Endotracheal tube placement confirmation and displacement monitoring

Abstract

Systems and methods for endotracheal tube (ETT) placement confirmation and / or displacement monitoring includes an ETT that includes an optical fiber running substantially along a length of the ETT. The ETT has an insertion end and the optical fiber includes a tip proximate to the insertion end of the ETT. A source of near-infrared (NIR) wavelength light is operably connected to the optical fiber. A detection system is spaced apart from the ETT and the optical fiber. The detection system is configured for noninvasive detection of NIR wavelength light emitted from the optical fiber and diffused through the patient. A processor is communicatively connected to the detection system and configured to receive data from the detection system. The processor is configured to determine an initial position of the tip of the optical fiber and to continuously monitor subsequent position of the tip of the optical fiber.

Description

ENDOTRACHEAL TUBE PLACEMENT CONFIRMATION AND DISPLACEMENTMONITORINGSTATEMENT REGARDING FEDERALLY SPONSORED RESEARCH AND DEVELOPMENT

[0001] This invention was made with government support under 1R21HD105021-0I and 5R21HD105021-02 awarded by National Institutes of Health (NIH) I National Institute of Child Health and Human Development (NICHD). The government has certain rights in the invention.CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This Application claims priority of U.S. Provisional Patent Application No. 63 / 621,058, filed on January 15, 2024, and is hereby incorporated by reference herein in its entirety.BACKGROUND

[0003] An endotracheal tube (ETT) is a pliable plastic duct placed into a patient’ s trachea to facilitate mechanical ventilation when a person is unable to breathe on their own. This may occur when anesthesia or substantial sedation is administered during surgery, when a patient is in the intensive care unit, or when a patient suffers from some other injury or pathologic process. Correct positioning of the ETT inside the trachea to fully ventilate the lungs is critical. Improper placement of the ETT may cause severe complications or even death, particularly in newborns and children. When an ETT is misplaced or becomes displaced, prompt intervention is required to correct the position. Depending upon the degree of displacement, replacement of the ETT may be necessary, however such procedure may be difficult and can be associated with subsequent patient complications. A simple and cost-effective technology does not currently exist to continuously confirm the position of the ETT after initial placement and rapidly alert clinical care teams of ETT displacement, particularly, detection and alert of minor ETT displacements may indicate a risk of a more significant displacement (e.g. extubation).

[0004] There are several techniques for clinical confirmation of the ETT position. Chest X-ray (CXR) is the current gold standard, but it has several drawbacks, including repeated exposure to radiation, the need for specialized equipment that generally requires interruption ofpatient care, and the ability to provide ETT position information only at a single time point. In patients who arc intubated, respiratory function (tidal volume, airway pressure, and flow), peripheral oxygen levels, and ETT carbon dioxide levels are generally measured as a proxy for appropriate placement of an ETT and proper ventilation. However, these measures do not provide specific or immediate feedback on ETT placement and can only provide information late after EET displacement resulting in serious patient complications. Other methods for determining ETT placement have been proposed, such as ultrasound or fiberoptic visualization; however, these techniques are not intended for continuous monitoring and are limited in clinical practice.

[0005] There are different methods to secure an ETT after placement, but keeping the tube at the correct position in various situations, such as those in which a patient must be moved or transported, may be challenging. There are specific patient circumstances such as surgical reconstruction of the airway or trachea or when there exists a limited distance for proper placement, especially in children and neonates, which increase the critical nature of maintaining the proper placement of an ETT. Additionally, displacement of the ETT can result in movement of the tube, which may result in mainstem bronchi ventilation and primarily aeration of only one lung. Often, identifying this difficulty happens later in a patient’s clinical course and can be associated with significant complications. In each of these situations, a monitoring system to easily and continuously confirm ETT position would provide an important adjunct to clinical care and would have the ability to improve patient outcomes and limit patient complications.

[0006] Various patents and publications disclose features of ETT systems and methods of monitoring and placement of ETT systems. These include US11,382,524, “Airway Maintenance Device;” US 10,206,607, “Methods and Apparatus for Optoacoustic Guidance and Confirmation of Placement for Indwelling Medical Apparatus;” US10,194,792, “Optical Device, Sheath and Endotracheal Intubation System;” US9,949,629, “Endotracheal Tube Insertion Device;” US9, 167,962, “Airway Management;” US7,992,573 “Optically Guided System for Precise Placement of a Medical Catheter in a Patient;” US5,285,778, “Endotracheal Tube with Fibers Optic Illumination and Viewing and Auxiliary Tube;” US4,898,175, “Out-body Observing Apparatus;” US2013 / 0096379, “Double-lumen Endotracheal Tube Devices, Systems and Methods;” US2012 / 0296162, “Method for Positioning a Disposable Sterile Endotracheal Tube, and Corresponding System for Intubation;” and US2008 / 0039715, “Three-Dimensional Optical Guidance for Catheter Placement,” each of which are incorporated by reference herein in theirentireties. Papers Ullah, et al. “Real-Time Optical monitoring of Endotracheal Tube Displacement” Biosensors, 10, 174 (2020), Lu ct al. “Continuous Monitoring of Endotracheal Tube Position with Near Infrared Light” J Biomed Opt. 2024 Mar;29(3):035001, and Lu et al. "Realtime and noninvasive assessment of endotracheal tube displacement using near-infrared and visible cameras" Biomed. Opt. Express 15, 6355-6369 (2024) are also incorporated by reference herein in their entireties.BRIEF DISCLOSURE

[0007] Endotracheal tube (ETT) intubation is an important airway management procedure to enable appropriate ventilation and gas exchange in the lungs of critically ill patients or patients recovering from a major surgical intervention. Given that patients with ETT intervention are not able to breathe and ventilate on their own, ETT misplacement or intra-treatment movement is of particular' concern. The occurrence of ETT misplacement is particularly high in newborns and young children due to their shorter trachea. Currently, there is no clinical system that can be utilized to provide accurate and real-time information to care teams about the position of the ETT.

[0008] Systems and methods as described herein provide an optical approach to confirm that the ETT tip is placed in an intended position and to continuously monitor the ETT tip thereafter. The disclosed device includes a near-infrared (NIR) sensor that uses an optical fiber embedded in an ETT to emit NIR wavelength light, and a detector outside the body for NIR light detection. NIR detector configurations are also disclosed, including a photodetector array configured for physical contact with the skin at the intended ETT tip position and a dual-camera for noncontact imaging of the NIR pattern on the skin, as well as a signal / image processor, which may for example, include a data acquisition card, a computer, and image processing algorithms.

[0009] A system for endotracheal tube (ETT) placement confirmation and / or displacement monitoring includes an ETT that includes an optical fiber running substantially along a length of the ETT. The ETT has an insertion end and the optical fiber includes a tip proximate to the insertion end of the ETT. A source of near-infrared (NIR) wavelength light is operably connected to the optical fiber. A detection system is spaced apart from the ETT and the optical fiber. The detection system is configured for noninvasive detection of NIR wavelength light emitted from the optical fiber and diffused through the trachea and skin. A processor is communicatively connected to the detection system and configured to receive data from the detection system. The processor isconfigured to determine an initial position of the tip of the optical fiber and to continuously monitor subsequent position of the tip of the optical fiber.

