Optical system for a body treatment device

Abstract

A sensing system for use with a body care or body treatment device, such as a personal care device. The sensing system may be integrated in a device. The sensing system, during use, obtains optical sensing data proximate a tissue region of the body care device, processes the sensing data to extract a representation of a superficial vein or vascular pattern at the tissue region, and using the representation of the vein pattern to infer information about a user category (e.g., a demographic category) or a body location to which the sensed tissue region belongs. Optionally, processing-related information may be generated using the information. This may be an operating parameter setting, or may be monitoring information such as overlay information.

Description

TECHNICAL FIELD

[0001] The present invention relates to the field of body treatment devices, and in particular to handheld body treatment devices. BACKGROUND

[0002] The present invention relates to body care devices, in particular those employing a handheld unit having an operational treatment area that is passed over a surface of a user's body to perform a body treatment function. Examples of such devices include grooming devices, such as hair removal devices, shavers and beard trimmers. One class of hair removal devices is Intense Pulsed Light (IPL) hair removal devices. Additionally, there are other devices designed for skin treatment, such as optical skin treatment devices. SUMMARY

[0003] The inventors have recognised that in the case of any device that is mobile, e.g. hand operated, and thus can be applied to a variety of different specific body areas and / or used by a variety of different users, it is valuable to allow the function to be varied according to the person or body location at which the device is being applied. The standard way of achieving this is to provide a user control interface that allows the user to manually adjust the settings of the device. However, the inventors have recognised that it would be advantageous to be able to automatically configure the settings of the device. The inventors have recognised that in order to facilitate this, means for automatically identifying or classifying the user or body part at which the device is being applied would be valuable. Indeed, this functionality would have value beyond setting configuration, e.g. for performance or coverage monitoring.

[0004] The inventors have also recognised that it would be advantageous to be able to automatically track the coverage of the device over the relevant body area, thereby being able to provide the user with insights into potential missed areas and overexposed areas.

[0005] The invention is defined by the claims.

[0006] According to examples in accordance with an aspect of the present invention, there is provided an optical sensing system for a body treatment device. The system comprises a vein visualisation subsystem for integration in a body treatment device, for detecting patterns of superficial veins of a user. The vein visualisation subsystem comprises an optical arrangement comprising: a light source for directing light into surface tissue of a user during use of the body treatment device, and an optical detector for detecting a pattern of light reflected from the surface tissue, the surface tissue having a vein pattern represented therein. The system further comprises a classification module for applying to the output from the vein visualisation subsystem to provide one or more classifications of the biological structure or region to which the vein pattern represented in the detected light pattern belongs. The system further comprises a processing module adapted to generate treatment-related information for at least one detected vein pattern using the output from the classification module.

[0007] The detected light pattern may, for example, be an image.

[0008] The inventors have recognized that a solution to the above problems can be achieved by optically scanning the skin while using the device and exploiting predictable variations in the pattern of superficial vein structures at different areas of the body and for different demographic categories of users.

[0009] As will be discussed further below, it is known from the literature that the superficial vein pattern is unique for each part of the body, e.g. leg, arm, foot, hand, etc. It is also known from the literature that the vein pattern varies with the age and gender of the individual. The inventors therefore propose to use this predictable variation in the superficial vein pattern as a function of the body part and the demographic to identify the demographic category of the person or body part to which the device is applied. In other words, the wider biological structure or region thereof that is scanned by the optical system is classified. This classification can subsequently be used to determine processing related information. This may, for example, include appropriate operational settings for the device, or it can include monitoring information, e.g. body coverage tracking information. Thus, it is proposed to use the vein pattern to automatically infer location context information about to whom or to which region the device is being applied.

[0010] More specifically, in some embodiments, the one or more classifications generated by the classification module can include any one or more of: an identification of the body part to which the vein pattern belongs; a classification of one or more demographic characteristics of the individual to which the vein pattern belongs, e.g. age and / or gender; and / or an identification of the unique patch of skin in which the vein pattern is located.

[0011] In some embodiments, the vein visualization subsystem can further comprise a pattern extraction module for extracting a representation of the vein pattern from the detected light pattern. The classification module can be adapted to operate on the extracted representation of the vein pattern.

[0012] In some embodiments, the representation of the vein pattern comprises a vein pattern mask.

[0013] In some embodiments, the pattern extraction module can be a trained neural network trained to extract a vein pattern mask from the detected light pattern. The neural network can be trained on training data comprising images depicting different vein pattern masks, each vein pattern mask annotated with a substantially true vein pattern mask.

[0014] In some embodiments, the classification module can comprise a trained neural network. As one example, this can be a convolutional neural network, e.g. having a U-net architecture. In some embodiments, the neural network employed by the classification module can be trained to receive a vein pattern mask as input to generate at least one classification of the relevant one or more classifications as output.

