Imaging apparatus for wound analysis

Abstract

A hand-held imaging apparatus for characterization of infection at a wound site has a radiant source configured to direct light having an excitation wavelength toward tissue at the wound site. One or more cameras are configured to acquire spectral image content from fluorescence response of the infected tissue at the wound site to the excitation wavelength. An enclosure having a recessed surface side is configured to house the radiant source and optics for the one or more cameras. A display surface side, opposite the recessed surface side, is configured to mount one or more displays that show the camera field of view prior to wound imaging and show processed spectral content from wound fluorescence following image capture. A processor is in signal communication with the one or more cameras and is configured to analyze the spectral image content to identify one or more microbial infections.

Description

IMAGING APPARATUS FOR WOUND ANALYSIS

[0001] This is an International patent application claiming priority from US provisional application No. 63 / 730,526 filed 11 December 2024 which is hereby incorporated by reference in its entirety.TECHNICAL FIELD

[0002] The present disclosure generally relates to methods and apparatus for wound assessment and analysis using fluorescence imaging and more particularly to portable devices that automate wound analysis according to spectral characteristics across a range of bandwidths. The apparatus and method may be suitable for assessing and documenting pressure ulcers, leg ulcers, diabetic foot ulcers and other chronic wounds.BACKGROUND

[0003] Effective wound assessment and treatment is recognized to be an important part of patient care that presents special challenges to the medical practitioner. In contrast to significant advances in other areas of treatment, wound analysis is largely based on visual observation, and in numerous cases effective treatment is delayed and significant information is not collected that might help to increase diagnostic accuracy and speed the process of repair and recovery. Among recognized shortcomings of existing wound analysis practice is a lack of uniform assessment tools and unavailability of suitable portable solutions for collecting and analyzing wound metrics.

[0004] Assessment of infection presents a special problem. Wound infection can be difficult to detect in its early stages, so that suitable treatment may be delayed or deferred, complicating the treatment task. Methods for assessment can be random and haphazard; sampling methods can present risks of spreading infection. Inaddition, once infection has been detected, it can be difficult to assess the effectiveness of wound cleansing or other treatment regimen for infection.

[0005] Imaging solutions have been proposed for taking advantage of the fluorescence response of many bacteria to stimulating light at short, high energy wavelengths, including intense blue (near-UV) and ultraviolet (UV) light. It has been shown that different strains of bacteria exhibit different spectral “signatures”, allowing spectral analysis apparatus to be used for bacterial detection and identification. However, there are a number of drawbacks to proposed applications. In order to accurately capture the spectral information that is generated from UV or near-UV excitation, proposed systems typically utilize a series of narrow-band filters that must each be successively and rapidly switched into place in order to detect energy at various wavelengths. Sensitive filters can be costly; apparatus for rapid switching of filters introduces complex mechanical components and further drives up cost. Ambient lighting is best reduced or eliminated with filtered light approaches; otherwise the resulting energy degrades the signal-to-noise level needed for diagnostic accuracy; this requirement for reduced ambient light works against visibility of the wound and can be impractical or even unsafe within the clinical environment.

[0006] Characteristics of a suitable system solution can include: Portability. A hand-held measurement apparatus would allow access the wound site with little or no discomfort to the patient. Compactness.Usable in ambient light, not interfering with other tasks and functions for medical personnel and patient.- Accuracy and repeatability.- Processing speed, with results displayed as the wound is being illuminated. Real-time or near-real-time results reporting would allow the practitioner to readily view and review the wound site with the assistance of processed information.No need for paper rulers or markers.Protection from radiation harmful to the eyes.

[0007] Thus, it can be seen that there is need for a wound analysis apparatus that can be used to aid diagnostic assessment and treatment.SUMMARY

[0008] It is an object of the present disclosure to advance the art of wound analysis, particularly with respect to detecting infection.