[0010] In examples of the system for ETT placement confirmation and / or displacement monitoring, the tip of the optical fiber is inset from the insertion end of the ETT. The optical fiber may be inset by 1cm from the insertion end of the ETT. The tip of the optical fiber includes a 45° angle and a mirror. The NIR wavelength light may be between 660nm - 2500nm, optionally the NIR wavelength light is between 780nm-850nm, optionally, the NIR wavelength light is around 810nm, the NIR wavelength light is optionally 660nm, 780nm, 800nm, 805nm, 810nm, 830nm, or 850nm. The optical fiber may be inserted into a channel in the wall of the endotracheal tube.

[0011] In examples, the detection system may include an array of photodetectors. The array of photodetectors includes a first photodetector configured to be positioned on the tracheal skin aligned to the tip of the optical fiber. Second and third photodetectors are linearly aligned with the first photodetector along a longitudinal dimension relative to the ETT. Fourth and fifth photodetectors are linearly aligned with the first photodetector along a lateral dimension. The lateral dimension is perpendicular to the longitudinal dimension. An adhesive layer is configured to secure the array of photodetectors to skin of a patient. The processor may be communicatively connected to the first, second, and third photodetectors, and the processor is configured to calculate a first ratio of a first signal from the first photodetector to a second signal from the second photodetector and a second ratio of a third signal from the third photodetector to the second signal and to calculate a longitudinal displacement of the endotracheal tube from at least one of the first ratio and the second ratio. The processor may be communicatively connected to the fourth and fifth photodetectors, and the processor is configured to calculate third ratio of a fourth signal from the fourth photodetector to the second signal and a fourth ratio of a fifth signal from the fifth photodetector to the second signal and to calculate a rotational displacement of the endotracheal tube from at least one of the fourth ratio and the fifth ratio. The first and second ratios are further calculated from an average of the second, fourth, and fifth signals as the second signal. The third and fourth ratios are further calculated from an average of the second, first, and third signals as the second signal.

[0012] In additional examples, the detection system may include an infrared light camera configured to provide infrared images to the processor and a visible light camera configured to provide visible images to the processor. The processor may be configured to calculate total pixelintensities in a horizontal dimension in the infrared images against pixel position in a vertical dimension and the processor is configured to determine a rotation of the endotracheal tube from shift in the peak pixel position of the intensity profile from a reference peak pixel position. The processor may be configured to calculate total pixel intensities in a vertical dimension in the infrared images against pixel position in a horizontal dimension and the processor may be configured to determine an endotracheal tube displacement from shift in the peak pixel position of the intensity profile from a reference peak pixel position. The processor may be configured to fit a Gaussian distribution to the total intensity profile from which the peak position is determined. The processor may be further configured to associate a visible image of the visible images to a corresponding NIR image of the infrared spectrum images collected concurrently. The process or may be configured to determine an artifact patient motion between visible spectrum images. The processor may be configured to calculate a corrected endotracheal tube displacement / rotation from the endotracheal tube displacement / rotation and the artifact patient motion. The processor may be configured to identify a reference mark in the visible images and configured to calculate a shift in the reference mark between visible images as the artifact patient motion. The processor may be configured to apply image segmentation to the visible images, and to identify corresponding feature points between visible images, and to calculate a mean disparity between the corresponding feature points between the visible images as the artifact patient motion. The processor may be configured to identify a pulse artifact within a sequential series of the infrared images. Two or more additional light sources with one exemplarily between 660nm - 780nm and another exemplarily between 830nm - 950nm may be coupled to the optical fiber for use in measuring hemoglobin oxygen saturation.

[0013] A method of estimating displacement of an endotracheal tube (ETT) includes providing an optical fiber configured to emit near-infrared light in the ETT. A first NIR image is obtained and a subsequent second infrared image is obtained with an infrared camera having a field of view including an expected endotracheal tube position. A first visible image is obtained and a subsequent second visible image is obtained with a visible camera having a field of view including the expected endotracheal tube position. A total pixel value is calculated in a first dimension of the first infrared image and the second infrared image. A first pixel location is identified in a second dimension that corresponds to a peak total pixel value in the first dimension for the first infrared image. A second pixel location is identified in the second dimension thatcorresponds to a peak total pixel value in the first dimension for the second infrared image. An endotracheal tube shift is calculated from a difference between the first pixel location and a second pixel location. Matched point pairs are identified between the first visible image and the second visible image. A mean disparity between the matched point pairs is calculated to calculate a patient motion artifact. A corrected endotracheal tube shift in pixels is calculated from the endotracheal tube shift and the patient motion artifact. A calibration function is applied to the corrected endotracheal tube shift in pixels to calculate an endotracheal tube displacement in millimeters. The endotracheal tube displacement is compared to a predetermined displacement threshold value. An alarm is produced if the endotracheal tube displacement exceeds the threshold value.BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 depicts an example of an endotracheal tube with an illumination assembly.

[0015] Figure 2 is a detailed view of an example of an insertion end of an endotracheal tube with an illumination assembly.

[0016] Figure 3A depicts an example of the endotracheal tube with an illumination assembly and a detector array detection system.

[0017] Figure 3B is a detailed view of an example of a detector array.

[0018] Figure 4A is a graph of a ratio of Vpi / Vp2 plotted against a ratio of Vps / Vpi.

[0019] Figure 4B is a graph of the calibration process based on the ratios rn = Vpi / Vp and 1’32 = Vp3 / Vp2 at the selected displacements (either -10, -5 and 0 mm or 0, +5 and +10 mm).

[0020] Figure 5 depicts an example of the endotracheal tube with the illumination assembly and a dual-camera detection system.

[0021] Figure 6A depicts the geometry of the image detection for dual-camera detection.

[0022] Figure 6B represents the camera alignment with the anterior-posterior axis of the patient.

[0023] Figure 7 depicts images and corresponding graphs to describe an example of using NIR images to calculate the uncorrected ETT displacement or rotation. Figures 7A, 7D, and 7D depict exemplary NIR images (white patterns) overlay with corresponding visible images captured by the dual-camera method with an ETT position at -15, 0 and +15 mm, respectively.

[0024] Figures 7B, 7E, and 7H are 1 -dimensional NIR intensity profile graphs in the horizontal direction corresponding to images 7A, 7D and 7G.

[0025] Figure 7C, 7F, and 71 are 1 -dimensional NIR intensity profile graphs in the vertical direction corresponding to images 7A, 7D and 7G.

[0026] Figure 7J is a scattered plot of the horizontal peak positions extracted from the NIR images captured at all positions plotted against the horizontal displacements measured by the reference sensor with the exemplary positions in Figures 7A, 7D and 7G highlighted by the three circles.

[0027] Figure 8 graphically depicts an example of a method of estimating ETT longitudinal displacement Dxfrom the NIR and visible images.

[0028] Figure 9 is a flow chart that depicts an example of the method of estimating ETT displacement.

[0029] Figure 10 presents experimental data showing the correlation between the determined peak position shift in pixel and the measured longitudinal displacement of the ETT.

[0030] Figure 11A depict a natural or intentionally placed visible mark for mark-based motion detection.