[0015] For example, the trained neural network employed by the classification module can be a network trained using training data comprising vein pattern masks annotated with ground truth information indicating the biological structure or region to which the vein pattern represented in the vein pattern mask belongs. For example, the ground truth information can indicate one or more of: the body part to which the vein pattern belongs; the age of the individual to which the vein pattern belongs; the gender of the individual to which the vein pattern belongs, and a unique skin patch identifier of the skin patch to which the vein pattern belongs.

[0016] In some embodiments, the classification module is for integration within a body treatment device.

[0017] In some embodiments, the classification module comprises a memory storing the neural network and a processor for controlling the application of the neural network.

[0018] In some embodiments, the neural network can be quantized prior to integration into a device to reduce the size of the model, enabling it to be deployed by hardware with small memory and processing capabilities.

[0019] In some embodiments, the processing-related information comprises electronic control instructions for adjusting an operating parameter of the body treatment device in dependence on the classification.

[0020] In some embodiments, the operating parameter comprises a selection between available treatment modes, wherein a first mode is a shaving mode and a second mode is a hair removal mode.

[0021] In some embodiments, the body treatment device is a shaving device, and wherein the operating parameter comprises a cutting speed / rate of a cutting mechanism of the shaving device. For example, a higher cutting rate can be set when the device is applied to a face than when the device is applied to a chest.

[0022] In some embodiments, the body treatment device is an Intense Pulsed Light (IPL) hair removal device, and wherein the operating parameter comprises an intensity level of the pulsed light generated by the device.

[0023] As mentioned above, in addition to or instead of using the classification to configure the operating settings of the device, the classification can be used to track the coverage of the device over the body during a treatment session.

[0024] In this regard, in some embodiments, the classifier module can be adapted to generate a classification comprising an identification of the unique patch of skin on which the vein pattern is located. In some embodiments, the system further comprises a coverage tracking module comprising one or more processors adapted to receive, during a body treatment session, an indicator of each detected unique patch of skin; and store, in a memory module, a record of each patch of skin detection to compile a body treatment session history. In this case, the previously mentioned treatment-related information can comprise the body treatment session history.

[0025] In some embodiments, the processing module is adapted to identify, from the treatment session history, any unique patches of skin for which multiple detections are recorded in the session history. In this case, the treatment-related information can comprise an indication of each unique patch of skin for which multiple detections are recorded. This allows identification of areas of skin that have been covered more than once during a given treatment session.

[0026] In some embodiments, the processing module can be constituted by an analysis module. The analysis module can be adapted to retrieve, from the memory module, a body treatment session history for at least one body treatment session. The analysis module can also be adapted to access a data structure recorded in a data store that records a set of known or assumed unique patches of skin comprised by the user. In this case, the treatment-related information can comprise coverage feedback indicative of coverage and / or non-coverage of different unique patches of skin during a treatment session.

[0027] In some embodiments, the data structure recorded in the data store can comprise a body map comprising an indicator of a body location of each patch in the set of unique patches of skin.

[0028] In some embodiments, the coverage feedback can comprise a graphical representation of the body map comprising a visual indication of coverage and / or non-coverage of each unique patch of skin. Optionally, the processing module can be adapted to communicate the coverage feedback to a user interface or mobile computing device.

[0029] Consistent with at least any of the above embodiments, the system can further comprise one or more of the following features.

[0030] In some embodiments, the light source can be an infrared (IR) light source.

[0031] In some embodiments, the light source comprises an array of illumination elements, such as an elongate or linear array of illumination elements.

[0032] In some embodiments, the optical detector comprises an array of light detection elements, such as an elongate or linear array of light detection elements.

[0033] In some embodiments, the light source comprises an elongate or linear array of illumination elements, and the optical detector comprises an elongate or linear array of light detection elements parallel to the elongate or linear array of illumination elements.

[0034] This provides a structurally efficient device in which illumination and detection can be performed simultaneously by moving the optical device in a direction perpendicular to the linear array.

[0035] In some embodiments, the vein visualisation subsystem further comprises a proximity sensor for detecting a proximity distance between the optical detector and the surface tissue of the user. In some embodiments, the vein visualisation subsystem further comprises a patch normalisation module for applying to the detected light patterns a positive or negative magnification of a size determined in dependence on the sensed proximity distance, so as to normalise each detected light pattern size to a standardised size.

[0036] Another aspect of the present application is a body treatment device comprising an optical sensing system according to any example or embodiment described in the present disclosure.

[0037] In some embodiments, the body treatment device is an Intense Pulsed Light (IPL) hair removal device, or the body treatment device is a shaver or other mechanical hair removal device.

[0038] In some embodiments, the vein visualisation subsystem comprises an optical input / output region arranged for application against the user's skin during use of the device, and wherein the light source and optical detector emit and detect light via the optical input / output region. This can comprise an optical window, for example. The region can form a tissue contact region.