[0009] The present disclosure provides a hand-held imaging apparatus for characterization of infection of tissue at a wound site, the apparatus comprising: a radiant source configured to direct light having an excitation wavelength toward the tissue at the wound site, one or more cameras configured to acquire spectral image content from fluorescence response of the infected tissue at the wound site to the excitation wavelength, an enclosure having a recessed surface side configured to house the radiant source and optics for the one or more cameras and having a display surface side, opposite the recessed surface side and configured to mount one or more displays that show the camera field of view prior to wound imaging and show processed spectral content from wound fluorescence following image capture; and a processor that is in signal communication with the one or more cameras and that is configured to analyze the spectral image content to identify one or more microbial infections.The present disclosure further provides a hand-held imaging apparatus for characterization of infection at a wound site, the apparatus comprising: a radiant source configured to direct light having an excitation wavelength toward tissue at the wound site,one or more cameras configured to acquire spectral image content from a fluorescence response to the excitation wavelength of infected tissue at the wound site, an enclosure having a first surface side configured to house the radiant source and optics for the one or more cameras and having a display surface side, opposite the first surface side and configured to mount one or more displays that show the camera field of view prior to wound imaging and show processed spectral content from wound fluorescence following image capture; and a processor that is in signal communication with the one or more cameras and that is configured to analyze the spectral image content to identify one or more microbial infections.The first surface side can be a recessed surface side.Preferably the radiant source generates UV light below a wavelength of about 450 nm, preferably below about 410 nm and preferably at a wavelength of about 405 nm. Advantageously the one or more cameras include a multi-spectral camera, a hyperspectral camera, a thermal camera, a reflectance camera, and a plenoptic camera.Conveniently the radiant source comprises at least one, preferably an integer from 2 to 8 and preferably four LEDs.Preferably the LED or LEDs emit monochromatic light with a wavelength of about 405 nm.Advantageously the radiant source is configured to be selectively activated during image acquisition, enabling the detection of bacterial fluorescence.Preferably the apparatus further comprises a visible light source at the first surface side.Conveniently the visible light source comprises at least one LED and preferably comprises a LED ring.Advantageously, the one or more cameras comprises a 3D camera for wound dimension analysis.Optionally the 3D camera and / or the processor is configured to generate a wound surface image by generating point clouds and constructing triangular or polygonal meshes of neighboring points.Preferably the one or more cameras comprises an RGB imaging module or camera configured to capture a color image of the wound surface.Optionally the processor is configured to analyze the color image using artificial intelligence to perform tissue classification, optionally providing classification as granulation tissue, fibrin / slough and / or necrotic tissue.Conveniently the one or more cameras comprises a thermal sensor configured to provide a temperature map of the wound and surrounding skin.Advantageously the temperature map highlights any temperature variation, to enable a clinician to identify signs of inflammation and / or infection risk.Preferably the one or more cameras is configured to detect fluorescence in the spectral range of approximately 620-700 nm or approximately 490-560 nm. Conveniently the one or more cameras employs at least one narrow-band filter to capture fluorescence in the spectral range of approximately 620-700 nm or approximately 490-560 nm while suppressing reflected excitation light and background signals from surrounding tissue.In a preferred arrangement the radiant source comprises at least one LED and the apparatus comprises: at least one white light source; at least one multispectral or hyperspectral camera configured to acquire spectral image content from a fluorescence response to the excitation wavelength; at least one RGB imaging module or camera; at least one 3D camera; and at least one thermal camera or thermal module.The present disclosure also provides a method of documenting a wound comprising:a) providing an apparatus as described herein and as claimed, b) selecting a wound for documentation, c) analyzing a wound dimension using a 3D camera, d) capturing an RGB image of the wound using an RGB camera, e) providing light to the wound from a radiant source, the light having an excitation wavelength, f) acquiring a spectral image from a fluorescence response to the excitation wavelength; and g) presenting captured images to a user.The present disclosure also provides a method of documenting a wound comprising: a) providing an apparatus as described herein and as claimed, b) selecting a wound for documentation, c) in ambient light and / or visible light providing light to the wound from a radiant source, the light having an excitation wavelength, d) acquiring an excitation image comprising a fluorescence response to the excitation wavelength and ambient and / or visible light; and e) either before or after steps c and d, acquiring a reference image without illumination from a radiant source, the second image only containing ambient and / or visible light, f) subtracting the reference image from the excitation image to provide an image representative only of fluorescence response.Advantageously the method further comprises enhancing the image representative only of fluorescence response, optionally using artificial intelligence, and distinguishing different fluorophores, such as porphyrin-associated red fluorescence, and / or pyoverdine-associated green fluorescence to produce an enhanced fluorescence map.Conveniently the method further comprises quantifying the enhanced fluorescence map pixel-by-pixel to compute a bacterial burden score, for example:• total fluorescence area,• mean intensity,• maximum local intensity,• spatial distribution patterns.The present disclosure also provides a method of documenting a wound comprising: a) providing an apparatus as described herein and as claimed, b) selecting a wound for documentation, c) acquiring a thermal image of the wound and surrounding skin to measure absolute temperature and temperature gradients; d) acquiring a fluorescence image under radiant light with an excitation wavelength to reveal the presence of bacterial fluorophores such as porphyrins or pyoverdines, e) capturing a RGB image to extract texture features such as colour patterns, surface irregularities, tissue roughness, and the visual characteristics of the wound edge f) isolating the wound bed and surrounding peri wound tissue; g) combining with a multimodal fusion engine:• local temperature elevations or asymmetries,• fluorescence signal intensity and distribution,• texture-based indicators of inflammation or tissue breakdown; h). processing the resulting fused data through a machine-learning classifier trained to recognize infection-specific patterns.Preferably the excitation wavelength is below a wavelength of about 450 nm, preferably below about 410 nm and preferably at a wavelength of about 405 nm. These objects are given only by way of illustrative examples, and such objects may be exemplary of one or more embodiments of the disclosure. Other desirable objectives and advantages inherently achieved by the disclosure may occur or become apparent to those skilled in the art. The invention is defined by the appended claims.BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The foregoing and other objects, features, and advantages of the disclosure will be apparent from the following more particular description of the embodiments of the disclosure, as illustrated in the accompanying drawings. The elements of the drawings are not necessarily to scale relative to each other.