[0031] Figure 11B depicts heterogenous skin movement of the patient for subject-based motion detection.

[0032] Figure 11C is a graph of estimated motion results on a pixel by pixel basis between an application of the mark-based motion detection and a subject-based motion detection.

[0033] Figure 1 ID is a scatter plot of estimated motion from the mark-based motion detection against the estimated motion from the subject-based motion detection.

[0034] Figure 12 is a graph of experimental results comparing ETT position to position estimates using the detector array system.DETAILED DISCLOSURE

[0035] Systems and methods of endotracheal tube (ETT) placement and positioning detection are disclosed herein. The systems and methods use a detection system in combination with an ETT having an illumination assembly.

[0036] Figure 1 depicts an example of an ETT 10 with an illumination assembly 12 as described in further detail herein inserted into a patient 14. In a tested example, the ETT 10 is exemplarily a 7mm uncuffed ETT, although it will be recognized that other dimensions and constructions of ETT may be used within the scope of the present disclosure, including but notlimited to ETTs with a size between 3-7mm. An insertion end 18 of the ETT 10 is inserted into the trachea 44 of the patient 14. Proper placement of the ETT locates the insertion end 18 proximate to, but superficial of, the carina 34. In adults this position range may be about 2-5 cm , while this range may be 0.2 -2.0 cm for neonates. The ETT exemplarily includes a wye joint 38 with one branch connected to a ventilator 40 for providing respiratory support to the patient. The other branch is exemplarily connected to a syringe 42 for delivery of fluids through the ETT 10.

[0037] The ETT 10 includes an illumination assembly 12 which exemplarily includes an optical fiber 16 attached to the ETT 10. As described herein, the optical fiber 16 may be a sidefiring optical fiber and may exemplarily be a 200pm fiber. It is recognized that the optical fiber 16 may be another size than 200pm while remaining within the scope of the present disclosure. The illumination assembly 12 further includes an NIR light source 36 to which the optical fiber 16 is connected and driven to an output, exemplarily up to 2.3mW. It is similarly recognized that the fiber may be driven with more power than 2.3 mW so long as the power remains below an appropriate maximum exposure (MPE) limit for NIR light in tissue. For example, MPE limits may be provided by, but are not limited to IEC 60825 and ANSI Z136.1. As explained later below, the light source 36 may be able to sequentially emit multiple wavelengths of light, for combination of ETT monitoring with detection of other physiological parameters.

[0038] Figure 2 is a detailed view of an example of a insertion end 18 of an ETT 10. The optical fiber 16 may be embedded into the material of the ETT or may exemplarily be secured to the exterior of the ETT, for example with adhesive. As depicted in Fig. 2, the ETT 10 includes a wall 20 that defines the lumen 22 of the ETT 10. A channel 24 is defined within the wall 20, the channel 24 being dimensioned so as to have an inside diameter larger than an outer diameter of the optical fiber 16. The optical fiber 16 extends down the length of the ETT 10 through the channel 24 and ends in the channel 24 at a defined location relative to a tip 26 of the ETT 10. The channel 24 may end in a blind end interior to the wall 20 of the ETT. In another example, the channel 24 may extend the entire length of the ETT, but the channel is filled with an epoxy, silicone, or other biocompatible sealant to close the channel 24 at the insertion end of the ETT and / or to secure the optical fiber in place. The wall 20 between the channel 24 is exemplarily sufficiently thin and transparent enough such that the light emitted from the tip 32 of the optical fiber 16 passes through the wall 20 with sufficient intensity to also pass through the intervening tissue of the patient for detection as described in further detail herein. In a still further example, the wall of the ETT 10may further include a cut out exposing the tip 32 of the optical fiber 16, which may then be sealed using an epoxy, silicone, or other biocompatiblc sealant that is transmissive to the light wavelengths emitted from the tip 32. The tip 32 of the optical fiber may be set back from the tip 26 of the ETT 10, for example but not limited to 10 mm from the tip 26 of the ETT. As described in further detail herein, an offset of the optical fiber tip 32 from the ETT tip 26 facilitates detectability of the optical fiber tip 32 even if the ETT is slightly over-inserted to move to or beyond the carina 34 (Fig. 1). The optical fiber tip 32 is exemplarily located such that when the ETT tip 26 is properly positioned within the patient, the optical fiber tip 32 is within the trachea of the patient. It has been found that the tracheal region is thinner and is free of bone obstructions (e.g. clavicle, rib, sternum) and therefore enables better light transmission to the outside of the patient.

[0039] The tip 32 of the optical fiber 16 is cut or polished to a 45° angle and coated with a silver mirror 30. The mirror 30 at the tip 32 reflects the light 28 from the optical fiber 16 into the tracheal tissue at a generally 90° angle to the optical fiber 16. The mirror 30 of the angled tip 32 directs light 28 from the optical fiber 16 generally outwards from the ETT 10 through the trachea towards the anterior skin.

[0040] In another example, the optical fiber is epoxied into a V-groove in an ETT. Other manners of securing the optical fiber to the ETT will be recognized by a person of ordinary skill in the ait based upon the present disclosure.

[0041] The optical fiber 16 is connected to a source 36 of near- infrared (NIR) light which is, for example, an LED (e.g Model M810F2, Thorlabs, Newton, NJ). In examples, the light may be a wavelength or wavelengths between 660nm - 2500nm. In still further examples, the light may have a wavelength between 780nm-850nm. In examples, the wavelength may be near 810nm. Examples of NIR wavelengths include 700nm, 780nm, 800nm, 8O5nm, 810nm, 830nm, 850nm, 900nm, and 950nm. In an example, wavelengths between 800nm - 810nm, including the 800nm, 805nm and 810nm wavelengths may provide advantages as providing high penetration depth in biological tissue compared to other wavelengths and minimizes the effects of blood oxygenation because it is near the isosbestic wavelength of hemoglobin. It will be recognized that these values and ranges are exemplary and do not preclude other values or ranges as may be recognized in view of the present disclosure. NIR light is found to be advantageous over visible light because of itslower absorption and scattering in human tissue which promotes passing through the patient’s tissue for in vivo detection.

[0042] Two examples of detection systems 100 are described herein. It will be recognized that features, operations, or components of the detection systems may be applicable to the other or combined with the other to arrive at still further combinations of detection systems that are within the scope of the present disclosure. The detection system 100 is positioned on the chest skin for noninvasive detection of light emitted from the optical fiber through the trachea of the patient. From noninvasive detection of the emitted light, the position of the ETT 10 (and specifically the ETT tip 26) can be determined and monitored for movement. The monitored movement may be one or both of a longitudinal displacement (Lp) (e.g. in the direction of the axis of the ETT) or rotation (R) (e.g. about the axis of the ETT). As disclosed herein, detection systems can identify either movement of longitudinal displacement or rotation and produce an alarm to alert a clinician as to the movement of the ETT. While longitudinal displacement is directly relevant to clinical outcomes of ETT movement (e.g. extubation, or impaired intubation airflow delivery or performance), it has further been found that ETT rotational movement may also have clinical importance as an indicator for patient assessment. ETT rotational movement may reduce the NIR light intensity for detection by the system disclosed herein, and reduce the system’s ability to monitor ETT movement. Additionally ETT rotational movement may indicate that the patient is coming out of anesthesia prematurely, that the ETT has been inadvertently moved by an object or clinician, or that the ETT is susceptible to longitudinal displacement. Therefore, in examples, detection of ETT rotation movement my result in a produced alarm or alert for a clinician to review or evaluate the patient and ETT placement.