[0039] In some embodiments, the body treatment device comprises an operational region at a surface of the device for engagement with tissue to perform a body treatment function, and wherein the optical sensing system comprises a plurality of optical devices comprising at least one optical device disposed on one side of the operational region and at least a second optical device disposed on an opposite side of the operational region. This allows directional information to be obtained.

[0040] Another aspect of the present application provides an optical sensing method for use during application of a body treatment device to a user's body, comprising: detecting a vein pattern of superficial veins of the user, wherein the detecting comprises: directing light into surface tissue of the user during use of the body treatment device; and detecting a pattern of light reflected from the surface tissue in response to the directing of light. The method further comprises applying a classification module to provide one or more classifications of biological structures or regions thereof to which the vein pattern represented in the detected light pattern belongs. The method further comprises generating treatment-related information for at least one detected vein pattern using output from the classification module.

[0041] In some embodiments, the classification module comprises a trained neural network, and wherein the method comprises a training process for training the neural network, the training process comprising training the network using training data, the training data comprising vein pattern masks annotated with ground truth information indicating a biological structure or region to which a vein pattern represented in the vein pattern mask belongs. For example, the ground truth information can indicate one or more of: a body part to which the vein pattern belongs; an age of an individual to which the vein pattern belongs; a gender of an individual to which the vein pattern belongs; and a unique skin patch identifier of a skin patch to which the vein pattern belongs.

[0042] In some embodiments, the method further comprises quantizing the neural network after training. The neural network can be quantized prior to integration in a device to reduce the size of the model, enabling it to be deployed by hardware with small memory and processing capabilities.

[0043] Another aspect of the invention is a computer program product comprising machine executable instructions that, when executed by a processor, are configured to cause the processor to: control vein pattern detection of superficial veins of a user by:

[0044] communicating with a light source to control the light source to direct light into surface tissue of the user during use of a body treatment device; and

[0045] communicating with an optical detector to control the optical detector to detect a pattern of light reflected from the surface tissue in response to the directing of light;

[0046] applying a classification module to provide one or more classifications of a biological structure or region thereof to which a vein pattern represented in the detected pattern of light belongs; and

[0047] generating treatment related information for at least one detected vein pattern using output from the classification module.

[0048] Another aspect of the invention is a method of training an artificial neural network, comprising: receiving training data, wherein the training data comprises a plurality of training data entries, each training data entry comprising a vein pattern mask annotated with ground truth information indicating a biological structure or region to which a vein pattern represented in the vein pattern mask belongs; and training the artificial neural network using the training data.

[0049] In some embodiments, the ground truth information indicates one or more of: a body part to which the vein pattern belongs, an age of an individual to which the vein pattern belongs, a gender of an individual to which the vein pattern belongs and a unique skin patch identifier of a skin patch to which the vein pattern belongs.

[0050] These and other aspects of the application will become apparent upon reference to the following examples. BRIEF DESCRIPTION OF DRAWINGS

[0051] For a better understanding of the present application, and to show how it can be put into effect, reference will now be made, purely by way of example, to the accompanying drawings in which:

[0052] Figure 1 Example vein patterns detectable on arms and feet are schematically shown;

[0053] Figure 2 Example images of vein patterns taken on forearms of each of a plurality of different subjects in different age / gender categories are shown;

[0054] Figure 3 Steps of an example method according to one or more embodiments of the application are outlined;

[0055] Figure 4 is a block diagram of an example system according to one or more embodiments of the application;

[0056] Figure 5 is a block diagram of an example system according to one or more embodiments of the application, in which the classifier module acts directly on the detected light pattern or imaged light pattern;

[0057] Figure 6 is a block diagram of an example system according to one or more embodiments of the application, in which the classifier module acts on a vein pattern mask extracted from the detected light pattern or image by a pattern extraction module;

[0058] Figure 7 and Figure 8 An example body treatment apparatus in which an optical sensing system according to one or more embodiments of the application is integrated is schematically shown;

[0059] Figure 9 The structure of an optical component of an example optical sensing system according to one or more embodiments of the application is schematically shown;

[0060] Figure 10 An example vein pattern mask derived from measured light pattern data is schematically shown;

[0061] Figure 11 At least part of a processing flow according to one or more embodiments is schematically outlined;

[0062] Figure 12 A processing flow for configuring operational parameter settings based on classification according to one or more embodiments is schematically outlined;

[0063] Figure 13 A processing flow according to a set of particular embodiments is schematically outlined, wherein the operating mode of a body care device is varied according to the classification;

[0064] Figure 14 A processing flow according to a set of particular embodiments is schematically outlined, wherein the motor speed of a hair cutting device (e.g. a shaver) is adjusted according to the classification;

[0065] Figure 15 A processing flow according to a set of particular embodiments is schematically outlined, wherein the intensity level of a treatment light of an optical hair removal device is adjusted according to the classification; and

[0066] Figure 16 is a block diagram of another example system according to one or more embodiments of the present application, including an overlay tracking module. DETAILED DESCRIPTION

[0067] The present application will be described with reference to the accompanying drawings.

