyze, and / or interpret the data. Typically, this analysis is presented to a user, sometimes in conjunction with a visible image taken from the endoscope, to aid in diagnostic or therapeutic decision making. Often, anomalies have an increased heat signature compared to normal tissue. The devices of the present disclosure utilize this signature to add an additional parameter for anomaly detection. Without wishing to be bound by theory, collecting thermal data from multiple sensors on the device as the device moves through a hollow organ or space during a procedure results in a unique anomaly signature. The present disclosure includes methods and systems for detecting and interpreting this thermal data, optionally in conjunction with an optical image, to provide more accurate detection of anomalies and more complete imaging of anomalies during endoscopic examinations. By utilizing algorithms provided by the machine learning processes described herein, devices of the present disclosure collect and utilize unique signatures associated with multiple sensor measurements, resulting in simpler and faster anomaly detection than devices that do not use the multiple sensors and / or machine learning anomaly detection described herein.

[0012] As used herein, a distal end of a tubular portion of an endoscope is provided, the distal end comprising a distal end of an endoscope tube and a wall of the tube proximal to the distal end, the wall comprising: a) a plurality of far-infrared and / or temperature sensors distributed around the wall; b) optionally a camera disposed at the distal end for imaging objects in front of the tube (e.g., the camera's field of view includes the major longitudinal axis of the tube, which extends beyond the distal end); and a tip section including a plurality of far-infrared sensors, a temperature sensor (e.g., one or more temperature sensors), and a camera configured to communicate with each other and transmit data for processing and analysis. The endoscope may include a visible light source that reflects from an organ or space being imaged into the camera to form an optical image. In various implementations, the field of view of at least one infrared sensor includes an axis perpendicular to the wall surface or perpendicular to the longitudinal axis of the tube. In some embodiments, the field of view of each far-infrared sensor of the plurality of far-infrared sensors does not overlap with the field of view of the camera. In other embodiments, the field of view of at least one far-infrared sensor of the plurality of far-infrared sensors overlaps with the field of view of the camera. The far-infrared sensor (or a portion thereof) may be arranged linearly (e.g., along the major longitudinal axis of the tube), circumferentially or partially circumferentially around the tube wall, or a combination thereof. In some embodiments, the far-infrared sensor (or a portion thereof) may be attached to the distal end of the endoscope. In various implementations, at least one far-infrared sensor may have a field of view that overlaps with the field of view of the camera. In some embodiments, the far-infrared sensors (or a portion thereof) can be circumferentially arranged around the tube wall such that each sensor lies in a plane perpendicular to the primary longitudinal axis of the tube. In various implementations, the far-infrared sensors (or a portion thereof) can be circumferentially arranged around the tube wall such that each sensor lies in a plane that is not perpendicular to the primary longitudinal axis of the tube. In some embodiments, multiple far-infrared sensors are integrated into the tubular and / or steering portion of the endoscope so as not to increase the diameter of the tubular portion. In some embodiments, the distal end includes 1-15 far-IR sensors (e.g., 2-15 far-IR sensors, 2-10 far-IR sensors, 8 far-IR sensors).

[0013] Also provided are devices that are attached to the tubular portion of an endoscope. These devices are attachable (e.g., removably attachable) to a wall proximal to the distal end of the endoscope tube. Typically, the endoscope includes a camera located at the distal end for imaging objects in front of the tube (e.g., the field of view of the camera includes the major longitudinal axis of the tube that extends beyond the distal end), and the device for tip augmentation includes a plurality of far-infrared sensors distributed on a substrate that is attachable to a wall proximal to the distal end of the endoscope, and the plurality of far-infrared sensors, a temperature sensor (e.g., one or more temperature sensors), and the camera are configured to cooperate with each other to transmit data for processing and analysis, respectively. In various implementations, the field of view of at least one infrared sensor includes an axis perpendicular to the wall surface or perpendicular to the longitudinal axis of the tube. In some embodiments, the field of view of each of the plurality of far-infrared sensors does not overlap with the field of view of the camera. In other embodiments, the field of view of at least one of the plurality of far-infrared sensors overlaps with the field of view of the camera. The far-infrared sensor (or a portion thereof) can be circumferentially or partially circumferentially disposed around the tube wall. In some embodiments, multiple far-infrared sensors are integrated into the tubular portion of the endoscope such that the device does not increase the diameter of the tubular portion or increases the diameter or circumference of the tube by less than 20% (or 0.1%-20%) or less than 10% or less than 5%. In some embodiments, the far-infrared sensor (or a portion thereof) can be circumferentially disposed around the tube wall such that each sensor lies in a plane perpendicular to the major longitudinal axis of the tube. In various implementations, the far-infrared sensor (or a portion thereof) can be circumferentially disposed around the tube wall such that each sensor lies in a plane that is not perpendicular to the major longitudinal axis of the tube. In various implementations, multiple far-infrared and / or thermal sensors are disposed on an ingestible camera, such as a pill camera, configured to be swallowed and pass through the digestive tract and record far-infrared and thermal measurements of the digestive tract as it passes from the mouth to the anus. In other embodiments, the described sensors can be positioned and arranged to replace the visible light camera entirely.In some embodiments, the device includes between 1 and 15 far IR sensors (eg, between 2 and 15 far IR sensors, between 2 and 10 far IR sensors, 8 far IR sensors).

[0014] These devices can be attached to an endoscope, for example, that has a camera, although the endoscope can also navigate within an organ or space without the use of a camera. The housing can include a cavity into which the distal end of the endoscope can be inserted. The housing can also include a hole on a surface adjacent the cavity into which an endoscopic camera and / or a visible light source can be aligned upon insertion, such that the camera maintains functionality when the device is attached to the endoscope. In some embodiments, the device can include a light source (e.g., a visible light source, an infrared light source) and reflected light can be detected by a camera and / or multiple IR sensors.

[0015] The relative orientation and field of view of the infrared (e.g., far infrared) sensors on these devices can be variables used to gather and interpret information regarding any anomalies detected by the sensors. For example, the fields of view of at least two of the far infrared sensors overlap so that different sensors can simultaneously acquire data of an object. The sensor configuration and field of view overlap can be parameters used in the detection algorithm. In some embodiments, the overlapping fields of view can also detect device problems such as sensor failure or obstruction (e.g., obstruction due to deposition of biological material such as waste on the sensor or along the tissue surface). In some embodiments, identification of an abnormal temperature (e.g., an abnormal temperature observed through an obstruction deposited on the tissue surface) can notify the endoscope operator, for example, to remove the obstruction from the potential area of ​​interest (e.g., by cleaning) and / or visually inspect the area one or more times after redirecting the visual portion of the endoscope to a particular location (e.g., after removal of the obstruction). In some embodiments, upon identification of an abnormal temperature, the distal tip is automatically repositioned for cleaning and visual inspection. Also, for obstructions on the sensors that are difficult to remove during an endoscopic procedure, machine learning algorithms can identify when such obstructions occur and / or compensate for detection using the remaining unobstructed sensors. These machine learning algorithms can provide feedback, for example, to recalculate the full field of view from the multiple IR sensors that are unobstructed during an endoscopic procedure, or to inform the user to clean or otherwise clear the sensor area. In some embodiments, the fields of view of at least two of the multiple far-infrared sensors do not overlap and can capture heat or temperature information from adjacent organ parts. In various implementations, at least two far-infrared sensors are arranged such that a first sensor images an organ part being viewed, and then after the endoscope moves (e.g., rotates, advances, retracts), a second sensor's field of view can image the same or substantially the same organ part being viewed.In some embodiments, when the device is attached to an endoscope, a portion of the plurality of sensors (e.g., a first portion, a second portion) are linearly arranged along the wall and the linear distribution is substantially parallel (e.g., ±5°, ±1°) to the primary longitudinal axis of the tube.

[0016] In some embodiments, at least two of the plurality of sensors detect IR light of different wavelengths. For example, each sensor may independently detect far IR (e.g., light having a wavelength between 15 and 1000 μm), mid IR (e.g., light having a wavelength between 1,000 nm and 15,000 nm), or near IR (e.g., light having a wavelength between 800 nm and 3,000 nm). In some embodiments, more than 50% (e.g., 50%-100%, 60%-100%, 70%-100%, 80%-100%, 90%-100%) of the sensors on the device are capable of detecting far-IR light (e.g., light having wavelengths between 15-1000 μm, or between 15-100 μm, 100-200 μm, 200-300 μm, 300-400 μm, 400-500 μm, 500-600 μm, 600-700 μm, 700-800 μm, 900-1000 μm). In some embodiments, sensors with overlapping fields of view detect the same wavelength range, different wavelength ranges, or a combination thereof.

[0017] The endoscope tube can be a tube with a curved outermost wall, such as a cylindrical tube. In some embodiments, the cylindrical tube is an elliptical cylinder or a cylinder. In devices designed to attach to these tubes, the substrate can be sized to match the dimensions of a shape such as a circular or elliptical ring designed to fit into the wall location proximal to the distal end of the tube.

[0018] The present disclosure is based in part on the discovery that certain far-infrared sensors improve assessment of abnormal thermal signatures at physiologically relevant distances. For example, at least one of the far-infrared sensors can have a field of view of less than 25° (or between 0.1° and 25°) (e.g., less than 22°, less than 18°, less than 15°, 1° to 20°, 4° to 13°, 5° to 12°). In various implementations, at least 50% (e.g., at least 60%, at least 70%, at least 80%, at least 90%, all) of the multiple far-infrared sensors have a field of view of less than 25° (or between 0.1° and 25°) (e.g., less than 22°, less than 18°, less than 15°, 3° to 22°, 1° to 20°, 4° to 13°, 5° to 12°, 1° to 2°, 2° to 3°, 3° to 4°, 4°-5°, 5°-6°, 6°-7°, 7°-8°, 8°-9°, 9°-10°, 10°-11°, 11°-12°, 12°-13°, 13°-14°, 14°-15°, 15°-16°, 16°-17°, 17°-18°, 18°-19°, 19°-20°, 20°-21°, 21°-22°, 22°-23°, 23°-24°). In some embodiments, at least 90% (e.g., all) of the plurality of far-infrared sensors have a field of view of 5°-12° alone. In various embodiments, at least one, some (e.g., greater than 50%, greater than 60%, greater than 70%, greater than 80%, greater than 90%), or all of the multiple infrared sensors have a detection rate of at least 1 Hz (e.g., at least 2 Hz, at least 3 Hz, at least 4 Hz, at least 5 Hz, at least 6 Hz, at least 7 Hz, at least 8 Hz, at least 9 Hz, at least 10 Hz, 2 Hz to 20 Hz, 3 Hz to 20 Hz, 4 Hz to 20 Hz, 5 Hz to 20 Hz, 6 Hz to 20 Hz, 7 Hz to 20 Hz, 8 Hz to 20 Hz, 9 Hz to 20 Hz, 10 Hz to 20 Hz).

[0019] Further analysis of thermal parameters can be performed by endoscopy systems. a) an endoscope having a distal end of the present disclosure or a device of the present disclosure attached to its distal end; b) a machine-readable medium configured to receive data transmitted from a plurality of far-infrared sensors, and optionally a camera; c) a processor including instructions for analyzing the data transmitted to the machine-readable medium and identifying anomalies in an organ or a set of organs (e.g., the subject's gastrointestinal tract) based on data transmitted from the plurality of far-infrared sensors and, optionally, the associated camera (e.g., diseases such as polyps present on the tissue surface of the organ, diseases present below the tissue surface, tumors, cysts, granulomas, circulatory abnormalities, inflammation); It can be provided with: In some embodiments, the processor may include instructions for the camera AI algorithm to detect anomalies from the camera images transmitted to the machine-readable medium. For example, the processor may include instructions for the far-infrared AI algorithm to detect anomaly data transmitted from a plurality of sensors to the machine-readable medium. In various implementations, the processor includes instructions for comparing the output of the camera AI algorithm with the output of the far-infrared AI algorithm. In some embodiments, the instructions for analyzing the data to identify anomalies include calculations involving the location of the distal tip within a hollow organ or space (e.g., the digestive tract) and / or the speed of movement (e.g., forward speed, backward speed, rotation speed) of the distal tip during data collection.