[0011] FIG. 1 is a view that shows a hand-held imaging apparatus for wound assessment according to an embodiment of the present invention.

[0012] FIG. 2 is a perspective view of the hand-held imaging apparatus for wound assessment according to an embodiment of the present invention.

[0013] FIG. 3 is a plan view that shows the underside of the hand-held imaging apparatus with camera and light sources.

[0014] FIG. 4 is a perspective view that shows the underside of the hand-held imaging apparatus with camera and light sources.

[0015] FIG. 5 is a plan view that shows the underside of the hand-held imaging apparatus with camera and light sources.

[0016] FIG. 6 is a simplified block diagram of handheld imaging apparatus 10 according to an embodiment of the present disclosure.DETAILED DESCRIPTION

[0017] Figures provided herein are given in order to illustrate principles of operation and component relationships according to the present disclosure and are not drawn with intent to show actual size or scale. Some exaggeration may be necessary in order to emphasize basic structural relationships or principles of operation. In the drawings and text that follow, like components are designated with like reference numerals, and similar descriptions concerning components and arrangement or interaction of components already described may be omitted.

[0018] Where they are used, the terms “first”, “second”, and so on, do notnecessarily denote any ordinal or priority relation, but may be used for more clearly distinguishing one element or time interval from another. The term “plurality” means at least two.

[0019] In the context of the present disclosure, positional terms such as “top” and “bottom”, “upward” and “downward”, and similar expressions are used descriptively, to differentiate different surfaces or views of the Applicant device, with its standard orientation.

[0020] In the context of the present disclosure, the term “coupled” is intended to indicate a mechanical association, connection, relation, or linking, between two or more components, such that the disposition of one component affects the spatial disposition of a component to which it is coupled. For mechanical coupling, two components need not be in direct contact, but can be linked through one or more intermediary components.

[0021] The term “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any aspect described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other aspects.

[0022] In the context of the present disclosure, the term “subject” may be used to designate the wound that is being examined and imaged. In camera optics terminology, this subject designation can correspond to the optical “object” that is imaged.

[0023] In the context of the present disclosure, ultraviolet (UV) light can have a range of wavelengths, smaller than about 410 nm.

[0024] The UV and near-UV illumination that is directed to tissue at the wound site has a wavelength that is centered about a nominal peak value of 405 nm. The full width at half-maximum (FWHM) value for the UV illumination is within the range of + / - 10 nm of the peak value.

[0025] FIG.1 shows a handheld imaging apparatus 10 configured for woundanalysis according to an embodiment of the present disclosure. Imaging apparatus 10 has one or more on-board cameras to capture an image of the wound and provides visible light illumination for general visibility in using the imaging apparatus 10 as well as illumination sources at wavelengths outside the visible spectrum for identifying an infection type. Results of wound analysis are available on a display 12, which can also provide a user interface for control of apparatus 10 and can also show relevant information, such as data from patient history, for example.