[0043] Figures 3A and 3B depict an example of a detection system 100 using a detector array 102. Fig. 3A is a schematic diagram of the ETT 10 and detection system 100. The detection system 100 includes a detector array 102. Fig. 3B is a detailed view of an example of the detector array 102. The detector array 102 is connected to a data acquisition card 104. The data acquisition card 104 may exemplarily be an Arduino board (MEGA2560) and provides the data to a computer 106 which exemplarily includes a controller 108, memory in the form of a computer readable medium (CRM) 110, an input device 112, and a graphical display 114)

[0044] In examples, the controller 108 is embodied in a processor which may be any suitable hardware processor or combination of processors, such as a central processing unit (CPU),a graphics processing unit (GPU), an accelerated processing unit (APU), a microcontroller, an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), etc. The processor is exemplarily connected to memory which may include any suitable storage device or devices that can be used to store instructions, values, data, etc., that can be used, for example, by the processor to generate data, to receive data, to analyze data, to present content using the display, or to communicate between components. The CRM is memory and can include any suitable volatile memory, non-volatile memory, storage, or any suitable combination thereof, including but not limited to random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), one or more flash drives, one or more hard disks, one or more solid state drives (SSD), one or more optical drives, etc.

[0045] The controller 108 is communicatively connected to or integrated with the CRM 110 which is non-transient and upon which is stored computer readable code in the form of computer programs or software configured for execution by the controller 108. It will be recognized that the controller 108 is exemplarily implemented with any of a variety of known controller circuits, integrated circuits, microcontrollers, or associated circuitry, for example as described above. The controller 108 may be part of a central processing unit (CPU) which includes integrated memory, although in embodiments the CRM may be a separate component or communicatively connected to the controller 108. The controller 108 accesses software or firmware in the form of computer readable code stored on the CRM 110 as either integrated memory or external memory. The controller 108 executes the computer readable code as an instruction set to carry out the methods and functions as described herein, including the receipt of input, calculations, and outputs as will be described, both herein as well as in the previously noted applications which have been incorporated by reference herein.

[0046] Referring to Fig. 3B, the detector array 102 exemplarily includes a detector board 116 which is exemplarily a printed circuit board (PCB). The detector board 116 includes a data connection 118 which is communicatively connected to the data acquisition card 104. Five photodetectors (Pl, P2, P3, P4, P5) arranged in a “cross” pattern are secured to the detector board 116. The photodetectors are exemplarily phototransistors (e.g. SMD2440-002, Honeywell, Charlote, NC). The phototransistors exemplarily have a peak sensitivity at 880nm, a spectral range of 350-1 lOOnm, and a view angle of 50 degrees. It will be recognized that this is merely exemplary and other photodetectors with other operating characteristics will be recognized from thedisclosure herein by a person of ordinary skill in the art to be within the scope of the present disclosure. In the example depicted, photodetector P2 is located centrally to the detector board 116. Photodetector Pl and photodetector P3 are spaced from photodetector P2 and from each other along a longitudinal axis LD of the detector array 102. When the detector array 102 is positioned on the patient 14, the longitudinal axis LD of the detector array 102 is generally aligned with the longitudinal axis Lp of the patient 14. Photodetectors Pl, P2, and P3 are exemplarily spaced apart at 10mm (center to center) along the longitudinal axis LD. Photodetector P4 and photodetector P5 are spaced from photodetector P2 and from each other along a width axis WD of the detector array 102. When the detector array 102 is positioned on the patient 14, the width axis WD of the detector array 102 is generally perpendicular to the longitudinal axis Lp of the patient 14. Photodetectors P4, P2, and P5 arc exemplarily spaced apart 3.5 mm (center to center) along the width axis WD-

[0047] The detector array 102 may further include an adhesive layer 120 of a biocompatible adhesive. The detector array 102 further includes a light blocking enclosure 122. The enclosure 122 may be constructed of an injection molded plastic and blocks infiltration of light to the photodetectors (Pl, P2, P3, P4, P5). In still further examples, an opaque adhesive film may extend over the detector array 102 to secure the detector array 102 to the skin of the patient, while also blocking light infiltration to the photodetectors (Pl, P2, P3, P4, P5) from the side. In such examples, the opaque adhesive film may be used in addition to or instead of the adhesive layer 120.

[0048] In use, when the ETT 10 is positioned and the illumination assembly 12 energized, illumination from the optical fiber tip 32 may be visible, with or without the assistance of a camera (depending upon the wavelength of emitted light) however, as described herein the wavelength is likely in the NIR range and exemplarily 800nm - 810nm, which would require a camera that includes NIR detection for a clinician to see. The detector array 102 may then be adhered to the patient’s skin at the trachea with an attempt to center the light emission from the ETT on the “central” photodetector (e.g. P2 in Figure 3(b)). It will be recognized that the patient’s skin at the trachea in alignment with the optical fiber tip 32, may depending upon patient anatomy, be characterized as the chest or upper chest of the patient, while the intended location places the optical fiber tip 32 within the trachea of the patient, for the better transmissive properties of this anatomical region. The tip 26 of the ETT 10 exemplarily extends further into the patient from the optical fiber tip 32. A feature of a detector array 102 adhered to the skin of the patient as describedherein is that the detector array 102 generally moves in correspondence with the patient, and particularly with the trachea of the patient. Therefore, this disclosed detector system is robust to patient motion artifacts, particularly with the minimal expected motion of a sedated intubated patient. As described above, the detector array 102 is adhered on the patient with the photodetectors in the longitudinal dimension aligned with a longitudinal dimension of the ETT.

[0049] As previously noted, the detector array 102 is communicatively connected to a computer. This is exemplarily a wired connection, although it will be recognized that wireless communicative connections may also be used. Through the communicative connection, the computer receives sensor data from each of the photodetectors (P 1 , P2, P3, P4, P5). NIR light from the optical fiber of the ETT propagates through the patient’s trachea and chest tissue and a portion of this light transmission is detected by the photodetectors of the detector array 102. Because of the position of the ETT and the transmission of the light through the tissue of the patient, each of the photodetectors will detect a different intensity of light. These differential intensities can be used by the computer as disclosed herein to monitor ETT position and to detect any ETT displacement.

[0050] As noted above, the detector array 102 is adhered to the skin of the patient in an attempt to position (called zero position) the “central” photodetector P2 on the point of emission of light from the optical fiber. This initial placement will maximize the intensity of NIR light detected by the “central” photodetector. Intensity is maximized as the emitted light has the shortest path through the patient’s trachea and skin to this photodetector. As the ETT displaces longitudinally, the signal data from a respective photodetector increases as the ETT displaces towards that photodetector and more light transmits to that photodetector. The signal data from a photodetector decreases as the ETT displaces away from that photodetector and less light transmits to that photodetector. Looking first at the longitudinal dimension, a graph 202 of a ratio of P1 / P2 of the signal data respectively from photodetectors Pl and P2 is plotted against a graph 204 of a ratio of P3 / P2 of the signal data respectively from photodetectors P3 and P2 is shown in Figure 4A. As can be seen in Figure 4A, as the ETT displaces towards Pl, the P1 / P2 ratio increases while the P3 / P2 ratio decreases. Conversely, as the ETT displaces towards P3, the P3 / P2 ratio increases while the P1 / P2 ratio decreases. This relationship can be modeled exemplarily using a fitted second-order polynomial for displacement estimation 206, 208.