[0020] Also provided is a method of observing an object using an endoscope system, comprising: a) detecting temperature data of an object by positioning an endoscope having a plurality of far-infrared sensors distributed around a wall proximal to a distal end of the tube such that the object is within a field of view of at least one of the plurality of far-infrared sensors; b) transmitting the thermal data to a machine-readable medium; and Also provided is a method comprising:

[0021] A method of observing an object using an endoscope system includes: a) providing a tip of the present disclosure having a plurality of sensors for detecting heat or temperature data, or an endoscope having a device of the present disclosure attached to the tip; b) disposing a light source capable of emitting light (e.g., white light, red light, blue light, green light, infrared light, near infrared light), the light emitted from the light source being reflected from the object into a camera and / or a plurality of far-IR sensors (or portions thereof) to form image data and / or collect reflection data at far-IR wavelengths; c) transmitting the image data to a machine-readable medium; and d) optionally, moving the distal end of the endoscope so that the object is within a field of view of one or more of the plurality of far-infrared sensors, and detecting thermal data of the object; e) transmitting the heat or temperature data to a machine readable medium; and may include.

[0022] In some embodiments, a method of observing an object using an endoscopic system includes: a) positioning an endoscope having a distal end of the present disclosure or a device of the present disclosure attached to the distal end such that an object is within the field of view of at least one of the plurality of far-infrared sensors to detect thermal data of the object; b) transmitting the thermal data to a machine-readable medium; and c) arranging a light source capable of emitting light (e.g., white light, red light, blue light, green light, infrared light, near-infrared light), such that the light emitted from the light source is reflected from an object into a camera to form image data; d) transmitting the image data to a machine-readable medium; and may include. In various embodiments, the light source is positioned by movement of the distal tip. The light source can be positioned prior to and / or after the thermal data is measured and analyzed. In some embodiments, the distal tip can be moved such that the object passes through the field of view of at least two of the plurality of far-infrared sensors. The sensor positioning can be performed before, during, or after the identification of the anomaly by the camera. In some embodiments, the analysis is performed on data acquired by both the camera and the plurality of IR sensors (e.g., the far-IR sensors) and is independent of the movement. In some embodiments, the method can include alteration of the tissue surface after the identification of the anomaly via the far-IR sensor data, such as initiating a cleaning step (e.g., releasing a liquid such as saline from the distal tip to remove deposits from the tissue surface and / or the sensor).

[0023] Typically, the method includes collecting and transmitting thermal data from each sensor during said movement to said machine-readable medium. In some embodiments, the machine-readable medium is in communication with a processor including instructions for analyzing the thermal and / or image data. In some embodiments, the thermal and / or temperature data identifies an anomaly (e.g., polyp, subcutaneous anomaly) at a location (e.g., tissue location), and a light source is positioned to reflect from the identified anomaly into a camera. In some embodiments, positioning of the light source includes positioning of both the light source and the camera (e.g., via movement of the distal tip).

[0024] The endoscopist may choose to flush any location of the tissue (e.g., with saline flowing from a nozzle proximal to the distal end) to remove obstructions and / or improve visibility of the location to be imaged. For example, in some embodiments, the method may further include flushing the location. In some embodiments, flushing may be performed before positioning the light source to observe the location of the anomaly. In another embodiment, flushing may be performed after positioning the light source to observe the location of the anomaly, and the method further includes repositioning the light source to reflect from the identified anomaly into the camera to observe the location of the flushed anomaly. Typically, these methods include transmitting and / or displaying the thermal and / or temperature and / or image data to an interface (e.g., a graphical user interface) to allow the endoscopist to observe (e.g., in real time) objects identified in the data (e.g., anomalies, flushed anomalies, potential anomalies). In some embodiments, the data may be stored and viewed at a later time (e.g., for use in machine learning algorithms described herein).

[0025] In general, embodiments of the invention include an infrared imaging device configured for side-scan infrared imaging, e.g., for medical applications, as described in U.S. Patent No. 10,791,916, which is incorporated herein by reference in its entirety. For example, in one embodiment of the invention, the imaging device includes a ring-shaped detection element including a circular array of infrared detectors configured to detect thermal infrared radiation, and a focusing element configured to focus incident infrared radiation toward the circular array of infrared detectors. [Brief description of the drawings]

[0026] [Figure 1A] 1 is a cross-sectional view of a tip of an endoscope including an array of far-infrared detectors of a particular geometry. [Figure 1B] 1 is a cross-sectional view of a tip of an endoscope including an array of far-infrared detectors of a particular geometry. [Figure 1C]1 is a cross-sectional view of a tip of an endoscope including an array of far-infrared detectors of a particular geometry. [Figure 1D] 1 is a cross-sectional view of a tip of an endoscope including an array of far-infrared detectors of a particular geometry. [Figure 1E] 1 is a cross-sectional view of a tip of an endoscope including an array of far-infrared detectors of a particular geometry. [Figure 2A] FIG. 1 is a top view of an endoscopic device including a device having multiple sensors mounted at a distal end. [Figure 2B] FIG. 1 is a front view of an apparatus and a distal end of an endoscope including an apparatus having multiple sensors mounted on the distal end. [Figure 3A] FIG. 1 is a diagram of an endoscope having multiple sensors distributed around the periphery of a housing at its distal end. [Figure 3B] FIG. 1 is a diagram of an endoscope having multiple sensors distributed around the periphery of a housing at its distal end. [Figure 4] 1 is an exemplary flow chart of an endoscopic procedure using the devices, systems and methods of the present disclosure. [Diagram 5] 1 is an example flowchart of an example machine learning procedure for constructing a computational unit used for anomaly detection. [Figure 6] FIG. 13 shows measured sensor data for various sensors as a function of distance from a heat source. [Figure 7A] FIG. 2 is a schematic diagram of a sensor measurement described in the examples. [Figure 7B] FIG. 2 is a schematic diagram of a sensor measurement described in the examples. [Figure 8A] FIG. 13 shows thermal far infrared IR measurements taken using a 5° FOV sensor when scanning the entire heating mask including holes for anomaly simulation at 0″ from the heating mask. [Figure 8B] FIG. 13 shows thermal far infrared IR measurements taken using a 5° FOV sensor when scanning the entire heating mask including holes for anomaly simulation at 1″ from the heating mask. [Figure 9A]FIG. 13 shows thermal far infrared IR measurements taken using a 12° FOV sensor when scanning the entire heating mask including holes for anomaly simulation at 0″ from the heating mask. [Figure 9B] FIG. 13 shows thermal far infrared IR measurements taken using a 12° FOV sensor when scanning the entire heating mask including holes for anomaly simulation at 1″ from the heating mask. [Figure 10A] FIG. 13 shows thermal far infrared IR measurements taken using various Melexis sensors when scanning the entire heating mask including holes for anomaly simulation at 0″ from the heating mask. [Figure 10B] FIG. 13 shows thermal far infrared IR measurements taken using various Melexis sensors when scanning 1″ from the heating mask across the heating mask including the hole for anomaly simulation. [Figure 11A] FIG. 13 shows thermal far infrared IR measurements taken using a 12° FOV Melexis sensor when scanning the entire heating mask including the hole for anomaly simulation at 0″ from the heating mask. [Figure 11B] FIG. 13 shows thermal far infrared IR measurements taken using a 12° FOV Melexis sensor when scanning the entire heating mask, including holes for anomaly simulation, at 1″ from the heating mask. [Figure 12A] FIG. 13 shows thermal far infrared IR measurements taken using a 5° FOV Melexis sensor when scanning the entire heating mask including the hole for anomaly simulation at 0″ from the heating mask. [Figure 12B] FIG. 13 shows thermal far infrared IR measurements taken using a 5° FOV Melexis sensor when scanning the entire heating mask, including holes for anomaly simulation, at 1″ from the heating mask. [Figure 13A] FIG. 13 shows an image of an artificial intestine with a resistive heater placed on the backside to simulate anomaly. [Figure 13B]FIG. 13B shows a negative control for scanning this protocol when the entire surface shown in FIG. 13A was scanned with all components at ambient temperature. [Figure 14A] FIG. 13 shows sensor measurements of 5° and 12° FOV Melexis sensors across the artificial intestine during anomaly simulation. [Figure 14B] FIG. 13 shows sensor measurements of 5° and 12° FOV Melexis sensors across the artificial intestine during anomaly simulation. [Figure 14C] FIG. 13 shows sensor measurements of 5° and 12° FOV Melexis sensors across the artificial intestine during anomaly simulation. [Figure 14D] FIG. 13 shows sensor measurements of 5° and 12° FOV Melexis sensors across the artificial intestine during anomaly simulation. [Figure 15A] 1 is an image of a curved experimental setup of artificial intestinal material. [Figure 15B] 1 is a thermal image of a curved material with a resistive heat source applied to the opposite side of the material. [Figure 16] FIG. 13 shows a far-infrared sensor scan of a curved artificial intestine with simulated abnormalities. [Figure 17A] FIG. 1 illustrates a horizontal configuration of multiple sensors for anomalies. [Figure 17B] FIG. 13 illustrates a vertical configuration of multiple sensors relative to anomalies. [Figure 18A] FIG. 13 shows sensor measurements of three sensors when scanned horizontally over a curved artificial intestine with simulated abnormalities. [Figure 18B] FIG. 13 shows sensor measurements of three sensors when scanned vertically on a curved artificial intestine with simulated abnormalities. [Figure 19A] FIG. 1 is a diagram of the sensor fixture (sensor housing and sensor array) used in the in vivo simulation experiments. [Figure 19B] FIG. 1 is a diagram of the sensor fixture (sensor housing and sensor array) used in the in vivo simulation experiments. [Figure 19C]FIG. 1 is a diagram of the sensor fixture (sensor housing and sensor array) used in the in vivo simulation experiments. [Figure 19D] FIG. 1 is a diagram of the sensor fixture (sensor housing and sensor array) used in the in vivo simulation experiments. [Figure 20A] FIG. 1 is a diagram of a sheath used in in vivo abnormality simulation experiments. [Figure 20B] FIG. 1 is a diagram of a sheath used in in vivo abnormality simulation experiments. [Figure 20C] FIG. 1 is a diagram of a sheath used in in vivo abnormality simulation experiments. [Figure 20D] FIG. 1 is a diagram of a sheath used in in vivo abnormality simulation experiments. [Figure 21A] FIG. 13 is a diagram of an attachment designed to orient a sensor fixture relative to a simulated anomaly in the intestine in an in vivo anomaly simulation experiment. [Figure 21B] FIG. 13 is a diagram of an attachment designed to orient a sensor fixture relative to a simulated anomaly in the intestine in an in vivo anomaly simulation experiment. [Figure 21C] FIG. 13 is a diagram of an attachment designed to orient a sensor fixture relative to a simulated anomaly in the intestine in an in vivo anomaly simulation experiment. [Figure 21D] FIG. 13 illustrates how the contacts between the sensor fixture, sheath and attachment work to enable sensor / anomaly alignment in these experiments. [Figure 21E] FIG. 13 illustrates how the contacts between the sensor fixture, sheath and attachment work to enable sensor / anomaly alignment in these experiments. [Figure 22] FIG. 1 illustrates an exemplary graphical user interface (GUI) used in the experiments that allows real-time data monitoring and tagging of potential anomalies based on far-IR measurement data. [Diagram 23]FIG. 13 illustrates two different thermistor measurements during thermal calibration and thermal balance of a resistor. [Figure 24] FIG. 13 shows an exemplary negative control experiment using a 5° FOV sensor. [Diagram 25] FIG. 13 shows measurements of simulated anomalies during an in vivo experiment, with the vertical line indicating the point in time when the anomaly entered the FOV of the sensor used. [Figure 26] 13 is a plot of data taken from 10 different runs at position 3. [Figure 27A] FIG. 1 shows two exemplary negative control tests using a 12° FOV sensor. [Figure 27B] FIG. 13 shows data of simulated anomalies collected using a 12° FOV sensor. [Figure 27C] FIG. 13 shows data from a 12° FOV sensor for 10 different trials at position 6. [Figure 28A] FIG. 13 shows IR data sensor data and corresponding actual anomaly locations (vertical solid lines) and estimated anomaly locations (horizontal dashed lines) based on blinded GUI users evaluating the real-time data, showing blinded measurements with a 5° FOV sensor. [Figure 28B] FIG. 13 shows IR data sensor data and corresponding actual anomaly locations (vertical solid lines) and estimated anomaly locations (horizontal dashed lines) based on blinded GUI users evaluating the real-time data, showing blinded measurements with a 12° FOV sensor. [Figure 29A] FIG. 1 compares raw data collected for far-IR simulated anomaly measurements with a machine learning based algorithm that identifies the location of anomalies (or "outliers") in the raw data. [Figure 29B] FIG. 1 compares raw data collected for far-IR simulated anomaly measurements with a machine learning-based algorithm that identifies the location of anomalies (or "outliers") in the raw data; boxed data is data identified as outliers by the machine learning algorithm. [Figure 30A]FIG. 1 shows data from 10 fully blinded studies taken from experiments performed with intraperitoneal placement of the intestine. [Figure 30B] FIG. 1 shows an exemplary fully blinded study with estimated polyp locations (vertical dashed lines) identified by a GUI operator. [Diagram 31] 13 is a plot of far-IR measurements after applying saline to the lumen prior to the measurements. [Diagram 32] 1 is a plot of far IR measurements of two trials resulting from applying feces to the sensor surface. [Figure 33A] FIG. 13 shows a plot of thermal measurements of polyps performed by a thermal sensor used during a colonoscopy, with boxed data relating to the polyps. [Figure 33B] FIG. 13 shows a plot of thermal measurements of polyps performed by a thermal sensor used during a colonoscopy, with boxed data relating to the polyps. [Figure 33C] FIG. 13 shows a plot of thermal measurements of polyps performed by a thermal sensor used during a colonoscopy, with boxed data relating to the polyps. [Figure 33D] FIG. 13 shows a plot of thermal measurements of polyps performed by a thermal sensor used during a colonoscopy, with boxed data relating to the polyps. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0027] Although detailed embodiments of the present disclosure are disclosed herein, it should be understood that the disclosed embodiments are merely exemplary of the present disclosure, which may be embodied in various forms, and the examples given in connection with the various embodiments of the present disclosure are intended to be illustrative and not limiting.

