An imaging device for monitoring wounds and a system and method for managing a treatment of a wound

Abstract

The present invention relates to an imaging device for monitoring wounds such as chronic wounds and a system for method of managing the development of a treatment of a wound. In particular, the invention relates to an imaging device comprising an optic camera in combination with a lux meter and a distance sensor for estimating the distance between the optic camera and the wound. The invention further relates to a system and method, where the imaging device captures hyperspectral images and other sensor data of the wound which are transmitted to a computing unit for processing. The computing unit analyses the collected data and estimates wound parameters relating to the state of the wound, which are displayed a user of the system. The computing unit further a machine learning model to predict the development of the healing process.

Description

[0001] An imaging device for monitoring wounds and a system and method for managing a treatment of a wound

[0002] Technical field

[0003] The present invention relates to a medical device used for monitoring wounds such as chronic wounds. In particular, the invention relates to a device comprising an optic camera in combination with a distance sensor for estimating the distance between the optic camera and the wound.

[0004] The invention further relates to a system comprising a wound management system and the device mentioned above.

[0005] Background of the invention

[0006] Non-healing wounds, often referred to as chronic wounds, present a substantial challenge to healthcare systems. Chronic wounds account for 2-5% of the total healthcare spending in the US and EU. At any given time about 2% of the western population suffer from chronic wounds. To make matters worse, chronic wounds are strongly correlated with diabetes rates and the rise of the aging population. Globally, diabetic-related wounds alone result in amputations occurring every 30 seconds. Additionally, the EU currently sees a dramatic increase in chronic wound patients. This phenomenon has been dubbed "The silent epidemic" by doctors and experts because of its severity, already costing more than 5% of all healthcare spendings and because of a lack of doctors and nurses due to their long treatment time. Chronic leg and foot ulcers last on average 12 to 13 months with an average of 3.5 treatment sessions a week.

[0007] Chronic wounds are at a higher risk of infection due to their open nature and impaired healing. Infections are present in 45% of cases and lead to more severe health issues like prolonged treatment or at worst, amputation. In some cases, chronic wounds can become so severe and difficult to manage that amputation may be necessary to prevent further complications. Approximately 26% of poorly treated chronic wounds result in amputation, and five years post amputation, the mortality rate is at an astounding 50%.

[0008] This increase in chronic wounds is not projected to stop or even slack in the foreseeable future. Demographics skewing towards the older population, longer life expectancy and the incidence of type 2 diabetes are the main drivers and will significantly increase the cost of wound care in the EU.

[0009] Currently the only way to solve this rising problem is to increase workforce output. However, current chronic wound care practices rely on personal experience and individual assessments based on visual clues with a widespread in professionals' evaluations of the wounds. The study "How precise is the evaluation of chronic wounds by health care professionals?" by Stefan Stremitzer et al. investigates this matter where 16 wound therapists describe a diabetic ulcer. They granulation tissue, fibrin coverage, necrosis, size and depth of the lesion, wound exudate and wound edges were judged, and the therapeutical consequences were determined. The study data showed an extensive inhomogeneity and a widespread of the results, like in no other field of medical diagnostics.

[0010] Because of these incomplete wound assessments, it is rare that the most optimal treatment is chosen. Resulting in wasteful spending on ineffective treatments, which leads to weeks or even months of no progress, increased patient pain, and a higher risk of amputation. Lack of access to reliable wound assessment hinders correct clinical decision making on selection of patient treatment.

[0011] The current workflow in the healthcare system is as follows: When a wound is diagnosed by a healthcare professional as chronic, the patient is referred to either a hospital, wound clinic or home care depending on the severity of the wound. Then a treatment plan based primarily on visual assessment of the wound is made by a dedicated wound nurse or doctor. However, not every treatment works for every patient, due to incomplete wound assessment stemming from a lack of data. Often, multiple treatment modalities are tried, leading to weeks or even months of no progress, increased patient pain, and a higher risk of amputation.

[0012] The current workflow incurs wasteful spending on ineffective treatments, which is further hampered by the fact that the patient is sent back and forth between hospitals, wound clinics, and home care depending on how severe the wound is at the current time of treatment. This creates a challenge by involving multiple individuals / personnel across hospitals, wound clinics and home care in a single treatment, leading to potential delays and communication errors.

[0013] One way to solve the problem of a rise in chronic wounds is by: 1) Improving wound stewardship and 2) Reducing and improving wound documentation. One way to do this while keeping the same quality of treatment and to save time is by digitalising, standardizing and fully embracing support decision based wound treatment.

[0014] One way of predicting wound healing is disclosed in US 2023 / 0181042, which uses a machine learning system to predict wound healing parameters based on images of the wound.

[0015] US 2013 / 0162796 Al discloses a method of monitoring wounds using thermal and nonthermal image sensors. US 2020 / 0330028 Al discloses a method of identifying wound tissue using a deep neural network, which is trained to detect the type of wound tissue based on digital images of the wound.

[0016] WO 2018 / 173073 Al discloses a method and device for detecting and classifying pathogens in a tissue sample taken from the patient.