[0051] A similar relationship is present for the detection of rotation of the ETT 10. As the ETT rotates, because photodetectors P4 and P5 arc orthogonal to the longitudinal axis of the ETT 10 and of the detector array 102, as the ETT 10 rotates (R) about its longitudinal axis, the light intensity from the optical fiber tip 32 increases at the photodetector P4, P5 in the direction of the rotation and decreases at the opposite photodetector. Comparative ratios of P4 / P2 and P5 / P2 exhibit rotation of the ETT as the ETT rotates in the direction of one or the other of P4 or P5, with the respective ratio value increasing as the ETT rotates towards the respective photodetector. In a further example, instead of using solely the P2 intensity value, an average or weighted average of the values for P2, P4, and P5 may be used as the denominator for the longitudinal displacement detection, while an average or weighted average of the values for Pl, P2, and P3 may be used as the denominator for the rotational displacement detection. It will be recognized that ETT displacement will likely include a longitudinal displacement component and a rotational displacement component. Therefore, ongoing monitoring of both of these considerations provides a more robust evaluation of ETT movement.

[0052] As discussed above, displacement of the ETT is estimated using voltages acquired from three photodetector sensors, Pl, P2, and P3, denoted by Vpi, Vp2, and Vp3, respectively. Two voltage ratios, ri2=Vpi / Vp2 and r32=Vpa / Vp2, are calculated as input parameters for an estimation model. Prior to real-time monitoring, calibration may be needed to account for the diverse thickness and properties of tracheal tissue across individuals. Calibration involves measurements performed at exemplarily five ETT locations, -10, -5, 0, +5, and +10 mm, measured by the reference position sensor. It will be recognized that this is merely exemplary, but and that more or fewer measurements may be taken and used for calibration purposes. The “0” position is established by vertically aligning the tip of the side-firing fiber with the P2 sensor, as confirmed by the attainment of the maximum reading in P2. A positive displacement is exemplarily induced by pushing the ETT inward, while pulling out the tube results in an exemplary negative displacement. Second-order polynomial functions are employed for calibration, which defines the relationship between voltage ratios and displacement. The visualization of the calibration procedure is illustrated in Figure 4B. In the negative displacement range, a polynomial model, f^l3h, is used to establish the relationship between r12and displacement d, based on calibration measurements at -10, -5, and 0 mm. Similarly, a low signal intensity model f2°wis obtained using r32values measured at the three positions. In the positive displacement range, a high signalintensity model / 32,5hand a low signal intensity model f^°ware fitted using calibration measurements r32and r12at 0, +5, and +10 mm, respectively.

[0053] For displacement estimation, the displacement direction (positive or negative) is determined by comparing r12(or r32) to its values at d = 0 mm r12(0) (or r32(0)). A ratio r12(d) / r12(0)>1 indicates an ETT movement towards the negative displacement (outward), whereas r12(d) / r12(0)<l suggests an inward movement or positive direction displacement, or the opposite way if 1' 2 is used. As the fitted calibration curves are monotonic within their applicable ranges, displacement estimation is performed by obtaining an estimated displacement duigh from the high signal model and obtaining an estimated displacement di_ow from the low signal model. The larger one of the two ratios r12and r32, is considered more reliable due to better signal-to-noise ratio (SNR) and is thus weighted more for displacement estimation. The final displacement estimation is a weighted averaging of these two estimations, formulated as dFmai=w-dHigh+(l-w)-dLoW, where the weight w was empirically set to 0.95.

[0054] Experimental animal results of an example of ETT position estimates using an example of the detector array system are shown in Fig. 12. The experiment used a linear stage for positioning an ETT displacement and the graph plots linear stage displacement against estimated displacement using the sensor of the detector array system example as described above with respect to Fig. 2. These experimental data show a high coefficient of determination of 0.986. A Bland- Altman analysis of this experimental data showed a decreasing bias with the mean displacement, approximately 1.5mm at -15mm and -1.0 mm at +15mm. The estimated bias was close to zero near 3mm mean displacement. As a result of this experiment, a high level of association was observed between the displacement obtained by the reference position sensor and the deviceestimated position.

[0055] Figure 5 depicts an example of a detection system 100 that uses a dual camera assembly 150 as the detector. In the detection system 100 of Fig. 5, the detector array 102 of Fig. 3A is replaced with the dual camera assembly 150 of Fig. 5. It will be recognized that like reference numerals between Figs.l, 3A and 5 exemplarily reference like structures and that the descriptions thereof are incorporated throughout the specification. The dual camera assembly 150 includes a visible light camera 152 and an NIR light camera 154.

[0056] The visible light camera 152 and the NIR light camera 154 are positioned together within a frame 156 and suspended approximately 0.5-2.0 m above the tracheal area of the patient.The visible light camera 152 operates to acquire visible light images within a visible field of view (FOV) 158. The NIR light camera 154 operates simultaneously to the visible light camera 152 to acquire NIR light images within an NIR FOV 160. The visible FOV 158 and the NIR FOV 160 are the same or nearly the same, so that visible light images and IR light images captured of the skin of the patient in the region of the tracheal during ETT placement and use may be registered to one another. The NIR light camera 154 obtains images of the NIR FOV 160 on the patient in which the NIR light from the optical fiber is visible after transmission through the tracheal tissue. The visible light camera 152 produces images of the visible FOV 158 which include the same or similar portion of the patient without catching the NIR light.

[0057] As previously described, the ETT includes an optical fiber configured to emit NIR wavelength light through the tracheal tissue of the patient. An image analysis process as described herein is performed to map the visible light images with the NIR light images to precisely locate the detected NIR pattern on the skin of the patient. Typically, motion of a sedated and intubated patient is minimal in the z-dimension (e.g. perpendicular to the skin), for example because the ETT is mostly tightly contained inside the trachea. Therefore, it may be generally assumed that any relative motion is in the x-dimension (e.g. along the ETT) and / or the y-dimension (e.g. perpendicular to the ETT).