[0028] All terms used herein are intended to have their usual meaning in the art unless otherwise indicated. All concentrations are in terms of weight percent of the particular ingredient relative to the total weight of the topical composition unless otherwise specified.

[0029] As used herein, "a" or "an" means one or more than one. When used herein with the word "comprising," the words "a" or "an" mean one or more than one. As used herein, "another" means at least a second or a third or more.

[0030] All numerical ranges used herein include the endpoints and all possible values ​​disclosed between the disclosed values. All half-integer numerical exact values ​​are also contemplated as specifically disclosed and all subset limits of the disclosed ranges. For example, the range 0.1%-3% specifically discloses the percentages 0.1%, 1%, 1.5%, 2.0%, 2.5% and 3%. The range 0.1%-3% also includes subsets of the original range, including 0.5%-2.5%, 1%-3%, 0.1%-2.5%, etc. It will be understood that the sum of all percentages of individual components of a plurality does not exceed 100%, unless otherwise indicated.

[0031] In this disclosure, unless otherwise indicated, "infrared radiation" (within the context of measurements during endoscopy), "IR" (within the context of measurements during endoscopy), "thermal signature," and "thermal data" may be used synonymously. However, "thermal data" or "thermal signature" may include additional parameters related to IR measurements, such as temperature change relative to the ambient temperature, the field of view of one or more sensors, the movement of multiple IR sensors, or temperature measurements performed by other methods (e.g., by a temperature sensor).

[0032] The tip of the present disclosure comprises a distal end of an endoscope tube and a wall of the tube proximal to the distal end; a) a plurality of far-infrared sensors and / or one or more temperature sensors distributed around the wall; b) optionally a camera disposed at the distal end for imaging objects in front of the tube (e.g., the camera's field of view includes the major longitudinal axis of the tube, which extends beyond the distal end); and the multiple infrared sensors (e.g., far-infrared sensors), temperature sensor, and camera are configured to communicate with each other and transmit data for processing and analysis. In various implementations, the field of view of at least one infrared sensor includes an axis perpendicular to the wall surface or perpendicular to the longitudinal axis of the tube. In some embodiments, the field of view of each of the multiple far-infrared sensors does not overlap with the field of view of the camera. In another embodiment, the field of view of at least one of the multiple far-infrared sensors overlaps with the field of view of the camera. The far-infrared sensors (or a portion thereof) can be independently positioned circumferentially or partially circumferentially around and / or within the tube wall. In some embodiments, the far-infrared sensors (or a portion thereof) can be independently positioned around and / or within the tube wall such that each sensor lies in a plane perpendicular to the main longitudinal axis of the tube. In various implementations, the far-infrared sensors (or a portion thereof) can be independently positioned around and / or within the tube wall such that each sensor lies in a plane that is not perpendicular to the main longitudinal axis of the tube.

[0033] In some embodiments, the tip can include a ring array type structure of far infrared sensors. Such a tip can be used with an endoscopic camera and / or a swallowable camera and / or a capsule endoscope. The endoscope can include, for example, a housing on a tube, an optical fiber for transmitting optical information to and from the distal end, and / or a wire for electrical communication to the distal end including the plurality of far infrared detectors, and a lens connected to the optical fiber. The plurality of far infrared sensors and / or temperature sensors can be disposed around and / or within the tube housing and configured for communication and / or wireless transmission of data through the housing (e.g., via wires). Electrical wiring and interconnects can extend within the housing to provide power to the thermal IR ring array imager and to transmit real-time data of the thermal IR captured by the plurality of far infrared sensors, optionally along with image data from the endoscopic camera.

[0034] A device for endoscopic augmentation can be attached (e.g., removably attached) to a wall proximal to the distal end of the endoscope tube. Typically, the endoscope can include a camera disposed at the distal end for imaging objects in front of the tube (e.g., the camera's field of view includes the major longitudinal axis of the tube, which extends beyond the distal end), and the device for tip augmentation includes a plurality of far-infrared sensors distributed on a substrate, the substrate being attachable to a wall proximal to the distal end of the endoscope; The multiple far-infrared sensors, temperature sensors, and cameras are configured to communicate with each other and transmit data for processing and analysis. In various implementations, the field of view of at least one infrared sensor includes an axis perpendicular to the wall surface or perpendicular to the longitudinal axis of the tube. In some embodiments, the field of view of each of the multiple far-infrared sensors does not overlap with the field of view of the camera. In another embodiment, the field of view of at least one of the multiple far-infrared sensors overlaps with the field of view of the camera. The far-infrared sensors (or a portion thereof) can be arranged circumferentially or partially circumferentially around the tube wall. In some embodiments, the far-infrared sensors (or a portion thereof) can be arranged circumferentially around the tube wall such that each sensor lies in a plane perpendicular to the main longitudinal axis of the tube. In various implementations, the far-infrared sensors (or a portion thereof) can be arranged circumferentially around the tube wall such that each sensor lies in a plane that is not perpendicular to the main longitudinal axis of the tube.

[0035] The delivery assembly can operate in conjunction with an endoscope, such as, for example, a currently available endoscope. Another aspect of the present invention provides a delivery assembly that is easily attached to a currently available endoscope without requiring modification of such an endoscope. Another aspect of the present invention provides an endoscopic delivery assembly that is easy to use and requires minimal training of the endoscopist. Generally, these assemblies include a substrate having a plurality of far-infrared sensors disposed therein or thereabout, the substrate being dimensioned and including elements (e.g., clips, screws) for removably attaching to the endoscope.

[0036] In some implementations, the device can form a ring array type structure that can be implemented with an endoscopic device to provide real-time scanning. Such a device can be used with an endoscopic camera. The endoscope can include a housing on a tube, optical fibers for transmitting optical information to and from a distal end, and / or wires for electrical communication to a distal end that includes a plurality of far-infrared detectors, and a lens connected to the optical fibers. The ring device can fit around the housing and can be configured for wireless transmission of communication and / or data through the housing. The optical fiber bundle can pass through an opening in the substrate and can pass inside the ring-shaped detector elements. The thermal IR ring array imager is typically fixed to the front end of the endoscope. A detector circuit (not shown) can be formed on the substrate. Electrical wiring and interconnects can extend within the housing to provide power to the thermal IR ring array imager and transmit real-time data of the thermal IR captured by the plurality of far-infrared sensors, optionally along with image data from the endoscopic camera.

[0037] The infrared (IR) band typically covers a wavelength range of 700 nm to 1 mm, a frequency range of 430 THz to 300 GHz, and a photon energy range of 1.7 eV to 1.24 meV. Far infrared radiation (FIR) is found on the wavelength spectrum of 1 to 1000 μm with a frequency range of 20 to 0.3 THz, and a photon energy range of 83 to 1.2 meV. In some embodiments, each sensor in the plurality of near infrared sensors detects and measures the presence of electromagnetic radiation having a wavelength of 1 μm to 1000 μm. In some embodiments, the endoscope can further include one or more infrared sensors, including a near infrared sensor. In some embodiments, at least two of the plurality of sensors detect IR light of different wavelengths. For example, each sensor may be capable of independently detecting far IR (e.g., light having a wavelength between 15 and 1000 μm), mid IR (e.g., light having a wavelength between 1,000 nm and 15,000 nm), or near IR (e.g., light having a wavelength between 800 nm and 3,000 nm). In some embodiments, more than 50% (e.g., 50%-100%, 60%-100%, 70%-100%, 80%-100%, 90%-100%) of the sensors on the device are capable of detecting far-IR light (e.g., light having wavelengths between 15-1000 μm, or between 15-100 μm, 100-200 μm, 200-300 μm, 300-400 μm, 400-500 μm, 500-600 μm, 600-700 μm, 700-800 μm, 900-1000 μm). In some embodiments, sensors with overlapping fields of view detect the same wavelength range, different wavelength ranges, or a combination thereof.

[0038] While optical cameras can only be used to capture images of diseases that may be present on the tissue surface being examined (e.g., the gastrointestinal mucosa), IR (e.g., far-infrared) sensors can be used to capture images of diseases that may be present below the tissue surface, e.g., less than 4 mm (e.g., 0.5 mm to 4 mm) below the surface of the tissue. In practice, imaging below the tissue surface can be performed by measuring emissive thermal electromagnetic radiation, such as far-infrared radiation. Typically, far-infrared sensors can detect the thermal infrared portion of the electromagnetic heat spectrum, which has wavelengths between 2 μm and 15 μm. Also, looking at the spectrum emitted by living tissue, one can observe the radiative photons and heat generated by the biological activity of living cells. This radiative IR heat can provide a measurement of the subsurface properties by propagating through a certain thickness of tissue in the wall of the gastrointestinal tract. The magnitude of the radiative IR heat varies based on the density of the underlying cells and the type of biological activity the tissue performs. By filtering the IR thermal view for specific wavelengths (e.g., 4 μm, 11 μm), the device can separate and display tissues of different densities and thermal emissivities, thus obtaining useful data on potential abnormal conditions (e.g., tumors, cysts, granulomas, circulatory abnormalities, precancerous abnormalities, cancerous abnormalities, benign abnormalities, inflammation) that may exist in an early stage below the surface of the tissue and not be visibly exposed on the inner surface at the time of measurement. The device of the present disclosure can enable early detection of these subsurface abnormalities before any point at which the abnormality subsequently grows and acquires more cellular activity. The thermal IR imaging devices and techniques described herein can have the advantage of enabling early detection of such abnormal conditions that exist below the surface tissue.

[0039] Multiple sensors distributed on the distal end of the endoscope can include combinations of configurations described herein to expand the FOV of the endoscope. FIG. 1A is a cross-sectional view of an exemplary three-sensor array attached to an endoscope tube. In this embodiment, the endoscope tube has an outer diameter of 13.9 mm and is shown positioned within a colon with a diameter of 76 mm. The tube with these three Melexis MLX90614ESF-DCH-000-TU 12° FOV sensors has an overall diameter of 35 mm. It can be seen that the FOV of each individual sensor indicates the portion of the colon that can be measured when the array is placed in this configuration. The sensors can be oriented on the periphery of the outer circumferential surface of the endoscope tube in a plane perpendicular to the major longitudinal axis of the endoscope tube. These sensors partially circumferentially surround the tube wall (i.e., they only occupy a portion of the circumference, such as 10%-90% or 20%-80% of the circumference). These sensors can be housed in a container adapted to attach to the distal end of the endoscope. The container can be cylindrical with a diameter that increases the diameter of the inserted device and endoscope by more than 1.1 times (e.g., more than 1.5 times, more than 2 times, 1.1 times to 4 times, 1.5 times to 3 times) compared to the distal end of the endoscope itself. For example, if the container is configured to accommodate the sensor such that the sensor is fully embedded within the illustrated 35 mm diameter, the increase in diameter is 2.5 times (35 mm / 13.9 mm=2.5 times).