[0068] It should be understood that the detailed description and specific examples, while indicating exemplary embodiments of apparatuses, systems and methods, are intended for purposes of illustration only and are not intended to limit the scope of the present application. These and other features, aspects, and advantages of the apparatuses, systems and methods of the present application will become better understood from the following description, appended claims, and accompanying drawings. It should be understood that the drawings are only schematic and are not drawn to scale. It should also be understood that the same reference numerals are used throughout the drawings, wherein same or similar drawings are indicated with the same or similar reference numerals.

[0069] The present application provides a sensing system for use with a body care or body treatment device (e.g. a personal care device). The sensing system can be integrated in the device. The sensing system acquires optical sensing data of a tissue region in proximity of the body care device during use, processes the sensing data to extract a representation of a superficial vein or vessel pattern at the tissue region, and uses the representation of the vein pattern to infer information about a user class (e.g. a demographic class) or a body location to which the sensed tissue region belongs. Optionally, a treatment related information can be generated using this information. This can be an operating parameter setting, or it can be a monitoring information such as coverage information.

[0070] Therefore, it is proposed to use a vein visualisation technique to analyze optical sensor data, e.g. images, of a user’s superficial veins.

[0071] According to at least one set of embodiments, it is proposed to use an image classification model (e.g. an AI-based image classification model) to determine which region of the body a vein pattern belongs to and / or a demographic class of the person the vein pattern belongs to.

[0072] A brief background of superficial vein patterns will now be shown.

[0073] Superficial veins are veins that are close to the surface of the skin. This is in contrast to deeper veins that are further from the surface. While the deeper veins are usually each paired with a corresponding artery of the same name, the superficial veins are not.

[0074] Figure 1 The pattern of superficial veins of the forearm and foot are shown schematically. For example, with respect to the forearm, the basilic vein 2, median cubital vein 4 and median cephalic vein 6 are labelled.

[0075] It is known from the literature that the pattern of superficial veins is unique for each part of the body, for example leg, arm, foot, hand, etc. It is also known from the literature that the vein pattern varies with the age and gender of the individual.

[0076] Figure 2 A series of images showing the pattern of superficial veins on the forearm of subjects of different demographic groups is shown. Image al shows a 19 year old male. Image a2 shows a 16 year old female. Image a3 shows a 47 year old male. Image a4 shows a 72 year old male. Image bl shows a 19 year old male. Image b2 shows a 22 year old female. Image b3 shows a 50 year old female. Image b4 shows a 57 year old male. It will be appreciated that the vein pattern differs between different demographic groups. This variation is regular and predictable to some extent, and the inventors have recognised that this can be used to derive information about the individual or the region being treated.

[0077] The inventors therefore propose to use the predictable variation of the pattern of superficial veins as a function of the body part and demographic to identify the demographic class of the person or body part to which the device is applied. Currently, any hand held device in personal health (e.g. shavers, hair removal devices, blood pressure sensors) and healthcare devices (e.g. CT imaging, X-ray imaging) lack the ability to automatically detect the body to which the device is being applied.

[0078] Figure 3 The steps of an example method according to one or more embodiments are outlined in the form of a block diagram. These steps will be described in general terms before being further explained in the form of example embodiments.

[0079] An optical sensing method 10 for use during application of a body treatment device to a user's body is provided.

[0080] The method comprises detecting 12 a venous pattern of a user's superficial vein. The detecting 12 comprises directing 14 light into surface tissue of the user during use of a body treatment device, and detecting 16 a pattern of light reflected from the surface tissue in response to the directing of light. The method further comprises applying 18 a classification module to provide one or more classifications of biological structures or regions thereof to which the venous pattern represented in the detected pattern of light belongs. The method further comprises generating 20 treatment-related information for at least one of the detected venous patterns using output from the classification module.

[0081] The method can be implemented in the form of a software product, for example a computer program product.

[0082] For example, an aspect of the invention is a computer program product comprising machine executable instructions which, when executed by a processor, are configured to cause the processor to:

[0083] The method can be implemented in the form of a software product, for example a computer program product.

[0084] The method can be implemented in the form of a software product, for example a computer program product.

[0085] The method can be implemented in the form of a software product, for example a computer program product.

[0086] As mentioned above, the method can also be implemented in hardware, for example in the form of an optical sensing system configured to perform an optical sensing method such as described above.

[0087] To aid understanding, Figure 4 A schematic diagram of an example optical sensing system 30 according to one or more embodiments of the invention is shown.

[0088] The system 30 comprises a vein visualisation subsystem 40. This can be integrated in a body treatment device. The vein visualisation subsystem 40 is configured to detect a pattern of a user's superficial vein.