[0026] FIG. 2 is a perspective view of apparatus 10 from the display 12 surface side.

[0027] FIGs. 3 - 5 show the recessed underside of apparatus 10 that is configured to face the wound area and that houses radiant UV / near-UV light sources 22, which can be LEDs, one or more visible light sources 28 which can be LEDs, and lens optics for a sensor array 20 containing multiple cameras. Structurally, as best shown in FIG. 4, sensor array 20 is recessed between grips or supports 24 that provide convenient handles for controlling distance and direction for imaging. According to an embodiment of the present disclosure, sensor array 20 can have multiple cameras of different types. This can include any of the following, as shown in the example of FIG. 5:(i) 3D light field or plenoptic camera 32a. Plenoptic imaging acquires light field content from the object, recording light intensity and angular information that can be sampled and post-processed in order to view the image content at a range of focus lengths and at variable depth of field. The plenoptic camera is also advantaged for use at lower light levels.(ii) Reflectance light (RGB) camera 32b, useful for obtaining color images of the subject wound surface, allowing proper aim and focus for optimizing use of the imaging apparatus.(iii) Multi-spectral camera (not shown) that uses filters for spectral characterization of the light.(iv) Hyperspectral camera 32c that obtains a continuous spectral record of image content.(v) Thermal camera 32d for acquiring a heat map that can show infection having a variety of causes, and other conditions.Visible light sources 28, shown as a surrounding LED ring that encircles lens 26 of sensor array 20 in FIG. 3, working with the reflectance light camera and the display 12, assist the user in accurately positioning the sensor array 20 over the wound site. Lens 26 is configured for use over a range of wavelengths, from near-UV to infrared (IR). Lens 26 can further include a coating suitable for filtration of ambient light or other unwanted wavelengths.System Components

[0028] FIG. 6 shows a simplified block diagram of handheld imaging apparatus 10 according to an embodiment of the present disclosure. A control logic processor (CPU or simply processor) 60 controls operation of sensor array 20 cameras 32a - 32d. Image capture can be initiated through an operator interface 66, which can be a touch screen interface on display 12, or manual pushbuttons or other controls, or some combination of interface tools, as is well known in the digital devices art. Processor 60 also controls operation of radiant UV / near-UV sources 22 and visible light sources 28 for providing illumination to the subject region. Programmed logic stored on or accessible to processor 60 operates on the image data, interpreting the image content accordingly. A memory 62 can be provided on the device itself or can be remotely located, such as using a wireless connection through communications circuitry 64. Remote connection can allow storage of patient data as well as providing additional processing and communication with other sites, such as for identifying unknown infectious bacteria types, for example.

[0029] As noted previously, sensor array 20 can include a number of cameras that can operate independently or cooperatively, such as for simultaneous imaging of the wound site. Sensor array 20 can include a multispectral or a digital hyperspectralcamera 32c that captures image content that provides both spatial and spectral data related to a wound. Advantages of hyperspectral imaging include simplified operation, since the camera itself obtains the spectral content, with image content differentiated and stored over predetermined, built-in wavelength ranges.

[0030] Sensor array 20 can optionally include a multi-spectral camera that employs a bank of filters, each filter successively switched into position for image captures in rapid sequence. The resulting image sequence obtains a set of images of the same field of view, wherein the images are very nearly simultaneous.

[0031] Sensor array 20 can include plenoptic or light-field camera 32a that acquires a single image and records light field data that is accessible for forming image content at different focus settings.

[0032] Sensor array 20 can include thermal camera 32d that generates a heat map of the wound and surrounding skin. The resulting thermal image highlights temperature variations, enabling clinicians to identify signs of inflammation or infection risk.

[0033] Temperature values may be overlaid onto the wound image for combined clinical assessment, with a reference line indicating the mean temperature across the analyzed region.

[0034] Sensor array 20 may include reflectance light (RGB) camera or module 32b. The RGB module 32b captures high-resolution color images of the wound surface. These images may be analyzed by an artificial intelligence model, which processes the data to perform tissue classification, differentiating between granulation tissue, fibrin / slough and necrotic tissue.

[0035] The Al output includes a percentage breakdown of each tissue type within the wound area, visualized through color-coded overlays. All calculated values can subsequently be manually adjusted and stored, ensuring flexibility and traceability in wound documentation.