[0040] Different configurations of sensors can also be used to increase the total area of ​​the organ measured by multiple sensors. For example, Figures 1B-1D show a configuration in which the sensors are oriented in a linear configuration along an axis parallel to the primary longitudinal axis of the tube. Figure 1B shows a top view of a preferred linear configuration, where the field of view of each sensor is oriented tangentially or substantially tangentially to the tube wall (e.g., ±10°, ±5°, ±1° to the tangent of the tube defined at the central sensor location). As the tube moves through the hollow organ, anomalies can be introduced into each field of view and multiple measurements can be integrated to aid in detection. In this linear orientation, each FOV can be substantially parallel so that the same area is monitored by each sensor as the tube moves through the organ. Figure 1C shows an end view configuration of sensors where the sensors are arranged in a linear array, but each sensor has a rotated FOV relative to the other sensors to provide measurements of adjacent organ locations. Figure 1D shows an embodiment in which the FOV extends perpendicular or substantially perpendicular to the tangent of the endoscope tube. In various implementations, multiple far-infrared sensors (or portions of far-infrared sensors) are oriented circumferentially (or partially circumferentially) around the endoscope tube wall, linearly along the major longitudinal axis of the tube, or a combination thereof. In various implementations, the central axis of the FOV of each sensor is independently parallel, substantially parallel, perpendicular, or substantially perpendicular to the tangent of the tube at the sensor's attachment point.

[0041] FIG. 1E is a schematic diagram of an endoscope tip of the present disclosure similar to FIGS. 1A-1D with dimensions showing a Melexis MLX90614ESF-DCI-000-TU 5° FOV sensor. The multiple far infrared sensors can be selected from, for example, an Excelitas 120° FOV sensor, an Excelitas 5° FOV sensor, a Melexis 50° FOV sensor, a Melexis 35° FOV sensor, a Melexis 90° FOV (DAA), a Melexis 12° FOV sensor, and a Melexis 5° FOV sensor. In a particular embodiment, the infrared sensor is a Melexis 12° FOV sensor and / or a Melexis 5° FOV sensor. In some implementations, the sensor can be a pixel sensor array, such as an Excelitas 59×3° FOV sensor, an Excelitas 22° FOV sensor, or an Excelitas 17° FOV sensor. In various implementations, at least 50% (e.g., at least 60%, at least 70%, at least 80%, at least 90%, all) of the multiple far-infrared sensors are at an angle of less than 25° (or between 0.1° and 25°) (e.g., less than 22°, less than 18°, less than 15°, 3° to 22°, 1° to 20°, 4° to 13°, 5° to 12°, 1° to 2°, 2° to 3°, 3° to 4°, 4°-5°, 5°-6°, 6°-7°, 7°-8°, 8°-9°, 9°-10°, 10°-11°, 11°-12°, 12°-13°, 13°-14°, 14°-15°, 15°-16°, 16°-17°, 17°-18°, 18°-19°, 19°-20°, 20°-21°, 21°-22°, 22°-23°, 23°-24°). In some embodiments, at least 90% (e.g., all) of the plurality of far-infrared sensors have a field of view of 5°-12° alone.In various embodiments, at least one, some (e.g., greater than 50%, greater than 60%, greater than 70%, greater than 80%, greater than 90%), or all of the multiple infrared sensors have a detection rate of at least 1 Hz (e.g., at least 2 Hz, at least 3 Hz, at least 4 Hz, at least 5 Hz, at least 6 Hz, at least 7 Hz, at least 8 Hz, at least 9 Hz, at least 10 Hz, 2 Hz to 20 Hz, 3 Hz to 20 Hz, 4 Hz to 20 Hz, 5 Hz to 20 Hz, 6 Hz to 20 Hz, 7 Hz to 20 Hz, 8 Hz to 20 Hz, 9 Hz to 20 Hz, 10 Hz to 20 Hz).

[0042] 2A and 2B are a top view of a portion of endoscope 1 (FIG. 1A) and a front view of the distal tip and tubular portion of the vessel along the major longitudinal axis of the endoscope. Endoscope 1 includes a bending section 2 and a distal tip 5. Bending section 2 includes pivot bins 3 and 4 that allow bending of the distal tip of the endoscope during use, as well as angulation wires 8. Distal tip 6 houses a camera 7 having a field of view, shown in dashed lines, and an optical light source 8. Attached to distal tip 6 is device 10 that includes a number of far-infrared sensors, including far-infrared sensors 12 and 13. Device 10 includes a cavity sized to accommodate distal tip 6, as well as holes connected to the cavity to allow subsequent operation of camera 7 and light source 8. Device 10 receives power and transmits data through wires 15 and 16. Wires 15 are connected to the device and extend outside the endoscope, while wires 16 extend inside the endoscope. In some embodiments, multiple far-IR sensors receive power and transmit measurements via wires that run inside the device. In some embodiments, the device includes a battery that powers the far-IR sensors. In some embodiments, the far-IR sensors transmit data wirelessly (e.g., via Bluetooth, wireless RFID communication algorithms).

[0043] FIG. 2A shows a series of far IR sensors arranged circumferentially around the distal tip, including far IR sensor 12 and its corresponding field of view (FOV). FIG. 2B shows an internal cross-sectional location of the sensors, including axis A, illustrating a circumferential arrangement of twelve far IR sensors, such as sensor 12, surrounding the entire circumference of the distal tip. As shown, each IR sensor on device 10 is embedded within the device housing, with minimal impact (e.g., less than 5% diameter change) on the exterior surface and configuration of the housing. FIG. 2A also shows three sensors arranged linearly along the major axis of the device.

[0044] The far-IR sensor 12 has a central axis of field of view oriented perpendicular or nearly perpendicular (e.g., 80°-100°) to the principal longitudinal axis of the distal tip 6. The far-IR sensor 13 has a central axis of field of view that is not oriented perpendicular (e.g., at an angle of 10°-80°, or 20°-70°) to the principal longitudinal axis of the distal tip 6. The field of view of each sensor may be individually overlapping or non-overlapping with any other sensor in the device. In general, overlapping and non-overlapping configurations of sensors, such as adjacent sensors, or sensors and cameras, allow the device or endoscope to take multiple measurements when needed and use these individually (e.g., by moving the distal tip to different measurement positions as determined by a machine learning algorithm trained on the particular geometry of the far-IR sensor) to reduce false positive detection rates. As shown, the FOV of the far-IR sensor 12 does not overlap with the FOV of the far-IR sensor 13. Meanwhile, the FOV of far-IR sensor 13 overlaps with the FOV of camera 7 at portion 14. In certain embodiments, all far-IR sensors with overlapping fields of view have the same detection frequency. In other embodiments, at least two far-IR sensors with overlapping fields of view have different detection frequencies. The machine learning algorithms described herein can utilize these various parameters to identify anomalies.

[0045] 3A and 3B show an endoscope 30 having a flexure 32 controlled by an angulation knob 34 that moves an angulation wire 36 via a chain and sprocket. Movement of the flexure 32 allows movement of a distal tip 38 as the endoscope moves through an organ or space under examination. The distal tip 38 includes a number of far-IR sensors, including far-IR sensors 40, 42, and 44. The distal tip 38 also includes a camera 46. As shown, the FOV of the sensor 42 and the field of view of the camera 46 overlap. In the illustrated embodiment, a portion of the far-IR sensors (e.g., far-IR sensors 40 and 44) ​​are embedded within the distal tip housing of the endoscope. Another portion of the far-IR sensors (e.g., far-IR sensor 42) is located on the wall of the distal tip.

[0046] In both the endoscope and endoscope-mounted device embodiments of the present disclosure, each far-IR sensor can be embedded separately in or on a wall of the distal housing or device housing, hi some embodiments, each far-IR sensor is embedded in a wall of the distal housing or device housing.

[0047] The disclosed endoscopic systems and methods can perform analysis of additional measured thermal or temperature parameters. These endoscopic systems and methods include: a) an endoscope having a device of the present disclosure at or attached to a tip thereof; b) a machine-readable medium configured to receive data transmitted from the plurality of far-infrared sensors and the camera; and c) a processor including instructions for analyzing the data transmitted to the machine-readable medium to identify anomalies in an organ or set of organs (e.g., the subject's gastrointestinal tract) based on the data transmitted from the plurality of far-infrared sensors and, optionally, on the images provided by the camera (e.g., diseases such as polyps present on the tissue surface of the organ, diseases present below the tissue surface, tumors, cysts, granulomas, circulatory abnormalities, inflammation); may include. In some embodiments, the processor may include instructions for the camera AI algorithm to detect anomalies from the camera images transmitted to the machine-readable medium. For example, the processor may include instructions for the far-infrared AI algorithm to detect anomaly data transmitted from a plurality of sensors to the machine-readable medium. In various implementations, the processor includes instructions for comparing the output of the camera AI algorithm with the output of the far-infrared AI algorithm. In some embodiments, the instructions for analyzing the data to identify anomalies include calculations involving the position of the distal end in the GI tract and / or the speed of movement of the distal end during data collection (e.g., forward speed, backward speed, rotation speed).

[0048] Temperature and / or far-IR sensors can also be used to collect thermal data regarding abnormalities within an organ or space (e.g., the digestive tract). For example, a temperature sensor can be contacted (e.g., by an endoscope, swallowable camera) with tissue (e.g., the digestive tract) to record abnormal temperatures compared to the surrounding tissue. For example, differences as low as 0.1°C or 0.5°C or 1°C (or up to 10°C) (e.g., 0.1°C-5°C, 0.1°C-2°C, 1°C-2°C, 1°C-1.5°C) can be used as benchmarks to identify abnormalities identified, for example, by the movement of the endoscope tip or the movement of a swallowable pill through the digestive tract. It will be understood that temperature differences that can be converted to Celsius temperature differences are also within the scope of the present disclosure. For example, a difference (e.g., 0.1°C to 5°C, 0.1°C to 2°C, 1°C to 2°C, 1°C to 1.5°C) as low as 0.1°C or 0.5°C or 1°C (or up to 10°C) relative to the raw output of one or more IR sensors (e.g., far-IR sensors) and / or temperature sensors (e.g., thermocouples) can be measured and analyzed. The system of the present disclosure can use this location as a potential location for further review or anomaly detection based on comparison with surrounding tissue as the temperature sensor moves through the GI tract. Temperature sensors of the present disclosure can include sensors capable of measuring temperature differences upon contact with a surface. Among other things, some temperature sensors can have a resolution of less than 0.5°C. For example, the temperature sensor can be a thermistor, a resistive temperature type detector, or a thermocouple (e.g., a T-type thermocouple). Devices utilizing these sensors can be used for diagnostic purposes and / or algorithm training purposes on swallowable cameras, for example, during colonoscopy.

[0049] The subject matter disclosed herein may be implemented in software or firmware, or a combination thereof, or as instructions stored on a machine-readable medium that may be read and executed by one or more processors. A machine-readable medium may include any medium and / or mechanism for storing or transmitting information in a form readable by a machine (e.g., a computing device). For example, a machine-readable medium may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media, optical storage media, flash memory devices, electrical, optical, acoustic, or other forms of propagated signals (e.g., carrier waves, infrared signals, digital signals), and the like.

[0050] Examples of hardware elements may include processors, microprocessors, circuits, circuit elements (e.g., transistors, resistors, capacitors, and inductors), integrated circuits, application specific integrated circuits (ASICs), programmable logic devices (PLDs), digital signal processors (DSPs), field programmable gate arrays (FPGAs), logic gates, registers, semiconductor devices, chips, microchips, and chipsets. In some embodiments, optionally in combination with any of the embodiments described above or below, one or more processors may be implemented as a complex instruction set computer (CISC) or reduced instruction set computer (RISC) processor, an x86 instruction set compatible processor, a multi-core, or any other microprocessor or central processing unit (CPU). In various implementations, one or more processors may be dual-core processors, dual-core mobile processors, and the like.