[0058] Figure 6A depicts the camera geometry of the system depicted in Fig. 5, while Figure 6B depicts the patient and ETT geometry for the system depicted in Fig. 5. Both ETT displacement and patient movement can be causes of shift in the NIR light pattern on the patient skin captured by the NIR light camera 154, while the visible light camera 152 captures the patient movement only. Motion of the patient M and the ETT’s displacement relative to the trachea D are projected onto the image plane (i.e., the cameras) in pixel units as m and d, respectively. Through the geometry relationship:where / is the focal length, L is the distance between the camera and skin, and is the pixel size of the camera sensor. In a tested example, f= 15.59 mm and e = 3.0 pm. This geometry is exemplarily shown in Figure 6A. The horizontal orientation of the camera module aligns with the anterior-posterior axis of the patient, as illustrated in Figure 6B. The movement on the skin plane (in mm) and image plane (in pixels) can be decomposed into x and y components:and:(in = x ■ mr+ y ■ m..( )( d = x ■ dx+ y ■ dy7

[0059] The total movement of the NIR pattern A on the skin attributes to the combined effects of the actual ETT displacement relative to the trachea D and the pig’s movement artifact M, and is expressed as:A= M + D (4)

[0060] For detection of longitudinal ETT displacement along the trachea (x-direction), Equation (4) is reduced to:Ax= Mx+ Dx(5)Similarly, in the image plane:8 — m + d (6)

[0061] While a two-term Gaussian function may be used as a model (see Equation (8)) to approximate the NIR light intensity distributions in Monte Carlo simulations, actual NIR light intensity distributions are not so ideal. Factors, including but not limited to, exact ETT position, variation in tissue thickness and composition, skin conditions, and possible peak saturation due to overexposure, contribute to non-ideal IR light transmission and intensity distribution in the NIR light images.

[0062] Figure 7 presents examples of light shifts representing ETT displacement. Figures 7A, 7D, and 7G depict 2-dimensional light distribution in the NIR image. These respective images were exemplarily captured from the same patient at different displacements and with zero rotation. These NIR light distributions are converted to a 1 -dimensional intensity profile by totaling the pixel values either along the y (vertical) dimension as represented in Figures 7B, 7E, and 7H or along the x (horizontal) dimension as represented in 7C, 7F, and 71. It will further be recognizedthat Figs. 7B and 7C are the respective NTR intensity profiles for Fig. 7A; Figs. 7E and 7F are the respective NIR intensity profiles for Fig. 7D; and Figs. 7H and 71 arc the respective intensity profiles for Fig. 7G. The 1-dimensional intensity profile is fitted to a two-term Gaussian function as in Equation 8 to obtain a fitted profile, where ai, a.2, / 1 and U2 are constants that can be obtained through fitting of the experimental or calibration data. In Figures 7B, 7E and 7H, the fitted intensity profile is along the ETT therefore a shift in the position of the peak intensity represents an estimate and direction of the ETT longitudinal displacement. Similarly, in Figures 7C, 7F and 71, the fitted intensity profile is perpendicular to the ETT therefore a shift in the position of the peak intensity represents an estimate and direction of the ETT rotation. Figure 7J presents an exemplary plot of all NIR intensity peak positions in the longitudinal dimension against the ETT displacement measured by a reference sensor in one experiment.

[0063] As demonstrated in Figure 7C, 7F and 71, a rotational displacement of the ETT is represented in the NIR light images by a shift of the NIR pattern, which results in a shift in peak position of the 1 -dimensional intensity profile along the y-direction. Therefore, pattern shift in the 2-d image can provide an estimate and direction of the ETT displacement and rotation.

[0064] Figure 8 graphically depicts an example of a method 300 of estimating ETT longitudinal displacement Dxfrom both the NIR and visible light images. Figure 9 presents the method 300 as a flow chart. At 302 NIR pattern images are acquired. The FOV of the acquired images include the expected position of the ETT and the light from the tip of the optical fiber in the ETT. As seen in Fig. 8, the visible channel image is also acquired at the same time at 314, to exemplarily form an image pair of one NIR pattern image and one visible light image, both of which exemplarily having a similar FOV. Each new NIR pattern image (and associated visible image) acquired at 302 is, at the time of acquisition, the current image 304. That image is evaluated as described herein to establish a current point location of the ETT at 306. From the identified NIR pattern, the pixel intensities as exemplarily shown in Fig. 7 are fitted to a Gaussian curve, and the pixel location of the curve peak found as a current position, xcurrent.

[0065] To detect displacement, Dxat any time point, an NIR reference point is needed. A reference point may be established at the start of the method or updated at any time during the ETT monitoring process. The NIR pattern image captured at the first ETT position or any NIR pattern image captured between the first and current ETT position is used to establish a reference point, xref . If the current position point of xcurrent. Is is to be used as the new reference, the referencecan be set at 308 as a new reference point, xref. The first reference point can be initialized as the first acquired xcurre?lt and be a zero position of the NIR pattern image.

[0066] It is recognized that the pixel position of the curve peak of the first NIR pattern image may always be initialized as the first reference point (called zero position) and may be subsequently updated with future xcurrentvalues to ensure frame-to-frame similarities, particularly if ambient lighting fluctuates. The method 300 continues to acquire subsequent NIR pattern images 302 which are processed as the current images 306 and may be used to update the reference point at 308.

[0067] At 310 a raw ETT shift (in pixels) in the image plane is calculated as total movement of the NIR pattern:

[0068] However, the raw ETT shift 6Xmust be corrected for artifacts of patient motion mx. At 312 a corrected ETT shift dxis calculated according to the equation (14) below. The patient motion mxmust be removed from the total movement <5xin the image to derive actual displacement dxof the ETT.

[0069] At 314 visible spectrum images are acquired with the visible light camera 152 (fig. 5) of the same or nearly the same FOV as the NIR pattern images captured at 302. As noted, the visible spectrum images and the NIR pattern images are exemplarily captured simultaneously as an image pair representing generally the same FOV at the same point in time. The ETT tip and the optical fiber tip arc expected to be within the FOV of the visible spectrum images and the NIR pattern images.

[0070] At 316 the visible spectrum images are segmented, for example but not limited to the use of the Segment Anything Model (SAM, https: / / ai.meta.com / research / publications / segment-anything / ). The model is applied to each visible spectrum image to delineate pixels associated with the subject in the visible image from those pixels associated with the background. Other models or techniques will be recognized as being suitable for image segmentation based upon the present disclosure.

[0071] At 318 at least one, if not more reference feature locations are identified in the visible spectrum images. The reference feature locations in a visible spectrum image captured concurrently with the NIR pattern image currently being used for the reference point, xref as explained above, provide the feature reference locations at the zero ETT position, while thereference feature locations in the current visible spectrum image (captured concurrently with the current NIR pattern image) provide the current reference feature locations. In subsequent visible spectrum images acquired at 314, the reference feature locations 318 are compared at 320 to corresponding feature locations in the subsequently acquired visible spectrum images. The locations of the reference features between current and reference frames may be identified at 320 using a rapid local feature matching model, for example, LightGlue (https: / / arxiv.org / abs / 2306.13643), although other suitable matching models or techniques may be recognized as being suitable based upon the present disclosure. Among the matched point pairs, those identified subject pixels in both the current and reference images, are considered valid and included at 322 for estimating the patient’s movement mx. In total, N pairs of matched points are identified, and the mean disparity of their x-coordinate values constitutes the patient's motion in the longitudinal or ETT direction:where,x^urrentanddenote the x-coordinates, measured in pixels, of the i-th matched point pair in the current and reference frames, respectively. Both SAM and LightGlue have demonstrated zero- shot capabilities in various computer vision tasks. Therefore, their original model weights may be used for subject segmentation and feature points matching without additional training. Given that the distance between the visible and NIR cameras is considerably smaller than their working distance, and the camera specifications are similar, it is reasonable to assume that both cameras have a virtually identical field of view, though calibration could enhance accuracy.Returning back to 312, the raw ETT shift is corrected for patient motion artifacts to calculate a corrected ETT shift, measured in pixels, in the image plane is given by: c = 8r— mr(14)