[0089] The vein visualisation subsystem 40 comprises an optical arrangement 50 for acquiring optical (e.g. image) data of a tissue patch. The optical arrangement comprises a light source 52 for directing light into surface tissue 32 of the user. The optical arrangement further comprises an optical detector 54 for detecting a pattern of light reflected from the surface tissue having a venous pattern represented therein.

[0090] The system also includes a classification module 34 for applying to the output from the vein visualisation subsystem 40 to provide one or more classifications of the biological structure or region to which the vein pattern represented in the detected light pattern belongs.

[0091] The system also includes a processing module 36 adapted to generate processing-related information for at least one detected vein pattern using the output from the classification module.

[0092] In some embodiments, the classification module 34 can be adapted to receive output directly from the optical detector 54. For example, the classification module can receive the detected pattern of light from the optical detector and can use the pattern of light to generate a classification. Figure 5 One example of such an arrangement is shown in

[0093] In some embodiments, the vein visualisation 40 can include a pattern extraction module 44 for extracting a representation of a vein pattern from the detected pattern of light, and wherein the classification module 34 is adapted to operate on the extracted representation of the vein pattern. In Figure 6 One example of such an arrangement is shown in

[0094] In any of the above examples, a controller (not shown) can be provided for controlling the operation of the vein visualisation subsystem, for example for controlling the operation of the light source and optical detector to detect the light pattern.

[0095] In any of the above examples, while the illustrated components are depicted in a particular architecture and spatial arrangement, this can be understood to be merely illustrative. The various components can in fact be provided in the form of a distributed system. In some embodiments, the different components can be spatially grouped in a different manner to that depicted in the figures. For example, in some embodiments each of the classification module 34, the processing module 36 and the optional pattern extraction module 44 can be integrated in a single controller module. Furthermore, while the various components of the system are shown as being provided as separate units, in fact the functional operations of some or all of the components can be performed by a single processing device in some embodiments. The various modules can be software modules or hardware modules.

[0096] In some embodiments, the optical detector 54 takes the form of a camera and the detected light pattern can be an image generated by the camera.

[0097] In some embodiments, the optical sensing system 30 can be integrated in a body treatment device. In Figure 7 and Figure 8Two examples are schematically illustrated in Figs. 2 and 3. Each figure illustrates a body treatment device 60 comprising components of an optical sensing system integrated therein. In both examples, the body treatment device comprises an operational treatment portion 64 adapted to perform a body treatment function. The operational treatment portion 64 defines an operational area 66 at a surface of the device for engagement with tissue to perform the body treatment function.

[0098] With reference to Figure 7 Adjacent to the operational area 66 of the body treatment portion 64 is an optical arrangement 50 of the optical sensing system 30 for acquiring a light pattern containing a representation of a vein pattern of the tissue area. An optical input / output area is provided at the surface of the device 60, which is arranged to be applied against the skin of a user during use of the device 60, and wherein the light source and the optical detector of the optical arrangement 50 emit and detect light via the optical input / output area.

[0099] The operational treatment portion 64 is operatively coupled to a treatment controller 72 adapted to control operation of the operational treatment portion. As non-limiting examples, the operational treatment portion can be adapted to perform a hair removal function, a hair cutting function or an optical skin treatment function. The optical arrangement 50 is coupled to a sensing controller 74 containing a classification module 34, a processing module 36 and an optional pattern extraction module 44. Different structural arrangements of the various components are of course possible, and the figures merely provide illustrative examples.

[0100] Figure 8 Differently from Figure 7 In difference to the embodiment of Fig. 2, a plurality of optical arrangements 50 is provided, including at least one optical arrangement 50a disposed on one side of the operational area 66 and at least a second optical arrangement 50b disposed on an opposite side of the operational area 66.

[0101] According to any of the embodiments, the body treatment device can in some embodiments be an Intense Pulsed Light (IPL) hair removal device. According to any of the embodiments, the body treatment device can in some embodiments be a shaver or other mechanical hair removal device.

[0102] As an alternative to integration in a body treatment device, the optical sensing system 30 can also be integrated in a separate unit or device, such as an accessory adapted to be removably coupled to a body of a body treatment device, or a wearable device, such as a wrist or hand or finger worn device.

[0103] Embodiments of the invention include functionality for acquiring a pattern of light reflected from surface tissue. Implementation options with respect to this functionality will now be discussed in more detail.

[0104] As mentioned above, the optical arrangement 50 comprising the light source 52 and the optical detector 54 can be employed.

[0105] With respect to the light source 52, it can comprise one or more illumination elements, each adapted to produce a light output for incidence onto a tissue surface of a user, e.g. during use of the body treatment device. In some embodiments, the illumination elements can be adapted to produce a light output in the near-infrared (NIR) or infrared (IR) frequency range. In some embodiments, each illumination element can be a LED illumination element. The light source can comprise a set of LEDs.