[0036] To the operator, sensor array 20 can appear to operate in similar manner to cameras integrated into cell phones, tablets, and other personal, hand-held devices. Operational features of sensor array 20 operation can include:Auto-focus;Brightness control or adjustment;Recognition software for determining the subject;Still or video options;Optional filter for reducing ambient light.

[0037] According to a preferred arrangement, sensor array 20 includes hyperspectral camera 32c, configured to obtain image content according to wavelength. Unlike multi-spectral cameras that employ a bank of filters and are limited to detecting a limited number of separate wavelengths in the imaged object, hyperspectral imagers generally use diffraction for spectral separation. This allows separation of the light to a number of contiguous wavelength bands, based on the desired spectral resolution. The hyperspectral imager provides a plane of x-y image pixels having image content in each of a number of contiguous wavelength bands, facilitating analysis of the material and chemical composition of the imaged object. In this way, hyperspectral camera 32c simultaneously performs both imaging and spectroscopy for the imaged wound site, enabling the assessment of tissue qualities as well as the presence of microbial infection, due to fluorescence response of bacteria and bacterial metabolic by-products of different types. Trained machine logic software can readily identify the bacteria type based on spectrographic analysis of response to high-energy light exposure, such as UV exposure.

[0038] As a hyperspectral imager of sensor array 20, camera populates a data cube, each image capture generating multiple sets of spectral data content for the same spatial image. Each set of spectral data can be distinguished from other sets according to wavelength band.

[0039] Hyperspectral camera 32c of sensor array 20 can be a scanning type that captures a portion of the field of view at a time or can be a non-scanning type.Unwanted wavelength ranges, not of interest for bacterial detection, can be filtered out from the image content. By isolating and amplifying this bacteria-associated emission, the system enables precise visualization of bacterial hotspots that exceed clinically relevant colonization thresholds (>104CFU / g). This targeted approach ensures that only diagnostically relevant fluorescence is detected and highlighted, providing clinicians with a reliable tool for assessing bacterial burden in wounds.

[0040] Image content from two or more cameras 32a - 32d can be combined in various ways for obtaining more comprehensive information for characterizing bacteria type and status of the infection. Temperature map information from thermal camera 32d can be combined with spectral information from hyperspectral camera 32c, for example, to not only determine a type of bacterial infection, but also the relative distribution and activity of the infectious organism. Plenoptic camera image content can be used in conjunction with hyperspectral image content in order to determine the relative position of infectious bacteria to the skin surface or to tissue at particular depths, for example. Combined information can be particularly useful as input to machine learning logic that is trained for decision-making when presented with disparate types of data.

[0041] Display 12 can be a touch screen display that allows operator command entry and scrolling behavior, similar to display behavior for cell phone and other personal electronics devices.

[0042] The visible light source 28 can be provided by a ring LED, as noted above, or from some other suitable type of low-power, high lumen illumination. In operation, visible light can be used for initial alignment of the sensor array 20 optics to the wound site, so that hyperspectral imaging can then be executed. A ring LED may be particularly advantageous because it has been found to provide shadow- free illumination with consistent quality across different clinical environments. Illumination intensity of the visible light source 28 can be adjusted in multiple steps such as 25%, 50%, 75% and 100% of maximum intensity to adapt to different wound conditions and ambient lighting in the clinical environment.

[0043] Radiant light source 22 provides UV light or near-UV light over a narrow waveband. This light, directed toward the wound region, causes excitation of various types of fluorescing bacteria, and fluorescent bacterial metabolic by-products such as porphyrins and pyoverdines, with spectral results varying according to bacteria type. According to an embodiment, emitted light for excitation is at 405 nm peak value. Fluorescence response from living tissue and bacteria generally occurs at higher wavelengths (lower frequencies) such as around 490-560 nm for pyoverdines and 620-700 nm for porphyrins. Illumination with radiant light source 22 can be selectively activated during image acquisition, enabling the detection of bacterial fluorescence in supplementary captures.

[0044] Visible and radiant light sources can be configured for different illumination modes including continuously energized, pulsing, timed, etc.