[0051] Examples of software may include a software component, a program, an application, a computer program, an application program, a system program, a machine program, operating system software, middleware, firmware, a software module, a routine, a subroutine, a function, a method, a procedure, a software interface, an application program interface (API), an instruction set, computing code, computer code, a code segment, a computer code segment, a word, a value, a symbol, or any combination thereof.

[0052] One or more aspects may be implemented by representative instructions stored on a machine-readable medium that, when read by a machine, represent various logic within a processor, which instructions cause the machine to produce logic for performing the techniques described herein. Such representations, known as "IP cores," may be stored on tangible machine-readable media and provided to various customers or manufacturing facilities to be loaded into fabrication machines that actually produce the logic or processors.

[0053] Some embodiments of the system include a computer-implemented method for training an artificial intelligence (AI, which as used herein may include any type of machine learning) to recognize any combination of images and data from the far-infrared sensor as above a threshold for anomaly identification. Some embodiments of the system include at least a portion of one or more computers including one or more processors and one or more non-transitory computer-readable media storing instructions for implementing at least a portion of each embodiment described herein. In some embodiments, the non-transitory computer-readable media includes instructions for executing a method for training an AI to recognize any combination of images and text ("text" as used herein includes any one or combination of one or more characters including letters, numbers, language symbols, and / or words) and identify a threshold for anomaly identification. In some embodiments, the system includes independently operating AIs, such as an AI for interpreting camera images, an AI for interpreting information read from multiple far-infrared sensors (or portions thereof), and an AI for combining information received from both data acquisition devices.

[0054] Image data may be processed by a camera capturing regular images while the endoscope is moving through the human GI tract. These regular images may be co-located to determine missed or poorly imaged regions in portions of the human GI tract that have already been passed by the endoscope. If missed or poorly imaged regions are detected, information regarding any missed or poorly imaged regions may be provided (e.g., by repositioning the camera to acquire images, or by interpolating the missed or poorly imaged data with thermal data from multiple IR and / or temperature sensors). In some embodiments, missed images may be cross-correlated to identify based on measurements received from multiple far-IR sensors. For example, if an anomaly is detected by data collected from multiple far-IR sensors but not from the camera, the distal tip may be repositioned to acquire camera image data of the region having an abnormal temperature. In some embodiments, the difference in anomaly detection between the far-IR sensor and the camera may be transmitted to the endoscope operator who repositions the distal tip. In some embodiments, the difference in anomaly detection causes the distal tip to be automatically repositioned.

[0055] The method may further include receiving from the camera a structured light image captured by the camera while the endoscope is moving through the human gastrointestinal tract, which is associated with the normal image, and deriving distance information of the normal image based on the structured light image. The normal image may be normalized according to the distance information and the optical magnification information of the normal image to facilitate editing of the normal image. The distance information may be used to determine whether a target region in a normal image is out of focus or in focus, and if the target region is out of focus in all normal images covering the target region, the target region is determined as a missed or underimaged region.

[0056] In some embodiments, the endoscope, or a device attached to the endoscope, may further include a motion sensing device that measures the movement of the tube within the human digestive tract. For example, the motion sensing device may be an accelerometer, or at least a gyrator configured to transmit data to a computer-readable storage medium. The motion sensing device may be used to determine the movement of the tube, the trajectory of the tube, the orientation of the tube, or any combination thereof. These parameters may be used in anomaly detection analysis. For example, an increase in the speed of movement may be correlated with an increase in the slope on the thermal data obtained from a single sensor. The integration of these multiple components taken independently from multiple far-infrared sensors and / or cameras may be used to identify appropriate thresholds for anomaly detection (e.g., via machine learning algorithms employed).

[0057] The endoscope may also be equipped with structured light to capture normal images along with structured light images (SLI). These SLIs are used to derive distances of different regions in the image, allowing stitching to be more accurate by using 3D information of the imaged regions. U.S. Pat. No. 9,936,151, which is incorporated by reference in its entirety, discloses a camera equipped with normal and structured light to capture both normal and structured light images using the same image sensor. The structured light image and the corresponding normal image are captured close in time such that motion between them is expected to be small. Thus, distance information derived from the structured light image can be correlated with the normal image. U.S. Pat. No. 9,936,151, which is incorporated by reference, discloses details regarding camera designs with SLI capture capabilities. In some implementations, the structured light is formed at a wavelength sufficient to reflect off the tissue surface and be detected by at least one of the multiple IR sensors.

[0058] The endoscope system of the present disclosure is capable of evaluating abnormalities such as lesions in biological tissue within an organ. an endoscope configured to image biological tissue within an organ; a plurality of far-infrared sensors for deriving thermal data from the organ; a processor including an evaluation unit configured to process the plurality of captured images and thermal data of the biological tissue to evaluate and / or identify the presence of an anomaly and / or evaluate and / or identify an anomaly (e.g., a type of anomaly) in the organ; may include. For example, the evaluation unit may be an image evaluation value calculation unit configured to calculate an image evaluation value indicative of an intensity of abnormality of the tissue in each of the plurality of tissue images from the pixel evaluation values ​​of each of the plurality of tissue images; a heat quantity calculation unit configured to calculate a heat intensity anomaly indicative of an intensity of an anomaly of the biological tissue in each image from a deviation (e.g., increase) of heat and / or temperature compared to a baseline or a heat threshold; an imaging position information acquisition unit configured to acquire, in association with each image, information on an imaging position within the organ at which each image was captured; an anomaly location calculation unit for determining the presence or absence of anomalies and / or thermal anomalies in each image based on whether the image evaluation value exceeds a predefined threshold and / or thermal threshold; a data correlation calculation unit configured to compare results from the image evaluation unit and the heat value calculation unit to obtain a start position and an end position of an area of ​​anomaly; and an organ lesion evaluation unit configured to set a length of a lesion portion to be evaluated from a start position and an end position as degree information of the abnormal portion, and to evaluate a degree of abnormality in the organ using the position and degree information and a representative value of an image evaluation value of a captured abnormal image acquired by imaging the lesion portion; may include one or more of: When an anomaly exists at multiple locations, the anomaly location calculation unit can use the location having the maximum depth length where the lesion extends continuously as the lesion portion to be evaluated. The length can be calculated based on the deviation of the temperature data, optionally in conjunction with one or more fields of view of the far-infrared sensor and / or the moving speed of the endoscope.

[0059] Also provided are systems for training an artificial intelligence to recognize the presence of anomalies. These systems include: The present invention may include one or more computers including one or more processors and one or more non-transitory computer readable media storing instructions that, when executed, cause the one or more processors to import far-infrared sensor data and, optionally, images from an endoscopic camera of the present disclosure, the processor: Identifying one or more anomalies based on threshold image data (e.g., captured from a camera); Identifying one or more anomalies based on threshold thermal data (e.g., captured from a plurality of far-IR sensors and / or temperature sensors); comparing the thermal data and the image data to identify whether both systems return a positive identification; rebalancing the thresholds based on a positive or negative correlation in the comparison; It can be configured as follows. In various implementations, the rebalancing of the thresholds may involve a positive identification of an anomaly, for example, by the person performing the endoscopy. Machine learning may be trained to recognize the presence of an anomaly by supervised learning, unsupervised learning, reinforcement learning, and / or ensemble learning. Algorithms utilized by the artificial intelligence may include anomaly detection such as linear, logistic, K-means, isolation forest, neural nets (especially for reinforcement learning), KNN, decision trees, random forests, SVMs, naive Bayes, or combinations thereof. In some embodiments, the machine learning algorithm (e.g., anomaly detection such as isolation forest) is selected to result in a proportion of outliers (or false positive rate) of greater than 20% (e.g., greater than 30%, greater than 40%, greater than 50%, 20% to 60%). The machine learning algorithm (and corresponding real-time analysis) can be selected based on the detection rate of the far-IR sensors in the device (e.g., each far-IR sensor can have a detection rate of at least 1 Hz (e.g., at least 2 Hz, at least 3 Hz, at least 4 Hz, at least 5 Hz, at least 6 Hz, at least 7 Hz, at least 8 Hz, at least 9 Hz, at least 10 Hz, 2 Hz to 20 Hz, 3 Hz to 20 Hz, 4 Hz to 20 Hz, 5 Hz to 20 Hz, 6 Hz to 20 Hz, 7 Hz to 20 Hz, 8 Hz to 20 Hz, 9 Hz to 20 Hz, 10 Hz to 20 Hz)). In some embodiments, multiple machine learning algorithms can be utilized (e.g., in a device including different sensors with different detection frequencies).

[0060] Machine learning algorithms can interpret the raw data output from the multiple sensors in real time to identify anomalies as the distal tip of the endoscope passes through various locations within the organ or space under examination. FIG. 4 illustrates an exemplary flow chart of an endoscopic procedure and system using multiple far-IR sensors as described herein. At step 100, a measurement procedure can be initiated. Step 100 can include any location of the distal tip as it moves through the organ or tissue. Step 100 can be initiated by an operator at some point during the process (e.g., as indicated by some visual marker) or can be initiated as soon as a baseline ambient temperature of the tissue is established (e.g., as measured by one or more far-IR and / or temperature sensors). In some embodiments, each "location" can include an aggregation of multiple measurements from each far-IR sensor. For example, a location can be considered an aggregation of 1-100 times the detection rate of data from at least one far-IR sensor. For example, if the computation unit is processing data from a far IR sensor with a detection rate of 1-10 Hz, each position may include measurements from 0.1 s to 100 s (e.g., 0.1 s to 50 s, 0.1 s to 20 s, 0.1 s to 10 s, 1 s to 10 s).

[0061] In step 102, the system may collect and structure far-IR sensor data. In some embodiments, in step 108, the system obtains far-IR sensor data directly from the far-IR sensor and transmits the raw data to a processor-accessible readable medium that includes instructions for anomaly detection. In some embodiments, in step 108, the data (e.g., data from multiple sensors, and data from multiple non-identical sensors with different fields of view and / or detection rates, etc.) is collected and structured for easy analysis by the processor. Optionally, in step 104, the endoscope may also collect and structure distal tip position information, such as speed of movement and rotation (e.g., as measured by an accelerometer). In some embodiments, in step 106, the system may also optionally collect and structure image data from an endoscope camera. Step 102, and optional steps 104 and 106, may each be transmitted to a processor in step 108 for analysis, interpretation, and identification of anomaly location. In some embodiments, in step 110, the field of view of multiple infrared sensors and optionally the camera is collected and transmitted to a processor for analysis of real-time data. In some embodiments, the computation unit is trained on a particular geometry of multiple far-IR sensors and optionally cameras, and step 110 is not required. The computation unit then interprets the data sent to the processor in step 112 using artificial intelligence and / or machine learning to identify the likelihood that an anomaly exists at the endoscope position. The computation unit identifies whether the endoscope position has a high likelihood of anomaly (114) (e.g., the computation unit detects a false positive rate of less than 20% (or between 0.1% and 20%), such as when there is a large deviation in the far-IR signal across multiple subsequent measurements and positions), a low likelihood of anomaly (116) (e.g., the computation unit detects a false positive rate of more than 80% (i.e., between 80% and 100%), or an indeterminate likelihood (e.g., between 20% and 80%) (118).

[0062] If the likelihood is high (114), the system may notify the endoscope user that an anomaly exists in step (120). For example, the computing unit may calculate and overlay the far-IR sensor data on the optical camera image and display it on the user's screen to receive a notification of the unidentified anomaly, and optionally provide the user with options such as cleaning the area of ​​the unidentified anomaly by orienting another set of sensors and / or camera in another location to observe the anomaly or to continue the endoscopic procedure in step 122, and repositioning the camera to further observe the anomaly (e.g., after cleaning the anomaly). If the likelihood is low (116), the endoscope is moved to the next location for inspection in step 122. If the likelihood is between the thresholds, the endoscope 122 may be repositioned to take measurements again and remeasure the potential anomaly in step 122. This remeasurement may involve measuring the same parameters as the previous measurement, or may involve a new measurement of the tissue (e.g., by overlaying a different far-IR set of sensors and / or camera field of view with the potential anomaly, cleaning the area to identify potential lesions before measuring). This data can be sent to a processor configured with machine learning and / or artificial intelligence algorithms in optional step 124 to allow real-time learning from these possible events and integration into subsequent calculations (which can occur during or after the endoscopic procedure). For example, if repeated measurements are made over the same location and ultimately a positive identification is made, the computation unit can integrate this information to allow similar data situations to be associated with that anomaly. As a result, the system is ready to proceed with measurements of new locations of the organ or space. In some embodiments, treatment can be performed based on the anomaly. For example, the endoscope (controlled by the user) can cut and / or collect some or all of the tissue (e.g., to remove the anomaly for further analysis after the endoscopy).In some embodiments, the endoscope (controlled by the user) can mark tissue at a location, for example by injecting a substance such as a fluorescent or chemiluminescent substance into the tissue at the location so that the location of the abnormality can be found at a later time, or administer a substance to aid in the treatment of the abnormality.