[0106] For example, an infrared emitting diode (IRED) is a solid-state light source that produces a light output in the near-infrared portion of the light spectrum. IREDs allow for cheap, efficient production of near-infrared light with only a small spatial footprint.

[0107] When light penetrates the body tissue, it is reflected by the haemoglobin in the blood, providing a reflected light pattern that mirrors the shape and structure of the superficial blood vessel pattern of the subject. Near-infrared (NIR) and infrared (IR) light is particularly well reflected by haemoglobin.

[0108] To capture the reflected light pattern, an optical detector 54 is provided. In simple cases, this can comprise a 2D photodetector array or the like. In some further examples, the optical detector can be provided in the form of a camera, such that the detected light pattern takes the form of an image acquired by the camera.

[0109] In some embodiments, the optical detector takes the form of a charge-coupled device (CCD). A CCD is a solid-state electronic device that is capable of converting light into an electrical signal. A CCD provides a form of image sensor. It comprises an array of light-sensitive elements (each referred to as a pixel), and wherein each light-sensitive element stores a charge proportional to the amount of light received over a certain measurement interval. In effect, the photons of light falling within the area defined by one pixel are converted into one (or more) electrons, and the number of electrons collected is proportional to the intensity at each pixel in the scene. The charge stored in each pixel is shifted through a series of gates to a readout register, where the charge is converted into a voltage signal that can be digitized and processed. The signals from a collection of pixels can be used to build an image of the scene of interest. The term "charge-coupled" refers to the coupling of electric potential that exists within the chemical structure of the silicon material that makes up the layers of the chip.

[0110] As the charge in each pixel is shifted sequentially, a CCD can produce high-quality images with low noise and high dynamic range. CCDs are small in size, simple to operate and easy to manufacture. This makes them ideal for integration in handheld devices.

[0111] The CCD can be configured to be sensitive to a variety of different wavelengths ranging from about 0.1 nm (soft x-rays) to about 400 nm (blue visible light) to about 1000 nm (near infrared), with peak sensitivity at about 700 nm. For purposes of embodiments of the present application, sensitivity across the near infrared (NIR) or infrared (IR) range can be preferred, although other wavelengths can also be used.

[0112] According to some examples, the light source 52 comprises an array of illumination elements, for example an elongate or linear array of illumination elements.

[0113] In some embodiments, the optical detector 54 can comprise an array of light detection elements, for example an elongate or linear array of light detection elements. For example, if a CCD camera is used, the pixels of the camera can form an elongate or linear array. This can be a ID linear array, or more preferably can be a 2D band of pixels.

[0114] In some embodiments, the light source comprises an elongate or linear array of illumination elements, and the optical detector comprises an elongate or linear array of light detection elements parallel to the elongate or linear array of illumination elements. This provides a structurally and functionally efficient configuration.

[0115] As one example, Figure 9 A configuration for the optical arrangement 50 is shown in accordance with one or more embodiments. The optical arrangement 50 comprises a light source comprising a first array of NIR or IR LEDs 52a and a second array of NIR or IR LEDs 52b, and an optical detector 54 comprising a NIR or IR sensitive CCD. Optionally, a NIR or IR filter 55 can be provided between the CCD and the surface of the tissue being illuminated, to restrict the light input to the CCD to the NIR or IR range only.

[0116] The light generated by the LED groups penetrates the skin of the subject and is reflected by the haemoglobin in the blood present in the superficial veins. The reflected light can pass through an optional IR or NIR filter before being captured by the CCD camera.

[0117] In some embodiments, the vein visualisation subsystem 40 can also comprise a proximity sensor for detecting the proximity distance between the optical detector 54 and the surface tissue of the user. A patch normalisation module can also be included for applying a positive or negative magnification to the detected light patterns, the magnitude of the magnification being determined in dependence on the sensed proximity distance, so as to normalise each detected light pattern size to a standardised size.

[0118] As mentioned above, in some embodiments, the vein visualization subsystem 40 can also comprise a pattern extraction module 44 for extracting a representation of the vein pattern from the detected light pattern prior to applying the classification module 34. In this case, the classification module is adapted to operate on the extracted representation of the vein pattern. This can be preferred in some embodiments, as the extraction of the vein pattern can ensure greater consistency in the classification and can make the classification process more reliable or straightforward.

[0119] In some embodiments, the representation of the vein pattern comprises a vein pattern mask. In some embodiments, the pattern extraction module can be a trained neural network trained to extract a vein pattern mask from the detected light pattern. Other example implementations of extracting a vein pattern can include more classical image processing or machine vision algorithms, such as edge detection, contour detection, threshold-based segmentation, region growing, skeletonization, and feature-based detection.