[0017] WO 2016 / 011534 Al discloses a method and device for evaluating and tracking bacterial load in a wound over time, where the wound is illuminated using a light source and an optical sensor is used to detect the fluorescence of the bacteria.

[0018] WO 2024 / 095284 Al discloses a device with a light source and imaging sensors for detecting problematic cellular entities in a wound using an analysis model.

[0019] Object of the invention

[0020] One object of the invention is to provide a device, system and method that overcomes the shortcomings of the abovementioned prior art or at least provide an alternative solution.

[0021] One object of the invention is to provide a device and method that enhances the flexibility and accessibility in the workflow for healthcare professionals.

[0022] One object of the invention is to provide a device and method that accurately reproduces images of the wound area.

[0023] One object of the invention is to provide a system and method that improves wound stewardship and reduces wound care costs.

[0024] One object of the invention is to provide a system and method that allows healthcare professionals to quickly identify whether the current treatment is effectively healing the wound or not.

[0025] Summary of the invention

[0026] The device, system and method according to the invention provides a way of improving the process for diagnosing and the process for the treatment of chronic wounds. Especially, the present device, system and method reduces the time and resources spend with setting up a treatment of a patient.

[0027] One object of the present invention is achieved by an imaging device, according to claim 1, for monitoring wounds comprising :

[0028] - at least one optic camera configured to capture images of a wound, - a distance sensor configured to measure the distance between the at least one optic camera and the wound, wherein the imaging device also comprises a lux meter, and optionally an IMU or a gyroscope, wherein the imaging device is configured to transmit a set of hyperspectral images and measured sensor data collected from the at least one optic camera, the distance sensor and the lux meter to a computing unit via a communications link, which is configured to estimate and present wound parameters, such as circumference, area and wound bed, to a user of the computer unit.

[0029] This provides a medical device that ensures the accurate reproducibility of images taken by the cameras. Thereby providing the healthcare professional with the best possible information for them to make their decision on. The present device is configured so that it can be operated by healthcare professionals such as wound doctors and nurses during wound treatment. Thereby allowing them to quickly assess the state of the wound and the progress of the healing process.

[0030] Thus, a first aspect of the present invention relates to an imaging or scanning device used to monitor wounds, such as chronic wounds. The imaging or scanning device for monitoring wounds comprising an optic camera, wherein the device also comprises means to estimate the distance between the optic camera and the wound such as a distance sensor and optionally a lux meter, and optionally an IMU i.e. a gyroscope, and the optic camera, the means for estimating the distance between the optic camera and the wound, and the optional lux meter are configured to transmit parameters to a computing unit to estimate parameters defining the state of the wound such as circumference, area, wound bed, etc..

[0031] The imaging device may comprise at least one colour camera, e.g. a RGB or 3D camera, configured to capture a coloured image, e.g. a RGB or 3D-image, of the wound and the surrounding tissue. The resolution, frame speed, focus point, shutter speed and other camera settings (e.g. number of spectral bands) may be selected based on the application and usage of the imaging device. For example, but not limited to, the colour camera may be a hyperspectral camera configured to capture a number of spectral images within a number of spectral bands, said number of spectral images and thus spectral bands being about 10, but could also be less than or greater than 10. This allows the health professionals to capture a detailed image of the wound area suitable for further processing in a local processor.

[0032] The imaging device may comprise a local processor and a memory configured to temporary store the captured image data. The processor may be configured to determine one or more wound parameters relating to the wound based on the captured image data. For example, a surface area of the wound, a circumference or diameter of the wound, a depth of the wound, colours of the wound, and other relevant wound parameters may be determined by the local processor. Alternatively, the image data may be transmitted to and processed in a separate computing unit as described later. The imaging device may thus be used to accurately measure one or more parameters relating to the wound area.

[0033] Optionally, the local processor may be configured to analyse the captured image data to identify different characteristics of the wound. The local processor may use predetermined colours to highlight the respective characteristics of the wound. The local processor may be configured to superimpose these predetermined colours onto the captured image data of the wound. For example, the colours may be used to highlight different types of tissue, such as epithelial tissue, granulating tissue, slough tissue, necrotic tissue or other relevant types of tissues. For example, but not limited to, epithelial tissue may be displayed as pink, granulating tissue may be displayed as red, slough tissue may be displayed as yellow, brown or gray, and / or necrotic tissue may be displayed as black. Healthcare professionals are thus able to easily track and document the visual progression of a wound over time. This also help healthcare professionals to assess the healing status, identify potential complications, and determine the effectiveness of treatment interventions.

[0034] According to one embodiment of the first aspect of the present invention, the imaging device comprises a fluorescent optical system configured to detect microorganisms optically, wherein the fluorescent optical system is configured to transmit parameters, such as a colour relating to specific microorganism, to the computing unit.

[0035] The imaging device may comprise a fluorescent optical system configured to excite bacteria in the wound so that they can be visually detected in the captured image data. Preferably, the fluorescent optical system may comprise one or more light sources configured to emit light at a predetermined wavelength and intensity onto the wound. The wavelength and intensity may be selected based on the fluorescent properties of a single or group of bacteria. This causes the bacteria to excite so they glow, e.g. scatters light, at a particular colour, e.g. red or cyan.