[0045] FIG. 7 shows, in exploded- view format, the arrangement of some of the more prominent internal components of imaging apparatus 10 according to an embodiment of the present disclosure. Covers 70 and 72 provide an encasement for cameras of sensor array 20 and associated light sources and logic processor, as described with reference to FIG. 6. Cover 70 sets a minimum distance to the wound site for imaging apparatus 10 when placed near the body of the subject. Sensor array 20 components are mounted on a circuit board 30 and associated lens 26 optics seat against cover 70. A display cover 72 supports one or more thin display 12 surfaces and is fastened against cover 70 to provide enclosure for inner components of the system.

[0046] The encasement or housing provided by covers 70 and 72 is designed for clinical robustness. It is sealed and may be disinfected, drop-resistant, and suitable for use in high-throughput environments such as hospitals, outpatient clinics, and nursing homes. Its ergonomic design ensures easy handling and compliance with hygiene standards. The device may feature a two-stage power switch system, with both switches sealed against each other by a pressure-resistant membrane to ensure durability and hygiene safety. Additionally, a side-mounted bracket may be providedto allow the device to be securely hung when not in use, providing convenient storage in clinical workflowsWavelengths

[0047] Wavelengths emitted by bacterial and tissue components upon exposure to UV or near-UV light can span a broad range of colors over the visible spectrum, including violet, blue, green, yellow, and red, for example. To optimize viewing, some filtering of the emitted light can be performed. Spectral response of particular interest for bacterial detection can include light within these ranges, for example:

[0048] 500 - 545 nm for cyan and green effects;

[0049] 600-670 nm for red and orange content.

[0050] The spectral range for particular bacteria can shift for detecting different types of organisms.Processing the Spectral / Hyperspectral Data

[0051] In order to provide an assessment of wound status that can be useful for diagnosis, processor 60 analyzes the data cube that is acquired for the image capture. Relevant wavelength bands are identified, and the combination of signals within each of the wavelength bands can be checked to identify various types of bacteria that can be present.

[0052] According to an embodiment, machine learning (“artificial intelligence”) can be provided in order to facilitate bacteria identification.

[0053] Typically, machine learning techniques employ multi-layer neural network architectures and can employ associated techniques that detect higher-level patterns from a volume of low-level data such as “deep learning”, for example. Thus, in the context of the present disclosure, terms such as “machine learning”, “artificial intelligence”, “deep learning”, and “neural networks” can be used equivalently to describe trained logic from various related aspects.

[0054] Training of a machine learning apparatus can use a training set of knownexamples for generating the Al decision-making logic.Power

[0055] Power can be provided to imaging apparatus 10 in a number of ways. In addition to standard AC line power, apparatus 10 can be operated using electrical energy from a power converter with a plug connection or can be provided by one or more batteries installed in the imager 10 housing.ExamplesI. Method of documenting a wound

[0056] An exemplary method of documenting a wound comprises the use of an imaging apparatus 10 as described above. The clinician must either select an existing wound record stored for a patient or create a new wound record. First the device 10 may analyze a wound dimension using a 3D camera, before capturing an RGB image of the wound using an RGB camera.

[0057] The device 10 illuminates the wound with an excitation wavelength of bacteria or bacterial products, such as by illumination with 405 nm monochromatic light from one or more LEDs. This causes any bacteria and / or bacterial metabolic products to fluoresce; for example, porphyrin metabolic products may fluoresce at a wavelength of 620-700 nm and pyoverdine products may fluoresce at a wavelength of 490-560 nm. This radiant illumination may be selectively activated at any point during image acquisition.

[0058] The hyperspectral camera 32c or multispectral camera detects fluorescence at the wavelengths mentioned above to visualize bacterial hotspots in the wound. This information may be visually presented to the user in a composite image which may also include the wound RGB image and / or a thermal image which indicates the presence of heat from infection. The user may use the displayed image to assess the wound as well as documenting and saving it. Saved images can be compared with future images of the same wound, to assess wound progression in terms of e.g. shrinkage and reduction in bacterial infection.II. Quantification of Bacterial Burden via Fluorescence Subtraction and AI-Based Signal Enhancement

[0059] This method provides an objective, quantitative estimation of bacterial burden by eliminating ambient light artifacts, isolating excitation-induced fluorescence of bacteria, and applying Al-based amplification and classification.

[0060] In this exemplary method a first image is captured, like in Example I, under 405 nm excitation, containing both bacterial fluorescence and ambient light.

[0061] Next a second image is captured immediately afterward with the 405 nm illumination turned off, containing only ambient light.

[0062] The reference image (excitation OFF) is subtracted from the excitation image (excitation ON), suppressing ambient illumination and isolating true bacterial fluorescence.