[0120] With respect to the option of applying a neural network to extract a vein pattern mask from the detected light pattern (e.g. the detected image), this can be implemented, for example, by training a convolutional neural network. The training data for the neural network can comprise a plurality of sample detected light patterns (e.g. images) containing a reflected representation of a vein pattern, the light patterns having the same format as the output from the optical detector, and wherein each sample detected light pattern is annotated with a ground truth vein pattern mask that has been extracted from the image, for example manually by a human operator or automatically via a different image processing tool or even another neural network for transfer learning.

[0121] Figure 10 An example vein pattern mask extracted from a captured light pattern (e.g. image) is depicted.

[0122] Reference is made, for example, to the project “Near-Infrared-to-Visible Vein Imaging via Convolutional Neural Networks and Reinforcement Learning project”, as of March 2023, which is located at the following URL: https: / / github.com / cviaai / VEINCV-RL. This method uses image segmentation as well as reinforcement learning to determine a mask of the veins.

[0123] Embodiments of the invention comprise functionality for providing one or more classifications of the biological structure or region to which the vein pattern represented in the detected pattern of light belongs. The system can comprise a classification module for performing this functionality.

[0124] By way of example, the one or more categories generated by the classification module can include one or more of: an identification of a body part to which the vein pattern belongs; a classification of one or more demographic characteristics of the individual to which the vein pattern belongs, such as age and / or gender; and / or a unique skin patch at which the vein pattern is located.

[0125] According to one set of embodiments, the suggestion classification module comprises a trained neural network.

[0126] For example, the trained neural network can be a network trained using training data comprising vein pattern masks annotated with ground truth information indicative of one or more of: a body part to which the vein pattern belongs, an age of the individual to which the vein pattern belongs; a gender of the individual to which the vein pattern belongs, and a unique skin patch identifier of a skin patch to which the vein pattern belongs.

[0127] In some embodiments, a method is applied which involves a training process for training a neural network, the training process comprising training the network using training data comprising vein pattern masks annotated with ground truth information indicative of a biological structure or region to which the vein pattern represented in the vein pattern mask belongs. For example, the ground truth information can be indicative of one or more of: a body part to which the vein pattern belongs, an age of the individual to which the vein pattern belongs; a gender of the individual to which the vein pattern belongs, and a unique skin patch identifier of a skin patch to which the vein pattern belongs.

[0128] In some embodiments, the method can further comprise quantizing the neural network after training.

[0129] The neural network can be quantized prior to integration into a device to reduce the size of the model, enabling it to be deployed by hardware with small memory and processing capabilities.

[0130] In some embodiments, the classification module can be for integration within a body processing device, and can comprise a memory storing the neural network and a processor for controlling the application of the neural network.

[0131] The training method can be provided as a separate aspect of the invention, or can be provided as part of the optical sensing method. For example, a method of training an artificial neural network can comprise receiving training data, wherein the training data comprises a plurality of training data entries, each training data entry comprising a vein pattern mask annotated with ground truth information indicative of a biological structure or region to which the vein pattern represented in the vein pattern mask belongs; and training the artificial neural network using the training data.

[0132] More details will now be given about the training of a suitable neural network model for generating a classification of a biological structure or region, which is the region to which the vein pattern represented in the detected light pattern belongs.

[0133] With respect to the generation of the training data set, the following considerations can be taken into account. The plurality of sample vein pattern masks included in the training data set can correspond to light patterns acquired from a population of people that is different in terms of demographics in order to ensure that the data set has a varied sample. In some embodiments, image enhancement can be applied to the acquired light patterns prior to extracting the vein pattern masks in order to take into account images acquired under different orientations and lighting conditions.

[0134] With respect to the neural network, according to some embodiments, it is proposed to use a convolutional neural network (CNN) having a U-net architecture.

[0135] A convolutional neural network (CNN) is an example of deep learning. Image classification involves extracting features from an input image and detecting patterns in the extracted features and certain classification class labels. Through a training process, the CNN learns the patterns associated with each class.

[0136] A method of training a CNN for classifying vein pattern masks can be as follows. An initial CNN model having a U-net architecture can be initially trained using a pre-existing image data set such as the COCO (Common Objects in Context) data set. The U-net is a widely used architecture for biomedical image classification.

[0137] After the initial training, further training can be performed using the specialized training data set discussed above, which includes vein pattern images annotated or labeled with one or more classifications. A simple approach is to train a single respective model to perform each individual classification that is desired. For example, if the classification module is desired to classify each of a user’s age, gender, and body part, then three separate models can be trained, each trained to generate one of the respective classifications.

[0138] Figure 11 The basic information flow achieved by using the classification module 34 is schematically illustrated.

[0139] The representation of the vein pattern 82, e.g. the vein pattern mask or the acquired light pattern, is provided as input to a classification module. In the current embodiment, three trained neural networks are applied to the input vein pattern: a first neural network 84 trained to classify the body part to which the vein pattern belongs; a second neural network 86 trained to classify the sex of the individual to which the vein pattern belongs; and a third neural network 88 trained to classify the age of the individual to which the vein pattern belongs. The combination of the outputs from these three neural networks provides a composite classification 90 of the combination of body part, age, and sex / gender.