[0036] One or more optical filters of the fluorescent optical system may be arranged relative to the colour camera, alternatively a second colour camera. The optical filters may be used to detect the reflected light from the excited bacteria so that the light is identifiable in the image data. This allows the healthcare professionals to visually determine if bacteria loads are present in the wound or not. Detecting regions of these harmful bacteria provide the healthcare professionals with an indication of the state of the wound and it also indicates whether the treatment is effective or not. Preferably, the bacteria may be detected in real time or near-real time. Bacteria in the wound may cause infection, delay tissue repair, and trigger an inflammatory response. Additionally, they can damage healthy cells and hinder new tissue formation, leading to increased pain and delayed wound closure, leading to health risks and possibly amputation. Therefore, the fluorescent optical system may be adapted to detect harmful bacteria, such as Pseudomonas aeruginosa, Staphylococcus aureus, Klebsiella pneumoniae, Enterococcus faecalis, Acinetobacter baumannii or other harmful bacteria.

[0037] According to one embodiment of the first aspect of the present invention, the imaging device comprises a near-infrared camera configured to detect light reflected from the wound, where the reflected light is indicative of an oxygenation level in the tissue.

[0038] The imaging device may also comprise a near-infrared (NIR) optical system configured to detect haemoglobin oxygen saturation in and around the wound, preferably in real time or near-real time. The NIR optical system may comprise one or more light sources configured to emit near-infrared light at a predetermined wavelength and intensity onto the wound. The wavelength and intensity may be selected based on the properties of the haemoglobin in the blood, where the haemoglobin absorbs and scatters light differently depending on its oxygenation level. The near-infrared light penetrates deeper into the tissue than visible light, thus allowing the blood in the veins to reflect the emitted light.

[0039] One or more NIR-cameras may be arranged relative to the NIR-light source and configured to capture images of the reflected light. The local processor may be configured to analyse these images to determine the intensity of the oxygen saturation and optionally the blood flow. Alternatively, the captured image data may be transmitted to the computing unit, which may be configured to analyse the image data to determine the intensity of the oxygen saturation and optionally the blood flow. This enables non-invasive monitoring of the oxygenation level in the tissue in the wound and of the surrounding tissue.

[0040] High haemoglobin oxygen saturation is important to track because it is vital for cellular metabolism and tissue repair. Increased oxygen levels enhance angiogenesis, promoting blood vessel formation and delivering nutrients to the wound site. Oxygen also supports immune function, aiding in fighting off infections. Additionally, oxygen helps generate energy for collagen synthesis, promoting wound closure. This makes detecting the oxygen saturation in and around the wound a vital parameter for healthcare professionals. The present tissue oxygenation measurements may also help to predict wound formation thereby preventing future chronic wounds.

[0041] According to one embodiment of the first aspect of the present invention, the imaging device comprises a thermal camera configured to detect a temperature or temperature gradient either inside the wound or outside the wound or between the inside of the wound and surrounding tissue.

[0042] The imaging device may further comprise a thermal optical system configured to detect one or more temperature gradients of the wound and the surrounding tissue. One or more thermal cameras may be arranged at the front end of the imaging device to capture thermal image data of the wound and the surrounding tissue. The local processor may be configured to analyse these images to determine the temperature at one or more predetermined spots. The local processor may further calculate the temperature gradients between spots within the wound or between spots in the wound and spots in the surround tissue. Alternatively, the captured image data may be transmitted to the computing unit, which may be configured to analyse the image data to determine the temperatures at the spots and optionally the temperature gradients between these spots. This enables non-invasive monitoring of the temperature, which may indicate the state of the wound.

[0043] An inflamed tissue will heal slower and is detrimental to effective healing, therefore a higher temperature gradient may indicate a higher inflammation of the tissue. The present temperature measurements may help to determine whether the wound is healing, stagnating or expanding.

[0044] According to one embodiment of the first aspect of the present invention, the at least one optic camera is facing in a first direction and the distance sensor is configured to measure the distance to the wound in the same first direction.

[0045] The optical cameras, the distance sensor, the lux meter and optionally other sensor units may all be arranged on a front end of the imaging device so that they face in the same direction. The distance sensor may thus measure the distance between the wound and the front end, e.g. the optical camera. Alternatively, the local processor may calculate the distance to the wound based on the captured image data.

[0046] The lux meter may be arranged relative to the optical cameras and be configured to measure the amount of reflected light from the wound and the surrounding tissue. Alternatively, the optical camera, e.g. the colour camera, may be used as a lux meter.

[0047] Optionally, an inertial measurement unit (IMU) may further be arranged in the imaging device, e.g. on the front end. The IMU may comprise at least a gyroscope configured to measure the orientation, e.g. angular position, of the imaging device. This allows the health professional to adjust the position and orientation of the imaging device relative to the wound as well as the emitted light intensity from the imaging device to ensure an accurate reproducibility of the images. Thereby providing them with the best possible information to make a decision on.

[0048] According to one embodiment of the first aspect of the present invention, the imaging device is a handheld device.