[0063] The computational unit may utilize algorithms developed by machine learning or artificial intelligence. Specifically, in embodiments where the computational unit requires optical imagery and far-IR sensor data, artificial intelligence is used (e.g., to correlate anomalies in the far-IR with locations on the optical image). The artificial intelligence may be a neural network (e.g., deep learning, deep convolutional, or recurrent neural network) that includes a set of neurons. Neurons are architectural elements used in data processing and artificial intelligence, particularly in machine learning of the weights of the inputs (e.g., data from one or more far-IR sensors and their corresponding geometry on the distal end, optical images) provided to a given neuron. As used herein, each neuron may be configured to accept a predefined number of inputs from other neurons in the neural network and provide outputs related and sub-related to the content of the optical image being analyzed in cooperation with the far-IR sensor data. Individual neurons may be strung together and / or organized in various configurations of neural networks to provide interaction and relationship learning modeling of how each optical image and the far-IR sensor data are related to each other.

[0064] For example, a neural network node functioning as a neuron may include multiple gates that process input vectors (e.g., a cross-section of an image, a vector of far-IR sensor data, a structured data array from multiple far-IR sensors, a geometry of the far-IR sensor), a memory cell, and an output vector (e.g., a contextual representation for display to an endoscope user or storage for subsequent examination). Typically, the input and output gates control the information flowing into and out of the memory cell, respectively. The weights and bias vectors of the various gates may be adjusted through a training phase, and once the training phase is complete, these weights and biases may be compiled for normal operation (or adjusted on the fly based on continued training). These neurons and neural networks may be constructed programmatically (e.g., via software instructions) or via dedicated hardware that connects the neurons together to form a neural network.

[0065] Neural networks can utilize features (e.g., patterns in images and / or far-IR sensor data) to analyze data and generate an assessment. These features can be individual measurable characteristics of the phenomenon under observation (e.g., increased temperature of an anomaly, changes in the surface of tissue). The concept of features is often related to the concept of explanatory variables used in statistical methods such as linear regression. Furthermore, deep features represent the output of nodes in the hidden layer of a deep neural network that can be identified during artificial intelligence training.

[0066] Control commands for the endoscope can also be generated based on a computer analysis of the data collected by the far-IR sensor. The computer analysis can include using AI / ML techniques for more accurate and precise control command generation and determining whether the endoscope (or device attached to the endoscope) is operating according to expected conditions / parameters. For example, data can be collected over time regarding the force and velocity of the distal tip or other time information. This data can be used to train one or more models to accurately and precisely generate control commands for the distal tip during surgery (e.g., to reposition, clean, provide surgical options, etc.).

[0067] The one or more models described herein can be any type of machine learning model, including but not limited to a correlation model, ordinary least squares, convolutional neural network (CNN), or supervised machine learning. Different models can be generated for different procedures / operations. For example, the artificial intelligence and machine learning model can be trained using YOLO (e.g., YOLOv2) or Detectron algorithms. Different models can also be generated for endoscopes with different types of sensors and distal tip geometries, or different capabilities. In some implementations, different models can also be generated for different patient profiles, demographics, or other patient information.

[0068] FIG. 5 is an exemplary flow chart of an ML / AI training algorithm of the present disclosure that utilizes data provided from multiple far-IR sensors and, optionally, an endoscope camera. Typically, a training session begins with collecting and structuring data from one or more far-IR sensors in step 202, optionally collecting data related to distal tip position (step 204) and camera images (step 206) (200). This data is transmitted to a readable medium accessible to a processor containing the machine learning algorithm. In some embodiments, the geometry of the multiple far-IR sensors is also provided to the readable medium. This data can be obtained from procedures (e.g., procedures in which a user directly forms a positive correlation between anomalies and far-IR sensor data) and / or synthetic data generated to help train the algorithm based on analysis of smaller data sets. The machine learning algorithm can interpret these data sets in step 212 to identify variables that enable threshold detection of anomalies in step 214. These variables can include temperature differences from the ambient temperature measured by the far-IR sensors, temperature differences associated with certain endoscope distal tip movements, and temperature differences correlated with optical images from the camera. A machine learning algorithm can leverage this information to shape the computational unit used in the endoscopic procedure. In some embodiments, the information from the computational unit can be reused in subsequent training sessions to further analyze new data sets based on previously identified data correlations. In some embodiments, the machine learning algorithm can learn during the endoscopic procedure to provide real-time changes to anomaly detection weights and correlations in the computational unit. EXAMPLES

[0069] The following examples illustrate specific aspects of the present invention, but should not be construed as limiting, as these examples are merely intended to provide a more specific understanding and practice of the embodiments and various aspects thereof.

[0070] Example 1: Sensor evaluation Multiple far-infrared sensors were tested to confirm their ability to consistently detect temperature at various distances from the heat source within the anatomical range. For the evaluation, a square blackbody radiator was placed in a test booth and set to produce a temperature 2°C higher than the ambient. The test booth was used to mimic physiological conditions by preventing ambient temperature fluctuations. An IR camera was used to ensure uniformity across the blackbody radiator.

[0071] The measurements from each sensor were compared to the thermocouple measurements. The sensors tested were Exceletis and Melexis sensors with fields of view (FOV) of 120°, 90°, 50°, 35°, and 5°. Figure 6 shows the accuracy (as a percentage difference from the thermocouple measurements) of the tested sensors, each evaluated at 0° and from 0 to 1 inch away from the blackbody radiator surface.

[0072] A plastic mask with a 5 mm hole was placed over a blackbody radiator and heated until uniform. The heating element was set at 26.0°, resulting in a uniform mask temperature of 24.1°.

[0073] To see the temperature variation related to the mask geometry, multiple sensors were moved across the face of the mask and past the holes. A diagram of this experimental protocol is shown in Figures 7A and 7B. The sensor was moved across the mask such that the holes were within the sensor FOV at several points during the movement. Temperature readings were monitored during the movement. Figure 8A shows the temperature readings when the Excelitas 5° FOV sensor was moved right next to the mask ("0' from the heating element"), and Figure 8B shows the readings when it was 1 inch away from the mask ("1' from the heating element"). Figure 9A shows the measurements when the Excelitas 120° FOV sensor was placed on the mask, and Figure 9B shows the measurements when the sensor was placed 1" away from the mask. The ambient temperature is also shown in these figures. As can be seen, an increase in temperature change can be seen as the sensor is moved across the holes. However, at 1" away from the mask, the 120° FOV was unable to detect any thermal variations. Also visible in FIG. 8B is an artifact due to heat leak between the heating element and the mask, causing a higher temperature to be measured.

[0074] The Melexis sensors were also measured and measurements are shown in Figure 10A (0 inches from the mask) and Figure 10B (1 inch from the mask). As shown, each sensor measures the mask differently at these locations and the measured temperature, location and size of the hole are affected by changes in FOV and sensor type. Figures 11A-11B show similar measurements for a 12° FOV Melexis sensor and Figures 12A-12B show measurements for a 5° FOV Melexis sensor. These sensors were able to accurately and consistently identify the shape and size of the hole at the anatomical limits tested.

[0075] Example 2: Sensor evaluation on an artificial intestine A resistor was placed under the artificial intestine to simulate sub surface temperature measurements. A negative control experiment was performed in which a sensor was moved across the artificial intestine while all temperatures were held at ambient temperature. FIG. 13A shows an annotated image of the experimental setup showing the resistor location under the artificial intestine and the power connections to the resistor, and FIG. 13B shows measurements from the negative control. FIGS. 14A-D show measurements of the 5° FOV and 12° FOV sensors when the sensor was placed across the artificial intestine at 0 and 1 inch from the mask. The ambient temperature for these measurements was 24.0° C., and the resistor temperature was ~26.0° C. The wider distribution compared to the mask holes in Example 1 is likely due to the thermal transmittance of the artificial intestine compared to the plastic. Nevertheless, the sensors were able to discern thermal fluctuations due to sub surface temperature fluctuations.

[0076] To more closely mimic the digestive tract, the artificial intestine was oriented in a curved shape, as shown in Figure 15A. A resistor was placed behind the curved artificial intestine. Figure 15B shows a heat map measured by an IR camera. Sensor measurements were taken along the longitudinal axis of the curve at different resistor temperatures (relative to ambient temperature) at various locations of the curved artificial intestine, and the 12° and 5° FOV sensor measurements are shown in Figure 16.

[0077] These measurements indicate that only certain sensors can provide accurate thermal measurements over the anatomical distances typically required for endoscopic measurements. Specifically, the 5° and 12° FOV sensors were highly accurate over the measured distances, and each provided some signal in all test conditions. Similar experiments can be performed to validate machine learning algorithms and train them to accurately identify the presence of anomalies, especially subsurface anomalies.

[0078] Example 3: Integrating Multiple Sensor Measurements Three identical Melexis 5° FOV sensors were placed on a substrate and moved across a curved artificial intestine with a resistor on one side to simulate a subcutaneous anomaly. The sensors were arranged on the substrate in linear rows so that scanning could be performed in a horizontal (FIG. 17A) or vertical (FIG. 17B) orientation relative to the anomaly. "Horizontal" and "vertical" in this context refer to the linear configuration of the sensors relative to the motion. A "horizontal" configuration can be thought of as analogous to a circumferential arrangement of sensors along a tube, and a "vertical" configuration can be thought of as analogous to a linear arrangement of sensors along the major longitudinal axis of the tube.

[0079] Thermal data was collected for sensors placed at each orientation and is shown in Figure 18A (horizontal orientation) and Figure 18B (vertical orientation). When the three sensors pass over the anomaly in the horizontal position, the anomaly can be measured as a whole, albeit at different locations on the substrate. In contrast, when the three sensors pass over the anomaly in the vertical position, a range of temperature fluctuations is identified. However, the center sensor (sensor 2) shows a larger thermal fluctuation compared to the end sensors (sensors 1 and 3) which record the same response.

[0080] Example 4: Animal Experiments Three custom fixtures were designed and used to collect data during animal testing. The studies were designed to determine whether the far-IR sensor could consistently detect simulated heat spots that were 2°C warmer than ambient in living animal tissue at multiple locations. The studies were also designed to characterize sensor performance in conditions representative of colonoscopy. For example, the studies could evaluate: 1) Use of an IR sensor to identify simulated polyps at a fixed distance of 0.1” (0.254 cm) (this distance was chosen to eliminate variability that may arise in experiments due to the proximity dependence of far IR); 2) Use of multiple sensors with different fields of view (FOV); 3) A controlled, blinded study to identify the heat source by observing the far-IR data on a graphical user interface (GUI) without direct visualization of the animal, the fixture, or the operator while the sensor-equipped fixture was moving along a controlled path to identify the beginning and end of a 2°C temperature difference (Δ or delta); and 4) Evaluation of sensor function following application of stressors that may be encountered during colonoscopy, including intentional cleaning of the intestinal surface with 0.9% saline, fecal matter placed on the sensor, and an air / CO2 insufflation environment.

[0081] Three custom fixtures were designed and used to collect data during animal studies.

[0082] Sensor Fixture and Design Prior to animal testing, a housing for the sensor array was designed, assembled, and leak tested. The housing held the sensors in a fixed, known position and provided protection from the biological tissue environment. The sensor fixture was designed in SolidWorks and fabricated in clear resin using a FormLabs Form3 3D printer. A UV-curable adhesive was used to secure the hardware within the housing and seal the housing. Figures 19A and 19B show a perspective view and a cross-sectional view, respectively, of the housing.

[0083] Figure 19C (top view) and Figure 19D (side view) show top and side views of the sensor housing, identifying the location of the saline spray tube, leak test tube, the thermistors (thermistor 1 and thermistor 2), the far IR sensor array, magnet, temperature and humidity sensors, and the relative locations of the cabling holes.