[0072] At 324 the corrected ETT shift is used to calculate the final ETT displacement, measured in millimeters. The final ETT displacement is calculated from the corrected ETT shift dx via a calibration function:where Dref is displacement at the reference point from the zero position. At the start of the measurement, the zero position is designated as the reference point, i.e., Dref- 0 mm. The reference frame is replaced by the current frame for the next image frame, or Dref= Dx. If the camera’s focuslength / is fixed, the calibration function / cct((') is determined by the distance between the camera module and the trachea skin L, which can be fixed or accurately measured using various technologies. Lastly, at 326 the ETT displacement is compared to a predetermined displacement threshold. IF the ETT displacement exceeds the predetermined displacement threshold, then an alarm or an alert is produced to notify the care team that the ETT has displaced by an amount sufficient to warrant further clinician review, evaluation, and / or action. It will be recognized that the predetermined threshold displacement may be patient-specific or at least patient demographic or physiology specific, where an adult may experience a larger distance of ETT displacement before arising clinician concern as compared to a neonate.

[0073] It is recognized that the above description is made exemplarily for the X or axial dimension of the patient and in general alignment with the trachea of the patient. As described above, rotation of the ETT can also be detected using the same methods as described above while using the pixel intensity values in the Y or lateral dimension perpendicular to the X or axial dimension.

[0074] Figure 10 presents experimental data showing the correlation between the determined pixel position shift and the longitudinal displacement of the ETT measured by the reference sensor, showing R2values between 0.95 and 0.99. The dual-camera detection system as disclosed herein thus experimentally produces a highly correlative measurement of ETT longitudinal displacement.

[0075] One challenge of the dual-camera detection system, as noted above, is the compensation for patient motion artifacts. Once the visible image data is mapped to the NIR image data, the visible image data can be used to determine patient movement M and calculate any corresponding movement of the light spot in the NIR image data m. At least two approaches are contemplated within the present disclosure and may be used within the method 300 as described above. In a mark-based approach, a natural or intentionally placed visible mark on the skin of the patient in the vicinity of the trachea is identified in the visible images (See Figure HA). Movement of this mark in the visible images is further identified and tracked though subsequent images. Movement of the mark is extrapolated to correct movement artifacts in the NIR image data. In examples, the mark may be a sticker or other marker of a defined size. This reference of a known size provides a scale for the movement measurements. In an alternative approach (called subjectbased motion detection method), the subject in the visible images is identified and heterogenousskin movement of the patient is determined using a Block Matching method. (See Figure 1 1B). Block matching divides a current frame into macroblocks and compares each of the macroblocks with a corresponding block and its adjacent neighbors in a temporally nearby frame. A vector is created that models the movement of the macroblock from one location to another. This movement calculated for all the macroblocks in a frame constitutes the motion estimated in a frame. Other motion estimation techniques may be used including but not limited to differential, pel-recursive, optical flow, and phase-correlation. Figure 11C provides estimated motion results on a pixel-by- pixel basis between an application of the mark-based motion detection and a subject-based motion detection. It can be recognized that the direction (+ / -) of the body movement is consistent between the two approaches as shown in Figure 11C even despite differences in magnitude of the motion. Figure 11D is a scatter plot of estimated motion artifacts from the mark-based determination against the estimated motion from the subject-based determination.

[0076] In additional applications, the dual-camera detection system may also be used to confirm the position of the ETT tip during placement. The dual-camera detection system generally provides a larger field of view as compared to the detector array detection system. This larger field of view includes an expanded portion of the patient’ s neck and chest, even including the patient’ s jaw and / or mouth. Because of this larger field of view, progression of the intubation may be identified and tracked.

[0077] Some examples of the systems and methods as described herein provide an advantage of continuously monitoring ETT position and / or displacement. Previous solutions relied upon X-ray, ultrasound, or fluoroscopy confirmation of ETT placement and position. In the case of X-ray and fluoroscopy, this exposes the patient to additional radiation dose. Ultrasound confirmation requires personnel with skills in ultrasonic imaging and, like X-ray, only provides a temporal snapshot of ETT position. The systems as described herein provide real-time and continuous monitoring of ETT position. The systems as described herein may be connected to an audio, visual, or push-notification alarm which is exemplarily activated by the computer in response to a detected threshold of ETT position change. In an example, this threshold may be a detected displacement of 5mm or may be a smaller or larger value than 5mm. In examples, this may depend upon whether the patient is an adult, a child, or an infant.

[0078] In a still further example, the detection system, whether it is the sensor array detection system or the dual-camera detection system, possesses pulse artifact in the detection- l-signal. This pulse artifact may be in the signals acquired by the photodetectors or may be in the NIR spectrum images. This pulse artifact may be analyzed by the computer to isolate and detect the patient’s pulse. Such leveraging of the detection systems as described herein, provide an additional source of a supplemental pulse measurement. The pulse measurement may further provide a leading indicator of a change in ETT placement, if for example, ETT displacement is causing sympathetic distress.

[0079] In a still further example, the light source 36 and / or illumination system 12 (e.g. Figs. 1, 3 A, and 5) may further include one or more additional light sources or otherwise be configured to emit multiple wavelengths of light. The light source may include one or more LEDs configured to emit one or more different wavelengths of light. These one or more additional wavelengths of light may be transmitted along separate additional optical fibers 16, although a more efficient configuration uses the optical fiber 16 with the different wavelengths of light sequentially pulsed down the optical fiber 16. The detector, either the detector array 102 or the camera assembly 150, is configured with a detection or frame rate to sufficiently acquire the different light wavelengths from the light source 36 as projected from the tip of the optical fiber 16. The computer 106 and detector array 102 may be synchronized with the light source 36 such that detection of the sequential wavelengths are expected. The camera assembly 150 may further have a frame rate sufficiently fast such that one or more image frames are acquired from each sequential wavelength emitted from the light source 36 and the tip of the optical fiber 16.

[0080] In an example of the above, the light source 36 is configured to emit an NIR wavelength between 800nm - 810nm, as well as emit red wavelength exemplarily between 650nm - 780nm and an NIR wavelength exemplarily between 830nm - 950nm. These three wavelengths are emitted sequentially. The 800nm - 810nm wavelength is used for ETT detection as describe above, while the other two emitted wavelengths are used to improve pulse detection or other physiological parameters, including but not limited to blood and / or tissue oxygen saturation, and pulse oximetry (SpCh).

[0081] In the above description, certain terms have been used for brevity, clarity, and understanding. No unnecessary limitations are to be inferred therefrom beyond the requirement of the prior art because such terms are used for descriptive purposes and are intended to be broadly construed. The different systems and method steps described herein may be used alone or incombination with other systems and methods. It is to be expected that various equivalents, alternatives, and modifications arc possible within the scope of the appended claims.