[0049] The imaging or scanning device may be portable, handheld device allowing healthcare professionals to carry it with them wherever they go. Thus, the health care professionals may bring this handheld device to the patient's bedside or move around the clinic / hospital without any restrictions. This portability enhances the flexibility and accessibility of the workflow of the healthcare professionals.

[0050] With a handheld device, the healthcare professional may take images at the point of care, eliminating the need to transport patients to a dedicated wound care centres or a specific room with a large device. The present imaging device can be operated by wound doctors and nurses during wound treatments either in a home setting or in a clinical setting. This also helps save time and resources.

[0051] Further, a second aspect of the present invention relates to a system, according to claim 7, for managing the development of a treatment of a wound, comprising :

[0052] - at least one imaging device as described above, and

[0053] - a computing unit configured to receive the collected hyperspectral images and sensor data from the imaging device, to analyse the collected hyperspectral images and sensor data, e.g. using a wound prediction algorithm, and to determine an output in form of a text or a picture or graphic image or value defining the state of the wound.

[0054] This provides a system for quickly assessing wound healing in a standardized, data-driven way and prevent unnecessary treatments, thus saving treatment costs and improving care. The present imaging device is combined with a computing unit using data processing and predictive analytics to predict wound comes.

[0055] The imaging device comprises a communications module configured to communicate with a corresponding communications module in the computing unit. The hyperspectral images and the sensor data is transmitted to the computing unit for further data processing. The computing unit may be a server-based or cloud-based unit. The computing unit comprises a processor and a memory, where a wound management platform is stored in the memory and executed by the processor.

[0056] The data processing may be mainly performed by the processor in the computing unit, where a minimum of data processing may be performed in the local processor in the imaging unit. Alternatively, the data processing may be shared between the local processor and the processor in the computing unit. This allows the healthcare professionals to synchronise the hyperspectral images and other sensor data collected by the imaging device, which is stored in the computing unit for further processing.

[0057] The computing unit is configured to analyse the image data and sensor data to determine a set of wound parameters. These wound parameters are then presented to the user of the system on a display, allowing the healthcare professionals to pick the most effective solution to increase the healing rate. The advanced wound imagery collected by the imaging device is further presented to the user of the system.

[0058] According to one embodiment of the second aspect of the present invention, the system further comprises a patient record database, wherein the collected hyperspectral images and sensor data received from the imaging device is stored in the patient record database, and wherein patient information, preferably anonymised patient information, is further inputted and stored in the patient record database.

[0059] The present invention further uses a wound care-oriented electronic medical record (EMR) system for documenting patient progress, where patient information is inputted into the computing unit and stored in a patient record database. The patient information may be inputted by the patient's own doctor using a dedicated user interface. The patient information, including the hyperspectral images, wound predictions and other relevant information, is then available to all the healthcare professionals associated with the patient's treatment across hospitals, wound clinics and home care.

[0060] Preferably the patient information is anonymised by the computing unit and then stored in the patient record database before further data processing. For each new patient, the total data collection increases. This data collection may then be processed by the computing unit to determine their impact on the wound healing, as they make up the background of a treatment plan. This allows the present system to take into account the underlying health factors of the patients in addition to the wound parameters when determining a treatment plan for a particular patient.

[0061] The patient information may comprise parameters relating to comorbidities which is vital as these conditions can significantly impact wound healing. Comorbidities such as diabetes, vascular disease, and obesity compromise the body's ability to heal wounds effectively. By analysing the comorbidity data, the present system may be used to identify risk factors. This information helps in developing personalized treatment plans, predicting complications, and wound healing outcomes. Predicting chronic wound outcomes is crucial for setting expectations, identifying high-risk cases, allocating resources effectively, tailoring treatment, advancing wound care practices, and improving patient outcomes and quality of life.

[0062] According to one embodiment of the second aspect of the present invention, the system further comprises a treatment plan module comprising a series of registered treatments, wherein the wound prediction algorithm is a machine learning model implemented in a healing prediction module, the machine learning model being configured to predict a treatment outcome and / or a new treatment based on the stored image and sensor data and the stored patient information.

[0063] The healthcare professionals may manually select a treatment plan amongst a list of treatment plans stored in the computing unit. The advanced hyperspectral imagery and wound parameters may form basis for the development of a personalised treatment plan.

[0064] Alternatively, a wound prediction algorithm implemented in a wound healing prediction module of the computing unit may be used to automatically predict healing outcomes, complications and developments in treatment plans. The wound prediction algorithm may be a machine learning model, which may be trained using a set of training data to generate a high accuracy wound prediction algorithm. The training dataset may a public available dataset or new patient information added to the patient record database. The wound prediction algorithm may be trained using any known training techniques. This allows healthcare professionals to quickly identify whether the current treatment is effectively healing the wound, or if there is no detectable change in the wound parameters.

[0065] For example, but not limited to, the wound prediction algorithm may be trained using coded vectors that includes several sub-values for tracking and predicting the healing state of each case as described in US 2020 / 0330028 Al. However, other training datasets and training techniques may be used to train the wound prediction model.