[0084] A saline spray tube was used to apply saline from a syringe outside the intestine to the intestinal wall.

[0085] Thermistor 1 protruded slightly above the surface of the housing to ensure good contact with the tissue as it slid through the colon. This thermistor reads the surface temperature of the tissue, and therefore thermistor 1 was placed directly under the heat source (a resistor attached to the exterior of the intestine) so that at the start of each test at a new location, the temperature of the heat spot was set and adjusted to an appropriate temperature above the baseline tissue temperature. Thermistor 2 serves the same purpose as thermistor 1, but measures the baseline tissue temperature. This difference (i.e., the delta between thermistor 1 and thermistor 2) is the temperature difference that is investigated in every experiment. A magnet was used to align the heat source with thermistor 1 at a position below the resistor contact.

[0086] During the experiment, a leak test tube was connected to a pressure gauge and a syringe to identify leaks in the housing.

[0087] Two types of far-IR sensors were evaluated: a 5° field of view (FOV) sensor (Melexis part number MLX90614ESF-CDI-000-TU), and a 12° FOV (Melexis part number MLX90614ESF-CDH-000-TU).

[0088] A sheath was developed that was inserted into the colon to guide the sensor immobilizer as it moved through the colon during the experiment. Figures 20A and 20B show perspective and cross-sectional views, respectively. Figure 20C shows a perspective view of the housing and sensor positioned within the sheath, and Figure 20D shows a front view of the housing and sensor within the sheath. The sheath is elliptical in cross-section, limiting the freedom of the housing to rotate about its long axis.

[0089] A resistor holder was also developed to place the resistor in a known location outside the intestine. The fixture held the resistor in a known position relative to the sheath, housing, and sensor location. A magnet was glued to the side of the sheath and connected to the magnet on the arm of the resistor, as shown in Figures 21A-21E. This attachment centers the resistor in the sheath during the experiment through the magnetic field of attraction between the sheath magnet and the attached magnet. Thermistor 1 would be aligned with the resistor when a magnet is inserted into the top hole on the attachment to align the housing through the magnet on the housing. In this configuration, thermistor 1 gets an accurate reading of the heat spot resulting from the resistive heater, thus providing a direct measurement of the heat spot compared to the surrounding tissue (thermistor 2). By changing the voltage parameters to the resistor, the heat spot can be set to the desired temperature difference with the surrounding tissue, while the housing and sensor are free to move within the sheath to obtain far IR measurements.

[0090] The sensor data was sent through a custom-made printed circuit board (PCB) connected via a USB connection to a laptop where a custom-made graphical user interface was configured to display all sensor data in real time. The GUI also allows the operator to mark specific events in the data file, such as the presence of polyps or the use of saline nebulization. A screenshot of the GUI is shown as FIG. 22, including A) indicators for board connection, B) a clock for operator notation of specific time events, C) options for selection of various sensors, D) input of the sampling rate of the sensor being tested, E) options to start or stop the measurement or plot, F) ability to mark data for polyp or saline detection, G) file name setting, H) sensor readings such as ambient humidity and temperature, I) plot of far-IR data during measurement, and J) plot of thermistor data. In these example traces, it can be seen that the far-IR sensor readings match the thermistor readings.

[0091] procedure Immature pigs were anesthetized and their abdomen was incised to expose the digestive tract within the abdominal cavity. The spiral colon, with its blood supply intact, was lifted from the abdominal cavity and exposed for access. Nine different locations of the colon were tested. For each location within the colon, an incision was made and a sheath was inserted through the incision. The resistor feature was then magnetically attached to the sheath as described above. Tissue adhesive (3M Vetbond, part number 1469SB) was applied to prevent the resistor from moving relative to the bowel during testing. Once the resistor was attached, the sensor housing was inserted into the sheath and advanced until the housing magnet was aligned with the top magnet of the resistor attachment.

[0092] Once the resistor was in place, a negative control test was performed. In the negative control, the sensor fixture was inserted into the sheath, data collection was started, and the sensor fixture (housing and sensor array) was moved under the resistor. At least one negative control test was performed at each position.

[0093] A heat spot was established after the negative control test. A magnet was placed in the top hole of the resistor fixture (FIG. 21A). The sensor fixture was then pushed through the sheath until the magnetic force indicated that the magnet was aligned with the resistor (FIGS. 21D and 21E).

[0094] Once the sensor fixture was in place, a voltage was applied to the resistor to begin the resistive heating and anomaly simulation. The difference in measurements between Thermistor 1 and Thermistor 2 was monitored and the voltage was adjusted in real time until a steady state difference was obtained. Figure 23 shows example measurements of the two thermistors as the voltage was varied to achieve a steady state Δ of 2.0°C. Once the desired temperature difference was obtained, the top magnet was removed to allow the sensor fixture to move freely within the sheath.

[0095] Two to ten trials were completed at each of the nine locations. Each trial included one to three passes under the simulated anomaly. In blinded trials (or blinded tests), an additional observer was placed in a separate room to watch the live stream data in the GUI and mark possible "polyps" based on the rise and fall of the far-IR data. To evaluate difficult conditions, 5 mL of saline was injected into the inside of the intestinal wall at the location of the heat spot or fecal matter was applied to the sensor face immediately before the test was performed. All measurements were taken at a fixed distance of 0.1" (0.254 cm) from the mucosal surface.

[0096] Positions 1-5 were used to test the 5° and 12° FOV sensor function. Measurements were started after the negative control by establishing a temperature delta of 2°. An exemplary negative control test is shown in FIG. 24. For the 5° FOV measurement, the sensor was passed over the simulated anomaly 10 times. FIG. 25 shows an exemplary experimental result at position 3, with the vertical line indicating the position when the anomaly entered the sensor's FOV.

[0097] Figure 26 shows the far-IR data for each of the 10 experiments at location 3. Each experiment successfully demonstrates the ability of the far-IR sensor to recognize temperature changes associated with simulated anomalies.

[0098] 27A-C show a negative control measurement at position 6 with a 12° FOV sensor (FIG. 27A), an example run at position 6 with a 12° FOV sensor showing with vertical lines the positions when anomalies entered and left the sensor's FOV (FIG. 27B), and data from 10 runs at position 6 (FIG. 27C). Similar to the 5° FOV sensor measurements, most measurements show temperature spikes associated with anomaly detection. Although the 12° FOV sensor measurements were less sensitive and exhibited smaller temperature deviations than the 5° FOV measurements, these sensors were also able to accurately identify the presence of anomalies within the FOV.

[0099] Controlled blinded study Ten blind trials were conducted at three different locations (2, 3, and 6) using both types of sensors. In almost all experiments, the correct number of simulated anomalies were detected by unbiased observers marking the data (i.e., estimated locations) in the live data stream to the GUI when they believed an anomaly was detected and when they believed it was not. The observers were not informed when the anomaly would appear in the test data or how many passes over the anomaly. Each trial included one to three passes over the simulated anomaly. Figures 28A and 28B show data from individual experiments correlating unbiased observations of anomaly detection (vertical dashed lines) with actual locations (vertical solid lines). As shown, the observers were able to accurately identify the presence of subsurface thermal anomalies using the far-IR data.

[0100] This type of information can be transmitted and used in the machine learning algorithms described herein. Figure 29A shows measurements at position 3 with a 5° FOV far IR sensor. A machine learning algorithm was developed based on the data collected here and applied to the raw data shown in Figure 29A in order to identify the location of anomalies (or "outliers"). Figure 29B shows the results of this data analyzed by the machine learning algorithm, with the boxed data showing data identified as outliers by the computation unit. The algorithm was able to correctly identify the anomaly location in addition to certain noise variations. It is expected that the outlier detection rate will decrease with larger datasets and improved machine learning, in addition to optimizing the device geometry.

[0101] Double-blind controlled tests were also performed in which the sensor attempted to identify the heat source without either the passive observer or the GUI operator knowing when the sensor encountered the simulated anomaly. Data from these measurements indicate that the sensor detected temperature changes from the simulated anomaly despite not always being able to precisely control the temperature difference and unknown anomaly location. In these measurements, placing the intestines back into the abdominal cavity resulted in a higher overall temperature than the measurements mentioned above, resulting in a temperature delta of 3°.

[0102] FIG. 30A shows the results of 10 fully blinded trials with a 5° FOV sensor at position 9. Significant temperature fluctuations are observed with a structure similar to measurements made when anomaly detection could be observed. The fluctuations in these measurements may also result from reduced gut integrity due to multiple tissue uses and thickening of the gut wall, as well as minimum trace lengths that prevented identification of a baseline temperature in some measurements. Nevertheless, the peak of the temperature delta measured from the far-IR sensor was measured as 3° C., a temperature difference associated with a resistor attached to the mucosal surface. FIG. 30B shows one of the experiments where a GUI operator estimated anomaly location based on fully blinded data collection.

[0103] Stress factor test A negative control was performed by attaching a resistor to a location outside the colon at either location 3 or 6, setting the heat spot temperature to 2 degrees above the baseline tissue temperature determined from thermistor 2, and applying 5 mL of saline to the inner surface of the intestine directly beneath the simulated anomaly. After application, the simulated anomaly was passed over the sensor as described above. Stress factor measurements were performed with both the 5° FOV sensor and the 12° FOV sensor.

[0104] The measurements taken when the sensor was passed over the simulated anomaly after a saline wash are shown in Figure 31. Although the sensitivity of these measurements was reduced, the simulated anomaly could still be clearly identified in the far-IR data.

[0105] Further stressor testing was performed by applying fecal matter to the surface of the far-IR sensor at position 8. Data from two experiments at position 8 is shown in FIG. 32. As shown, the presence of fecal matter on the sensor surface, a variable likely to be encountered during a colonoscopy, nullifies the simulated abnormality signal.

[0106] Example 5: Polyp temperature profile detection The temperature of polyps within the digestive tract was assessed during routine colonoscopy and compared to the surrounding normal tissue similar to the analysis of temperature rise anomalies found in Stefanadis, C., Journal of Clinical Gastroentereology 36.3 (2003), pp. 215-218, and Banic, M., Periodicum biologorum 113.4 (2011), pp. 439-444, each of which is incorporated herein by reference in its entirety. The central temperature of identified polyps was measured after measuring the temperature of the surrounding normal tissue 1-2 inches away from the polyp.

[0107] Temperature measurements were taken using an Omega HSTH-44000 series thermistor. One thermistor was utilized per patient. An Omega data logger was connected to the thermistor and a laptop for imaging and temperature data reading and analysis. Colonoscopes were provided by Charlottesville Gastroenterology Associates.

[0108] Routine colonoscopy was performed using a colonoscope containing a thermistor for contact thermal measurement of gastrointestinal tissue. Patients were excluded if cancer was identified during colonoscopy or if they had a history of inflammatory bowel disease. The thermistor was placed through the working channel of the colonoscope. After identification of one or more polyps using colonoscopy images, a clean thermistor was placed in contact with the polyp for 15-30 seconds. The thermistor was also placed in contact with the surrounding tissue 2-3 times, for 15-30 seconds each time.

[0109] These augmented colonoscopies were performed on seven patients over two days. One patient had three polyps identified during colonoscopy. The first polyp was identified in the rectum, measuring 6 mm in diameter and 37.2°C in temperature, compared to 36.0°C in the surrounding tissue. Figure 33A shows the temperature measurements around the polyp, measured in degrees Celsius by a thermistor (TM1). The second polyp was identified in the ascending colon, measuring 12 mm in diameter. At the base of the stalk the temperature was 38°C, compared to 37.5°C at the top of the pedunculated polyp. The surrounding tissue was 36.0°C-37.2°C, and the measurements may have been compromised by the presence of moisture in the dependent tissue. Figure 33B shows the temperature measurements around the polyp, measured in degrees Celsius by a thermistor (TM1). A third polyp was present in the descending colon with a diameter of 5 mm and a temperature of 37.5° C., compared to a normal surrounding tissue temperature of 36.0° C. FIG. 33C shows temperature measurements around the polyp measured in degrees Celsius by a thermistor (TM1). Another patient had one polyp in the transverse colon with an estimated diameter of 3 mm and a temperature of 36.5° C., compared to a surrounding tissue temperature of 36.0° C. FIG. 33D shows temperature measurements around the polyp measured in degrees Celsius by a thermistor (TM1).