[0063] A neural network enhances the differential fluorescence signal and distinguishes different fluorophores, such as:• porphyrin-associated red fluorescence, and• pyoverdine-associated green fluorescence.

[0064] The enhanced fluorescence map is quantified pixel-by-pixel to compute a bacterial burden score, for example:• total fluorescence area,• mean intensity,• maximum local intensity, and• spatial distribution patterns.

[0065] A composite “Bacterial Burden Index” (BBI) is calculated and displayed to the clinician.III. Infection Detection by Fusion of Thermal, Fluorescence, and TextureFeatures

[0066] This exemplary method of the invention detects early or developing wound infections by combining data from three independent sensing modalities, namely thermal imaging, fluorescence imaging, and RGB-based texture analysis, into a single diagnostic model.

[0067] A thermal image of the wound and surrounding skin is acquired using the thermal module or camera to measure absolute temperature and temperature gradients.

[0068] A fluorescence image is captured under controlled 405 nm excitation light as previously described to reveal the presence of bacterial fluorophores such as porphyrins or pyoverdines.

[0069] A high-resolution RGB image is captured to extract texture features including color patterns, surface irregularities, tissue roughness, and the visual characteristics of the wound edge.

[0070] A segmentation algorithm isolates the wound bed and surrounding peri wound tissue.

[0071] A multimodal fusion engine combines:• local temperature elevations or asymmetries,• fluorescence signal intensity and distribution, and• texture-based indicators of inflammation or tissue breakdown.

[0072] These fused data are processed through a machine-learning classifier trained to recognize infection-specific patterns and the user or clinician is presented with indications relevant to diagnosis of a bacterial infection in the wound.Conclusion

[0073] The present invention has been described in detail with particular reference to a presently preferred embodiment, but it will be understood thatvariations and modifications can be effected within the spirit and scope of the invention. The presently disclosed embodiments are therefore considered in all respects to be illustrative and not restrictive. The scope of the invention is indicated by the appended claims, and all changes that come within the meaning and range of equivalents thereof are intended to be embraced therein.

[0074] When used in this specification and claims, the terms "comprises" and "comprising" and variations thereof mean that the specified features, steps or integers are included. The terms are not to be interpreted to exclude the presence of other features, steps or components.

[0075] The invention may also broadly consist in the parts, elements, steps, examples and / or features referred to or indicated in the specification individually or collectively in any and all combinations of two or more said parts, elements, steps, examples and / or features. In particular, one or more features in any of the embodiments described herein may be combined with one or more features from any other embodiments) described herein.

[0076] Protection may be sought for any features disclosed in any one or more published documents referenced herein in combination with the present disclosure.

[0077] Although certain example embodiments of the invention have been described, the scope of the appended claims is not intended to be limited solely to these embodiments. The claims are to be construed literally, purposively, and / or to encompass equivalents.

Claims

CLAIMS1. A hand-held imaging apparatus for characterization of infection at a wound site, the apparatus comprising: a radiant source configured to direct light having an excitation wavelength toward tissue at the wound site, one or more cameras configured to acquire spectral image content from a fluorescence response to the excitation wavelength of infected tissue at the wound site, an enclosure having a first surface side configured to house the radiant source and optics for the one or more cameras and having a display surface side, opposite the first surface side and configured to mount one or more displays that show the camera field of view prior to wound imaging and show processed spectral content from wound fluorescence following image capture; and a processor that is in signal communication with the one or more cameras and that is configured to analyze the spectral image content to identify one or more microbial infections.

2. The apparatus of claim 1 wherein the first surface side is a recessed surface side.

3. The apparatus of claim 1 or claim 2 wherein the radiant source generates UV light below a wavelength of about 450 nm, preferably below about 410 nm and preferably at a wavelength of about 405 nm.

4. The apparatus of any preceding claim wherein the one or more cameras include a multi-spectral camera, a hyperspectral camera, a thermal camera, a reflectance or RGB camera, and a plenoptic camera.

5. The apparatus of any preceding claim, wherein the radiant source comprises at least one, preferably an integer from 2 to 8 and preferably four LEDs.

6. The apparatus of claim 5, wherein the LED or LEDs emit monochromatic light with a wavelength of about 405 nm.

7. The apparatus of any preceding claim, wherein the radiant source is configured to be selectively activated during image acquisition, enabling the detection of bacterial fluorescence.