[0066] Also, a third aspect of the present invention relates to a method, according to claim 10, of managing the development of a treatment of a wound, which method comprises the following steps:

[0067] 1) creating a patient profile in a patient record database, which patient profile comprises an identification of the patient and information relating to an initial assessment of the wound, and

[0068] - performing a first scan establishing an initial state of the wound and optionally of the surroundings of the wound with an imaging device as described above, wherein the collected hyperspectral images and sensor data from the imaging device are transmitted to and stored in the patient record database, - estimating wound parameters relating to the state of the wound by a computer unit as described above based on at least the collected hyperspectral images and sensor data, wherein the wound parameters are transmitted to and stored in the patient record database,

[0069] 2) after a time period t, performing a subsequent scan establishing a current state of the wound and optionally of the surroundings of the wound with the imaging device as described above, wherein the collected hyperspectral images and sensor data from the imaging device are transmitted to and further stored in the patient record database,

[0070] - estimating wound parameters relating to the state of the wound by a computer unit as described above based on at least the collected hyperspectral images and sensor data, wherein the wound parameters are transmitted to and further stored in the patient record database,

[0071] 3) performing a comparation, by the computing unit, of the wound parameters of the current scan with the wound parameters of the latest scan or with averaged values of the wound parameters from several earlier scans, and outputting, by the computing unit, delta values indicative of changes in the wound parameters from the latest scan or from the earlier scans, and wherein steps 2) and 3) are repeated until it is determined that the wound has healed or until it is determined that another medical action, e.g. amputation, is performed.

[0072] This provides a method that allows healthcare professionals to quickly assess the wound healing in a standardized, data-driven way to determine the effectiveness of a current treatment plan. Thereby preventing unnecessary treatments and thus saving treatment costs and improving care.

[0073] Initially, a patient profile is created and stored in the patient record database using a dedicated user interface of the wound management system. The patient profile may comprise the identification of the patient and information relating to an initial assessment of the wound. The initial assessment may comprise observations and / or medical information relating to the patient.

[0074] An initial scan is performed of the wound using the abovementioned imaging device, where the collected hyperspectral images and sensor data are added to the patient information stored in the computing unit. The wound parameters estimated by the local processor in the imaging device or by the processor in the computing unit may further be stored together with the patient information.

[0075] The patient information may be anonymised by an anonymization module before being added to the data collection used by the computing unit to predict healing outcomes, complications and developments in treatment plans. A subsequent scan is performed after a predetermined time period, t, using the abovementioned imaging device. The collected hyperspectral images and sensor data are added to the stored patent information to document the development of the wound.

[0076] The computing unit then analyses the stored data to determine the wound parameters relating to the wound. The computing unit may further compare the wound parameters of the current scan with the wound parameters of the last scan to determine delta values for each wound parameter. Alternatively, the wound parameters of the current scan may be compared with an averaged value of the wound parameters of several earlier scans.

[0077] The step of performing a subsequent scan and determining the delta values is repeated until it is determined that the wound has healed or until another medical action is required. This medical action may be an amputation of a limp. This allows healthcare professionals to quickly to assess the effectiveness of a treatment and make adjustments to the treatment plan, if needed.

[0078] According to one embodiment of the third aspect of the present invention, the step 3) further comprises generating a report, by the computing unit, indicative of the current state of the wound, preferably the report comprises at least one of i) a suggestion of advanced treatment by a doctor or nurse, ii) a warning relating to poor healing profile, iii) a suggestion relating to increased or reduced scanning frequency, and iv) a suggestion for treatment adjustments.

[0079] The computing unit may generate an output in form of a text or a picture or graphic image or value defining the state of the wound. This output may be a report indicating the current state of the wound.

[0080] Further, the computing unit may use the healing prediction model to predict the wound healing outcome and complications or risk factors. The computing unit may combine the output of this healing prediction model with other wound parameters to determine if adjustments in the current treatment plan are needed. The adjustment may be selected from a list of adjustments stored in the treatment plan module.

[0081] The healthcare professionals may then adjust the treatment plan accordingly to optimise the effectiveness of the healing process.

[0082] According to one embodiment of the third aspect of the present invention, one or more or a combination of the following wound parameters are estimated at each scan: area of the wound, circumference of the wound, colour of wound and wound edge, topography of wound bed, wound tissue type, moisture of wound, peri-wound temperature, wound bed temperature, hot spot temperature, types of microorganisms in the wound, area and size of areas of microbial activity in the wound, intensity of microbial activity in the wound, blood flow of the wound and surrounding area, blood oxygenation of surrounding wound area, and blood flow, e.g. local hot / cold spots, of the wound and surrounding area.

[0083] The imaging device and / or the computing unit may determine a set of predetermined wound parameters after each scan. Optionally, the set of wound parameters may be adapted to the type of wound during the setup of the patient profile. This allows the state of the wound to be determined in a standardised way.

[0084] For example, the colour camera may be used to determine wound parameters relating to the shape and dimensions of the wound, such as area of the wound, circumference of the wound, colour of wound and wound edge, topography of wound bed, wound tissue type, and moisture of wound. Other relevant wound parameters may also be determined.