[0110] Without wishing to be bound by theory, polyps present in any part of the colon (e.g., ascending colon, transverse colon, descending colon, rectosigmoid) will have a higher temperature than the surrounding tissue at each polyp's respective location. A mean difference of 1.215°C was identified between abnormal mucosal tissue (e.g., polyps, subcutaneous abnormalities) and normal mucosal tissue, providing confirmatory evidence for this hypothesis.

[0111] Since various modifications of the above-described subject matter may be made without departing from the scope and spirit of the present disclosure, it is intended that all subject matter contained in the above description or defined in the appended claims be interpreted as illustrative and exemplary of the present disclosure. Many modifications and variations of the present disclosure are possible in light of the above teachings. Accordingly, the present specification is intended to cover all such alternatives, modifications and variations that fall within the scope of the appended claims.

[0112] All documents cited or referenced herein, and all documents cited or referenced in the documents cited herein, together with any manufacturer's instructions, descriptions, product specifications, and product sheets for any products mentioned in this specification or any document incorporated herein by reference, are hereby incorporated by reference and may be employed in the practice of this disclosure. [Explanation of symbols]

[0113] 1 Endoscope 2 Bend 3 Pivoting bin 4 Pivoting Bin 5 Angulation Wire 6 Distal end 7. Camera 10 equipment 12 Far-infrared sensor 13 Far-infrared sensor 14 Overlap 15 Wire 16 Wire A-axis

Claims

1. a distal end of a tubular portion of an endoscope, the distal end comprising a distal end of an endoscope tube and a wall of the tube proximal to the distal end; a) a plurality of far-infrared sensors and / or one or more temperature sensors independently distributed around and / or inside the wall; b) a camera disposed at the distal end for imaging objects in front of the tube (e.g., the camera's field of view includes the major longitudinal axis of the tube, which extends beyond the distal end); wherein the plurality of far-infrared sensors, one or more temperature sensors, and camera are configured to communicate with each other and transmit data for processing and analysis, A tip portion characterized by:

2. the field of view of at least one infrared sensor includes an axis perpendicular to a wall surface or perpendicular to the longitudinal axis of the tube; The tip of claim 1 .

3. the field of view of each of the plurality of far-infrared sensors and the field of view of the camera do not overlap; The tip of claim 1 .

4. a field of view of at least one of the plurality of far-infrared sensors and a field of view of the camera overlap; The tip of claim 1 .

5. the fields of view of at least two of the plurality of far-infrared sensors overlap; The tip of claim 1 .

6. At least one of the far-infrared sensors has a field of view of less than 25° (or between 0.1° and 25°) (e.g., less than 23°, less than 18°, less than 15°, between 1° and 20°, between 4° and 13°, or between 5° and 12°). The tip of claim 1 .

7. At least 50% (e.g., at least 60%, at least 70%, at least 80%, at least 90%, or all) of the plurality of far-infrared sensors are tilted at an angle of less than 25° (or between 0.1° and 25°) (e.g., less than 22°, less than 18°, less than 15°, 3° to 22°, 1° to 20°, 4° to 13°, 5° to 12°, 1° to 2°, 2° to 3°, 3° to 4°, 4° to 5°, 5°-6°, 6°-7°, 7°-8°, 8°-9°, 9°-10°, 10°-11°, 11°-12°, 12°-13°, 13°-14°, 14°-15°, 15°-16°, 16°-17°, 17°-18°, 18°-19°, 19°-20°, 20°-21°, 21°-22°, 22°-23°, 23°-24°), The tip of claim 1 .

8. At least 90% (e.g., all) of the plurality of far-infrared sensors independently have a field of view of 1° to 20° (e.g., 5° to 12°); The tip of claim 1 .

9. a portion (e.g., a first portion, a second portion) of the plurality of sensors is disposed around or partially around the circumference of the wall; The tip of claim 1 .

10. a portion (e.g., a first portion, a second portion) of the plurality of sensors is embedded in a circumference of the wall; The tip of claim 1 .

11. the circumference is perpendicular to the major longitudinal axis of the tube; The tip of claim 10.

12. a portion (e.g., a first portion, a second portion) of the plurality of sensors is linearly arranged along the wall, and the linear dispersion is substantially parallel (e.g., ±5°, ±1°) to a horizontal axis of the tube; The tip of claim 1 .

13. The tube is a cylindrical tube and / or a tube with straight and / or curved edges, The tip of claim 1 .

14. The cylindrical tube is an elliptical cylinder. The tip of claim 13.

15. The cylindrical tube is a cylinder. The tip of claim 13.

16. a device attached to a tubular portion of an endoscope, the device for tip augmentation being attachable (e.g., removably attachable) to a wall proximal to a distal end of an endoscope tube, the endoscope including a camera disposed at the distal end for imaging objects in front of the tube (e.g., the camera's field of view includes a major longitudinal axis of the tube extending beyond the distal end); the device for tip augmentation includes a plurality of far-infrared sensors and / or one or more temperature sensors distributed on a substrate, the substrate being attachable to the wall proximal to the distal end of the endoscope; the plurality of far-infrared sensors, the one or more temperature sensors, and the camera are configured to communicate with each other and transmit data for processing and analysis, respectively; An apparatus characterized in that

17. the field of view of at least one infrared sensor includes an axis perpendicular to a wall surface or perpendicular to the longitudinal axis of the tube; 17. The apparatus of claim 16.

18. When the device is attached to the endoscope, the field of view of each of the plurality of far-infrared sensors and the field of view of the camera do not overlap.

17. The apparatus of claim 16.

19. When the device is attached to the endoscope, a field of view of at least one of the plurality of far-infrared sensors and a field of view of the camera overlap.

17. The apparatus of claim 16.

20. the fields of view of at least two of the plurality of far-infrared sensors overlap; 17. The apparatus of claim 16.

21. At least one of the far-infrared sensors has a field of view of less than 20° (or 0.1° to 25°) (e.g., less than 18°, less than 15°, 1° to 20°, 4° to 13°, 5° to 12°).

17. The apparatus of claim 16.

22. At least 50% (e.g., at least 60%, at least 70%, at least 80%, at least 90%, or all) of the plurality of far-infrared sensors are tilted at an angle of less than 25° (or between 0.1° and 25°) (e.g., less than 22°, less than 18°, less than 15°, 3° to 22°, 1° to 20°, 4° to 13°, 5° to 12°, 1° to 2°, 2° to 3°, 3° to 4°, 4° to 5°, 5°-6°, 6°-7°, 7°-8°, 8°-9°, 9°-10°, 10°-11°, 11°-12°, 12°-13°, 13°-14°, 14°-15°, 15°-16°, 16°-17°, 17°-18°, 18°-19°, 19°-20°, 20°-21°, 21°-22°, 22°-23°, 23°-24°), 17. The apparatus of claim 16.

23. At least 90% (e.g., all) of the plurality of far-infrared sensors independently have a field of view of 1° to 20° (e.g., 5° to 12°); 17. The apparatus of claim 16.

24. When the device is attached to the endoscope, a portion (e.g., a first portion, a second portion) of the plurality of sensors is positioned to surround or partially surround a circumference of the wall.

17. The apparatus of claim 16.

25. When the device is attached to the endoscope, the circumference is perpendicular to the major longitudinal axis of the tube.

25. The apparatus of claim 24.

26. When the device is attached to the endoscope, a portion (e.g., a first portion, a second portion) of the plurality of sensors are linearly arranged along the wall, and the linear dispersion is substantially parallel (e.g., ±5°, ±1°) to a horizontal axis of the tube.

17. The apparatus of claim 16.

27. the tube is a cylindrical tube, and the substrate is sized to be attached to the tube; 17. The apparatus of claim 16.

28. The cylindrical tube is an elliptical cylinder (e.g., the substrate is elliptical), 28. The apparatus of claim 27.

29. The cylindrical tube is cylindrical (e.g., the substrate is circular); 28. The apparatus of claim 27.

30. a) an endoscope having a tip portion according to claim 1 or a device according to claim 16 attached to said tip portion; b) a machine-readable medium configured to receive the data transmitted from the plurality of far-infrared sensors and, optionally, the camera; c) a processor including instructions for analyzing the data transmitted to the machine-readable medium to identify abnormalities in an organ or a set of organs (e.g., the subject's gastrointestinal tract) based on data from the far-IR sensor and, optionally, on the provided image (e.g., diseases such as polyps present on the tissue surface of the organ, diseases present below the tissue surface, tumors, cysts, granulomas, circulatory abnormalities, inflammation); An endoscope system comprising:

31. The processor includes instructions for a camera ML / AI algorithm to detect anomalies from the camera images transmitted to the machine-readable medium. The endoscope system of claim 30.

32. The processor includes instructions for a far-infrared AI algorithm to detect abnormal data transmitted from the plurality of sensors to the machine-readable medium. The endoscope system of claim 30.

33. the processor includes instructions for comparing an output of the camera ML / AI algorithm with an output of the far-infrared ML / AI algorithm.

33. The endoscope system of claim 32.

34. the instructions for analyzing the data to identify abnormalities include calculations involving the location of the distal tip within the gastrointestinal tract and / or the rate of movement of the distal tip during data collection. The endoscope system of claim 30.

35. the processor includes instructions for analyzing data transmitted from the plurality of far-infrared sensors to identify subcutaneous abnormalities. The endoscope system of claim 30.

36. 1. A method of observing an object using an endoscope system, comprising: a) detecting heat and / or temperature data of an object using an endoscope having a plurality of far-infrared and / or temperature sensors distributed around a wall proximal to a distal end of a tube, the endoscope positioned such that the object is within a field of view of at least one of the plurality of far-infrared and / or temperature sensors; b) transmitting said heat and / or temperature data to a machine-readable medium; A method comprising:

37. 1. A method of observing an object using an endoscope system, comprising: a) providing an endoscope having a tip section according to claim 1 or a device according to claim 16 attached to said tip section; b) arranging a light source capable of emitting light (e.g., white light, red light, blue light, green light, infrared light, near-infrared light), such that the light emitted from the light source is reflected from the object into the camera to form image data; c) transmitting the image data to a machine-readable medium; d) optionally, moving the distal end of the endoscope so that the object is within a field of view of one or more of the plurality of far-infrared sensors, and detecting thermal data of the object; e) transmitting thermal and / or temperature data from said plurality of far IR and / or temperature sensors to a machine readable medium; A method comprising:

38. 1. A method of observing an object using an endoscope system, comprising: a) positioning an endoscope having a distal end portion according to claim 1 or a device according to claim 16 attached to the distal end portion such that the object is within the field of view of at least one of the plurality of far-infrared sensors, and detecting heat and / or temperature data of the object; b) transmitting said heat and / or temperature data to a machine-readable medium; c) arranging a light source capable of emitting light (e.g., white light, red light, blue light, green light, infrared light, near-infrared light), such that the light emitted from the light source is reflected from the object into the camera to form image data; d) transmitting the image data to a machine-readable medium; A method comprising:

39. The light source is positioned by movement of the distal end.

38. The method of claim 37.

40. the distal end is moved such that the object passes through the field of view of at least two far-infrared sensors of the plurality of far-infrared sensors; 37. The method of claim 36.

41. thermal data from each sensor is collected during the movement and transmitted to the machine-readable medium; 41. The method of claim 40.

42. the machine-readable medium in communication with a processor including instructions for analyzing the thermal and / or image data; 38. The method of claim 37.

43. the thermal and / or temperature data identifies an anomaly (e.g., a polyp, a subcutaneous anomaly) at a location (e.g., a tissue location), and the light source is positioned to reflect off the identified anomaly into the camera; 38. The method of claim 37.

44. The method further comprises cleaning the location.

44. The method of claim 43.

45. the cleaning is performed before placing the light source to observe the location of the anomaly.

45. The method of claim 44.

46. the cleaning is performed after positioning the light source and observing the location of the anomaly, the method further comprising repositioning the light source to reflect off the identified anomaly into the camera and observing the location of the cleaned anomaly.

45. The method of claim 44.

47. transmitting and / or displaying the thermal and / or temperature and / or image data on an interface (e.g., a graphical user interface) to allow an endoscope user to observe (e.g., in real time) objects identified in the data (e.g., anomalies, cleaned anomalies, potential anomalies); 38. The method of claim 37.

Images