[0082] The functional block diagrams, operational sequences, and flow diagrams provided in the Figures are representative of exemplary architectures, environments, and methodologies for performing novel aspects of the disclosure. While, for purposes of simplicity of explanation, the methodologies included herein may be in the form of a functional diagram, operational sequence, or flow diagram, and may be described as a series of acts, it is to be understood and appreciated that the methodologies are not limited by the order of acts, as some acts may, in accordance therewith, occur in a different order and / or concurrently with other acts from that shown and described herein. For example, those skilled in the art will understand and appreciate that a methodology can alternatively be represented as a series of interrelated states or events, such as in a state diagram. Moreover, not all acts illustrated in a methodology may be required for a novel implementation .

[0083] This written description uses examples to disclose the invention, including the best mode, and also to enable any person skilled in the art to make and use the invention. The patentable scope of the invention is defined by the claims and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal languages of the claims.

Claims

CLAIMS1. A system for endotracheal tube placement confirmation and / or displacement monitoring, the system comprising: an endotracheal tube comprising an optical fiber running substantially along a length of the endotracheal tube, the endotracheal tube having a insertion end and the optical fiber comprising a tip proximate to the insertion end of the endotracheal tube; a source of near-infrared (NIR) wavelength light operably connected to the optical fiber; a detection system spaced apart from the endotracheal tube and the optical fiber, the detection system configured for noninvasive detection of NIR wavelength light emitted from the optical fiber; and a processor communicatively connected to the detection system and configured to receive data from the detection system to determine an initial position of the tip of the optical fiber and to continuously monitor a subsequent position of the tip of the optical fiber.

2. The system of claim 1, wherein the tip of the optical fiber is inset from the insertion end of the endotracheal tube.

3. The system of claim 2, wherein the tip of the optical fiber is inset by 1 cm from the insertion end of the endotracheal tube.

4. The system of claim 1, wherein the tip of the optical fiber comprises a 45° angle and a mirror.

5. The system of claim 1, wherein the NIR wavelength light is between 660nm - 2500nm, optionally the NIR wavelength light is between 780nm-850nm, optionally, the NIR wavelength light is around 810nm, the NIR wavelength light is optionally 660nm, 780nm, 800nm, 805nm, 810nm, 830nm, or 850nm.

6. The system of claim 1 wherein the detection system comprises: an array of photodetectors, the array of photodetectors comprising:a first photodetector configured to be positioned on the tracheal skin aligned to the tip of the optical fiber; second and third photodetectors linearly aligned with the first photodetector along a longitudinal dimension; and fourth and fifth photodetectors linearly aligned with the first photodetector along a lateral dimension, the lateral dimension being perpendicular to the longitudinal dimension; and an adhesive layer configured to secure the array of photodetectors to skin of a patient.

7. The system of claim 6, wherein the processor is communicatively connected to the first, second, and third photodetectors, and the processor is configured to calculate a first ratio of a first signal from the first photodetector to a second signal from the second photodetector and a second ratio of a third signal from the third photodetector to the second signal and to calculate a longitudinal displacement of the endotracheal tube from at least one of the first ratio and the second ratio.

8. The system of claim 7, wherein the processor is communicatively connected to the fourth and fifth photodetectors, and the processor is configured to calculate third ratio of a fourth signal from the fourth photodetector to the second signal and a fourth ratio of a fifth signal from the fifth photodetector to the second signal and to calculate a rotational displacement of the endotracheal tube from at least one of the fourth ratio and the fifth ratio.

9. The system of claim 7, wherein the first and second ratios are further calculated from an average of the second, fourth, and fifth signals as the second signal.

10. The system of claim 7, wherein the third and fourth ratios are further calculated from an average of the second, first, and third signals as the second signal.

11. The system of claim 1 wherein the detection system comprises: an infrared light camera configured to provide infrared images to the processor; and a visible light camera configured to provide visible images to the processor.

12. The system of claim 11 , wherein the processor is configured to calculate total pixel intensities in a horizontal dimension in the infrared images against pixel position in a vertical dimension and the processor is configured to determine a rotation of the endotracheal tube from shift in the peak pixel position of the intensity profile from a reference peak pixel position.

13. The system of claim 11, wherein the processor is configured to calculate total pixel intensities in a vertical dimension in the infrared images against pixel position in a horizontal dimension and the processor is configured to determine an endotracheal tube displacement from shift in the peak pixel position of the intensity profile from a reference peak pixel position.

14. The system of claim 12 or 13, wherein the processor is configured to fit a Gaussian distribution to the total intensity profile from which the peak position is determined.

15. The system of claim 12 or 13, wherein the processor is further configured to associate a visible image of the visible images to a corresponding NIR image of the infrared spectrum images collected concurrently, the processor is configured to determine an artifact patient motion between visible spectrum images, and the processor is configured to calculate a corrected endotracheal tube displacement / rotation from the endotracheal tube displacement / rotation and the artifact patient motion.

16. The system of claim 15, wherein the processor is configured to identify a reference mark in the visible images and the processor calculates a shift in the reference mark between visible images as the artifact patient motion.

17. The system of claim 15, wherein the processor is configured to apply image segmentation to the visible images, and to identify corresponding feature points between visible images, and to calculate a mean disparity between the corresponding feature points between the visible images as the artifact patient motion.- l-18. The system of claim 11 , wherein the processor is further configured to identify a pulse artifact within a sequential series of the infrared images.

19. The system of claim 1, wherein the optical fiber is configured to transmit at least one additional wavelength of light, the at least one additional wavelength of light being between 660nm-780nm.

20. The system of claim 19, wherein the optical fiber is configured to transmit at least one further wavelength of light, the at least one further wavelength of light being between 830nm-950nm.

21. The system of any of claims 19-20, wherein each of the wavelengths of light are transmitted sequentially, and the detection system and the processor are configured to detect each of the wavelengths of light.

22. The system of claim 21, wherein the processor is configured to determine a physiological parameter from the at least one additional wavelength of light and the at least one further wavelength, optionally wherein the physiological parameter is blood oxygenation.

23. The system of claim 1, wherein the optical fiber is inserted into a groove or channel of the endotracheal tube.

24. A method of estimating displacement of an endotracheal tube comprising an optical fiber configured to emit near-infrared light, the method comprising: obtaining a first infrared image and a subsequent second infrared image with an infrared camera having a field of view comprising an expected endotracheal tube position; obtaining a first visible image and a subsequent second visible image with a visible camera having a field of view comprising the expected endotracheal tube position; calculating a total pixel value in a first dimension of the first infrared image and the second infrared image;identifying a first pixel location in a second dimension that corresponds to a peak total pixel value in the first dimension for the first infrared image and identifying a second pixel location in the second dimension that corresponds to a peak total pixel value in the first dimension for the second infrared image; calculating an endotracheal tube shift from a difference between the first peak pixel location and a second peak pixel location; identifying matched point pairs between the first visible image and the second visible image; calculating a mean disparity between the matched point pairs to calculate a patient motion artifact; calculating a corrected endotracheal tube shift from the endotracheal tube shift and the patient motion artifact; applying a calibration function to the corrected endotracheal tube shift to calculate an endotracheal tube displacement; and comparing the endotracheal tube displacement to a predetermined displacement threshold value, and producing an alarm if the endotracheal tube displacement exceeds the threshold value.

Images