[0085] For example, the fluorescent optical system may be used to determine wound parameters relating to specific microorganisms in the wound, such as types of microorganisms in the wound, area and size of areas of microbial activity in the wound, and intensity of microbial activity in the wound. Other relevant wound parameters may also be determined.

[0086] For example, the near-infrared optical system may be used to determine wound parameters relating to the oxygen and haemoglobin levels in the wound, such as blood flow of the wound and surrounding area, blood oxygenation of surrounding wound area, and blood flow, e.g. local hot / cold spots, of the wound and surrounding area. Other relevant wound parameters may also be determined.

[0087] For example, the thermal optical system may be used to determine wound parameters relating to the temperature in and around the wound, such as peri-wound temperature, wound bed temperature, and hot spot temperatures. Other relevant wound parameters may also be determined.

[0088] Preferably, the present device, system and method described above is used to treat chronic wounds, preferably of a human body.

[0089] Within this context, the term "chronic wound" should be understood as a wound that fails to progress through a normal, orderly, and timely sequence of repair, or in which the repair process fails to restore anatomic and functional integrity after three months. See for example the article: "Chronic wound pathogenesis and current treatment strategies: a unifying hypothesis" by Thomas A Mustoe, et aL, published in Plastic and Reconstruction Surgery, June 2006; 117(7 suppl) : 35S-41S. Brief description of the figures

[0090] Figure 1 shows a prior art treatment plan of two patients for the treatment of a wound. Figure 2 shows a handheld embodiment of an imaging device according to the invention. Figure 3 shows four different pictures made with four different cameras according to an embodiment of the invention.

[0091] Figure 4 shows an example of information registered in a patient record database according to the invention.

[0092] Figure 5 shows an example of an output established based on two or more scans with an imaging or scanning device according to the invention.

[0093] Figure 6 shows an example of a system according to the invention comprising the imaging device.

[0094] Figure 7 shows an example of a workflow process for managing the development of the treatment of a wound.

[0095] The invention will now be described in further detail in the following.

[0096] Detailed description of the invention

[0097] Figure 1 shows a prior art treatment plan of two patients for the treatment of a wound. The treatment plans are described in the article "Diffuse near-infrared spectroscopy prediction of healing in diabetic foot ulcers: a human study and cost analysis". The study indicates that waiting for weeks to visually confirm treatment effectiveness needlessly prolongs patient recovery. Thus, finding the right treatment is a long, but vital process for the overall healing of wounds, particular for chronic wounds.

[0098] As illustrated in figure 1, patient A 's first treatment is judged to be inefficient after 8 weeks and is moved to another treatment plan that works. However, for patient B this was not the case. Efficient treatment was not diagnosed in time and even though new treatments were tried the wound led to an amputation of the lower limb. This research data clearly shows that minimizing the time spent on ineffective treatments is crucial for the patient, the overall healing of the wound and the healthcare system, as materials, bed space, and nurse / doctor salaries could have been spared.

[0099] Figure 2 shows a handheld embodiment of an imaging or scanning device according to the invention. The imaging device comprises a front end adapted to be positioned relative to the wound of a body of a patient. The front end comprises a colour camera 1, e.g. a 3D camera, for capturing digital images of the wound and the surrounding tissue. A fluorescent optical system is further arranged in the front end, where a first light source 5 emits light at a predetermined wavelength and intensity onto the wound. One or optical filters are arranged relative to the colour camera 1 to detect the reflected light which are visually identifiable in the captured image data. A thermal camera 2 is also arranged at the front end, which is used to detect the temperature of the wound and surrounding tissue.

[0100] A near-infrared camera 4, e.g. a haemoglobin camera, is further arranged at the front end. The near-infrared camera 4 is used to detect the oxygen and haemoglobin levels in the blood. A second light source 6 is arranged relative to the near-infrared camera 4 and used to emit near-infrared light onto the wound.

[0101] A lux meter 7 is arranged at the front end to measure the amount of reflected light from the wound. An inertial measuring unit 8 is used to measure the angle and orientation of the imaging device.

[0102] Figure 3 shows four different pictures made with four different cameras according to an embodiment of the invention.

[0103] A digital image 9 is captured by the colour camera 1 from which a local processor may determine the size of the wound. A fluorescent image 10 is captured by the fluorescent optical system indicating the presence of a particular bacteria. A near-infrared picture 11 is captured by the near-infrared camera 4 indicative of the intensity of the oxygen saturation in the blood. A thermal image 12 is captured by the thermal camera 2 indicative of the temperature of the wound and the surrounding tissue.

[0104] These hyperspectral images 9, 10, 11, 12 along with other sensor data, e.g. the measured distance and light intensity, are transmitted to a computing unit shown in figure 6.

[0105] Figure 4 shows an example of information registered in a patient record database according to the invention. Information 13, 14 relating to the identification of a particular patient is stored in a patient profile in the patent record database. Further, information 15, 16 relating to the wound and treatment thereof are also stored in the patient profile.

[0106] The predicted healing of the wound may be illustrated in a graph 17 on the graphic user interface. This graph is outputted by a healing prediction module in the computing unit. The predicted healing is estimated based on the patient information and the wound parameters stored in the patient record database.