8. The apparatus of any preceding claim, further comprising a visible light source at the first surface side.

9. The apparatus of claim 8 wherein the visible light source comprises at least one LED and preferably comprises a LED ring.

10. The apparatus of any preceding claim wherein the one or more cameras comprises a 3D camera for wound dimension analysis.

11. The apparatus of claim 10 wherein the 3D camera and / or the processor is configured to generate a wound surface image by generating point clouds and constructing triangular or polygonal meshes of neighboring points.

12. The apparatus of any preceding claim wherein the one or more cameras comprises an RGB imaging module or camera configured to capture a color image of the wound surface.

13. The apparatus of claim 12 wherein the processor is configured to analyze the color image using artificial intelligence to perform tissue classification, optionally providing classification as granulation tissue, fibrin / slough and / or necrotic tissue.

14. The apparatus of any preceding claim wherein the one or more cameras comprises a thermal sensor configured to provide a temperature map of the wound and surrounding skin.

15. The apparatus of claim 14 wherein the temperature map highlights any temperature variation, to enable a clinician to identify signs of inflammation and / or infection risk.

16. The apparatus of any preceding claim wherein the one or more cameras is configured to detect fluorescence in the spectral range of approximately 620-700 nm or approximately 490-560 nm.

17. The apparatus of claim 16 wherein the one or more cameras employs at least one narrow-band filter to capture fluorescence in the spectral range of approximately 620-700 nm or approximately 490-560 nm while suppressing reflected excitation light and background signals from surrounding tissue.

18. The apparatus of any preceding claim wherein the radiant source comprises at least one LED and comprising: at least one white light source;at least one multispectral or hyperspectral camera configured to acquire spectral image content from a fluorescence response to the excitation wavelength; at least one RGB imaging module or camera; at least one 3D camera; and at least one thermal camera or thermal module.

19. A method of documenting a wound comprising: a) providing an apparatus according to any preceding claim, b) selecting a wound for documentation, and, in any order: c) analyzing a wound dimension using a 3D camera, d) capturing an RGB image of the wound using an RGB camera, e) providing light to the wound from a radiant source, the light having an excitation wavelength, f) acquiring a spectral image from a fluorescence response to the excitation wavelength; and g) presenting captured images to a user.

20. A method of documenting a wound comprising: a) providing an apparatus according to any of claims 1-18, b) selecting a wound for documentation, c) in ambient light and / or visible light providing light to the wound from a radiant source, the light having an excitation wavelength, d) acquiring an excitation image comprising a fluorescence response to the excitation wavelength and ambient and / or visible light; and e) either before or after steps c and d, acquiring a reference image without illumination from a radiant source, the second image only containing ambient and / or visible light, f) subtracting the reference image from the excitation image to provide an image representative only of fluorescence response.

21. The method of claim 20, further comprising enhancing the image representative only of fluorescence response, optionally using artificial intelligence, and distinguishing different fluorophores, such as porphyrin-associated red fluorescence, and / or pyoverdine-associated green fluorescence to produce an enhanced fluorescence map.

22. The method of claim 21 further comprising quantifying the enhanced fluorescence map pixel-by-pixel to compute a bacterial burden score, for example:• total fluorescence area,• mean intensity,• maximum local intensity,• spatial distribution patterns.

23. A method of documenting a wound comprising: a) providing an apparatus according to any of claims 1-18, b) selecting a wound for documentation, c) acquiring a thermal image of the wound and surrounding skin to measure absolute temperature and temperature gradients; d) acquiring a fluorescence image under radiant light with an excitation wavelength to reveal the presence of bacterial fluorophores such as porphyrins or pyoverdines, e) capturing a RGB image to extract texture features such as colour patterns, surface irregularities, tissue roughness, and the visual characteristics of the wound edge f) isolating the wound bed and surrounding peri wound tissue; g) combining with a multimodal fusion engine:• local temperature elevations or asymmetries,• fluorescence signal intensity and distribution,• texture-based indicators of inflammation or tissue breakdown; h). processing the resulting fused data through a machine-learning classifier trained to recognize infection-specific patterns.

24. The method of any one of claims 19-23, wherein the excitation wavelength is below a wavelength of about 450 nm, preferably below about 410 nm and preferably at a wavelength of about 405 nm.

Images