[0107] This information of the patient and of the wound are presented to the use of the present system in a dedicated graphic user interface.

[0108] Figure 5 shows an example of an output established based on two or more scans with an imaging or scanning device according to the invention. The hyperspectral images 18 are displayed to the user of the system via the graphic user interface. The user may switch between the different images captured by the different cameras.

[0109] A superimposed image 19 of the wound is further displayed in the graphic user interface, where relevant information is superimposed onto the original image of the wound.

[0110] Information 20 relating to the wound and notes 21 relating to the wound taken by the healthcare professional is also displayed in the graphic user interface.

[0111] The calculated delta values 22 indicating the change between the current scan and at least the last scan are displayed together with the other information.

[0112] Figure 6 shows an example of a system according to the invention comprising the imaging device shown in figure 2.

[0113] The imaging device 23 is a portable, handheld device comprising a local power source (not shown) for supplying power to the electrical components, such as the local processor and memory. The imaging device 23 comprises an arrangement of optical cameras 1, 2, 4 configured to capture a set of hyperspectral images of the wound and the surrounding tissue.

[0114] The colour camera 1 is used to capture digital images, e.g. 3D images. The thermal camera 2 is used to capture thermal images. The fluorescent optical system 25 is used to capture fluorescent images. The near-infrared optical system 26 is used to capture near-infrared images. The captures images are temporary stored in the memory. Optionally, the local processor is configured to determine one or more wound parameters based on the captured hyperspectral images. These wound parameters are further temporary stored in the memory.

[0115] The sensor data of the lux meter 7, the distance sensor 3 and, optionally, the IMU 8 are used to accurately position and orientate the imaging device 23 relative to the wound to achieve the best possible reproducibility of the hyperspectral images during each scan.

[0116] The imaging device 23 is in communication with the computing unit 24 via a communications link. The hyperspectral images, the sensor data and, optionally, the wound parameters are transmitted to the computing unit 24 for further data processing. A computer program of a wound management module 27 is implemented in the memory of the computing unit 24, where the computer program is executed by the processor of the computing unit 24. The wound management module 27 comprises a patient record database 28 in which patient profiles of the respective patients are stored. The collected data from the imaging device 23 is added to the respective patient profile during each scan, which allows the healthcare professionals to document and track the healing process of the wound over time.

[0117] The stored patient information is anonymised by the anonymisation module 29 before being analysed and processed in at least the healing prediction module 30. A healing prediction model in the form of a machine learning model is implemented in this module 30 and trained to predict at least the wound healing outcome. The present healing prediction model takes into account the underlying health factors of the patients in additions to the wound parameters, this enabling it to also be used to predict complications or risk factors as well as suggest adjustments to the current treatment plan. The wound healing trajectory is visualized and displayed to the user of the system in a dedicated graphic user interface.

[0118] The wound management platform 27 comprises a treatment plan module 33 comprising a list of different treatment plans and, optionally, possible adjustments to the treatment plans. The treatment plan module 33 may be updated an administrator of the wound management platform.

[0119] A wound parameter module 31 analyses the collected data from the imaging device 23 and determined one or more wound parameters 37 relating to the state of the wound. The determination of the wound parameters may be shared between the processor in the imaging device 23 and the processor in the computing unit 24 depending on the data processing capabilities. The wound parameter module 31 may further combined these wound parameters into new wound parameters that can have relevance for the user.

[0120] Further, the wound parameter module 31 compares the wound parameters of a current scan to at least the wound parameters of the last scan and generate a delta value for each wound parameter. The wound parameters and delta values thereof are visualized and displayed to the user of the system in the dedicated graphic user interface.

[0121] An imaging module 32 processes the hyperspectral images and, optionally, superimposed one or more relevant information onto the respective image(s). The hyperspectral images and relevant information are thus displayed to the user of the system in the dedicated graphic user interface.

[0122] A report module 34 generates a report or output indicative of the current state of the wound. The output or report may further include suggestions to adjustment to the current treatment plan. Figure 7 shows an example of a workflow process for managing the development of the treatment of a wound.

[0123] Initially, patient information as well as information relating to an initial assessment of the wound are inputted into the computing unit 24. Further, an initial scan of the wound is performed by the imaging unit 23 and the wound parameters and hyperspectral images are stored in the patient record database 28.

[0124] The healthcare professionals can monitor and track the development of the wound healing process by evaluating the wound parameters 37, 38 as well as the hyperspectral images and predicted healing trajectory. This allows them to quickly assess the current state of the wound and the effectiveness of the treatment plan.

[0125] The healthcare professionals can generate a report via the wound management platform 27. For example, the wound management platform 27 can indicate what wound dressing is the most effective for wounds with certain parameters. If the wound shows low blood flow, high moisture and low infection, a certain wound dressing would be recommended. The report may thus comprise suggestions to adjustments of the current treatment plan.

Claims

Claims1. An imaging device (23) for monitoring wounds comprising :- at least one optic camera (1, 2, 4) configured to capture images of a wound,- a distance sensor (3) configured to measure the distance between the optic camera (1) and the wound, characterized in that the imaging device also comprises a lux meter (7), and optionally an inertial measurement unit (8) or a gyroscope, wherein the imaging device (23) is configured to transmit a set of hyperspectral images and measured sensor data collected from the at least one optic camera (1, 2, 4), the distance sensor (3) and the lux meter (7) to a computing unit (24) via a communications link, which is configured to estimate and present wound parameters (37), such as circumference, area and wound bed, to a user of the computer unit (24).

2. An imaging device according to claim 1, characterised in that the imaging device comprises a fluorescent optical system (1, 5) configured to detect microorganisms optically, wherein the fluorescent optical system (1, 5) is configured to transmit parameters, such as a colour relating to specific microorganism, to the computing unit (24).

3. An imaging device according to any one of claims 1 to 2, characterised in that the imaging device (23) comprises a near-infrared camera (4) configured to detect light reflected from the wound, where the reflected light is indicative of an oxygenation level in the tissue.

4. An imaging device according to any one of claims 1 to 3, characterised in that the imaging device (23) comprises a thermal camera (2) configured to detect a temperature or temperature gradient either inside the wound or outside the wound or between the inside of the wound and surrounding tissue.

5. An imaging device according to any one of claims 1 to 4, characterised in that the at least one optic camera (1, 2, 4) is facing in a first direction and the distance sensor (3) is configured to measure the distance to the wound in the same first direction.

6. An imaging device according to any one of claims 1 to 5, characterised in that the imaging device (23) is a handheld device.

7. A system for managing the development of a treatment of a wound, comprising :- at least one imaging device (23) according to any one of claims 1 to 6, and- a computing unit (24) configured to receive the collected hyperspectral images and sensor data from the imaging device (23), to analyse the collected hyperspectral images and sensordata, e.g. using a wound prediction algorithm, and to determine an output in form of a text or a picture or graphic image or value defining the state of the wound.

8. A system according to claim 7, characterised in that the system further comprises a patient record database (28), wherein the collected hyperspectral images and sensor data received from the imaging device (23) is stored in the patient record database (28), and wherein patient information, preferably anonymised patient information, is further inputted and stored in the patient record database (28).

9. A system according to claim 8, characterised in that the system further comprises a treatment plan module (33) comprising a series of registered treatments, wherein the wound prediction algorithm is a machine learning model implemented in a healing prediction module (30), the machine learning model being configured to predict a treatment outcome and / or a new treatment based on the stored image and sensor data and the stored patient information.

10. A method for managing the development of a treatment of a wound, which method comprises the following steps:1) creating a patient profile in a patient record database (28), which patient profile comprises an identification of the patient and information relating to an initial assessment (35) of the wound, and- performing a first scan establishing an initial state of the wound and optionally of the surroundings of the wound with an imaging device (23) according to one of the claims 1- 6, wherein the collected hyperspectral images and sensor data from the imaging device(23) is transmitted to and stored in the patient record database (28),- estimating wound parameters (37) relating to the state of the wound by a computer unit(24) according to any one of claims 7 to 9 based on at least the collected hyperspectral images and sensor data, wherein the wound parameters (37) are transmitted to and stored in the patient record database (28),2) after a time period, t, performing a subsequent scan establishing a current state of the wound and optionally of the surroundings of the wound with the imaging device (23) according to one of the claims 1-6, wherein the collected hyperspectral images and sensor data from the imaging device (23) is transmitted to and further stored in the patient record database (28),- estimating wound parameters (37) relating to the state of the wound by a computer unit (24) according to any one of claims 7 to 9 based on at least the collected hyperspectral images and sensor data, wherein the wound parameters (37) are transmitted to and further stored in the patient record database (28),3) performing a comparation, by the computing unit (24), of the wound parameters (37) of the current scan with the wound parameters (37) of the latest scan or with averagedvalues of the wound parameters (37) from several earlier scans, and outputting, by the computing unit (24), delta values indicative of changes in the wound parameters (37) from the latest scan or from the earlier scans, wherein steps 2) and 3) are repeated until it is determined that the wound has healed or until it is determined that another medical action, e.g. amputation, is performed.

11. A method according to claim 10, characterised in that the step 3) further comprises generating a report, by the computing unit (24), indicative of the current state of the wound, preferably the report comprises at least one of i) a suggestion of advanced treatment by a doctor or nurse, ii) a warning relating to poor healing profile, iii) a suggestion relating to increased or reduced scanning frequency, and iv) a suggestion for treatment adjustments.

12. A method according to claims 10 to 11, characterised in that one or more or a combination of the following wound parameters are estimated at each scan: area of the wound, circumference of the wound, colour of wound and wound edge, topography of wound bed, wound tissue type, moisture of wound, peri-wound temperature, wound bed temperature, hot spot temperature, types of microorganisms in the wound, area and size of areas of microbial activity in the wound, intensity of microbial activity in the wound, blood flow of the wound and surrounding area, blood oxygenation of surrounding wound area, and blood flow, e.g. local hot / cold spots, of the wound and surrounding area.

Images