[0140] Of course, it is not essential to provide multiple neural networks. In further embodiments, only a single neural network can be provided, trained to generate a single classification, e.g. one of age, sex and body part, or trained to generate multiple such classifications.

[0141] As will be discussed in more detail later, another example classification that a neural network can be trained to generate is the identification of the unique patch of skin in which the vein pattern is located. This can allow, for example, a navigation / localization function.

[0142] As mentioned above, embodiments of the application include a function to generate treatment-related information for at least one detected vein pattern using the output from the classification module. There are different choices for the type of treatment-related information that is generated.

[0143] In some embodiments, the treatment-related information comprises a setting for an operating parameter of the body treatment device.

[0144] For example, in some embodiments, the treatment-related information comprises electronic control instructions for adjusting an operating parameter of the body treatment device in dependence of the classification.

[0145] The system can include a memory module storing a look-up table allowing mapping between classifications and corresponding operating parameter settings. The specific mapping between classifications and settings can be pre-defined and / or can be configured by a user.

[0146] For example, by providing the ability to automatically detect the body part and / or the user demographics, a function can be provided that allows self-adjustment of the operating settings of the body treatment device without the need for external sensors or human intervention. This provides an improved experience for the user and also avoids the problem of the user applying incorrect settings. In the case of, for example, an optical hair removal device, incorrect settings with respect to the light intensity can be dangerous.

[0147] Figure 12An example processing logic according to one or more embodiments is schematically outlined, in which a classification generated by a classification module 34 is used to determine an operating parameter setting for a body treatment device. The classification module generates a classification 102. By way of example, the classification can include a classification of one or more of the following: an identification of a body part to which the venous pattern belongs; an age classification of the individual to which the venous pattern belongs; a sex or gender classification of the individual to which the venous pattern belongs. The classification can consist of a single one of these classifications or a composite classification including more than one. In the illustrated example, in a decision step 104, it is determined whether there is a custom user-defined setting for the mapping between the classification and the operating parameter setting. If not, a factory default mapping 108 is used. If so, a user-defined preference for the mapping 106 is used. For example, the user-defined settings can be configured via a user interface. They can be configured via an associated software application.

[0148] To further illustrate the concept of automatically configuring operating parameter settings based on venous visualization, some embodiments will be further illustrated below.

[0149] Figure 13 An example processing flow for enabling a multi-purpose body treatment device is shown according to a first example implementation. A multi-purpose body treatment device refers to a body treatment device that is switchable between a plurality of different body treatment modes, each mode performing a different type or class of body treatment. For example, the device can operate in each of a mechanical shaving mode and an optical hair removal mode. The workflow includes a user applying 112 the device to a body part; an optical arrangement acquiring 114 an optical pattern or image representation of a venous pattern of the body part; extracting 116 a venous pattern mask from the representation; classifying 118 a sex of the user by applying a trained classifier to the venous pattern mask, e.g., a trained neural network; classifying 120 a body part to which the venous pattern belongs by applying a trained classifier to the venous pattern mask, e.g., a trained neural network; classifying 122 an age of the user by applying a trained classifier to the venous pattern mask, such as a trained neural network; generating 124 control instructions for selecting an operating mode of the body treatment device according to the classifications, e.g., based on a lookup table.

[0150] For example, if the user is female and the mechanical hair removal mode is for male users, the system can be configured to select the optical hair removal mode. As a further example, if the body part is a face, the system can select the mechanical hair removal mode, or if the body part is a leg, the system can select the optical hair removal mode.

[0151] According to some embodiments, the device can be applied to a number of different body parts, and wherein the settings of the device vary according to the body part. For example, a shaver can be used on the face, chest, or armpit.

[0152] Figure 14 An example processing flow is shown according to an example implementation, wherein the body treatment device is a shaving device, and wherein the operating parameter comprises a cutting speed / rate of a cutting mechanism of the shaving device. For example, a higher cutting rate can be set when the device is applied to the face than when the device is applied to the chest. For example, a user can prefer a higher RPM shaver blade for beard cutting, and a lower RMP for cutting chest hair.

[0153] The workflow comprises a user applying 132 the device to a body part; the optical arrangement acquiring 134 an optical pattern or image representation of a venous pattern of the body part; extracting 136 a venous pattern mask from the representation; classifying 138 the body part to which the venous pattern belongs by applying a trained classifier to the venous pattern mask, such as a trained neural network; and generating 140 control instructions for selecting a cutting rate / speed of the shaver according to the classification, e.g. based on a lookup table.

[0154] According to some embodiments, the body care device can be an Intense Pulsed Light (IPL) hair removal device. Such a hair removal device allows for adjusting optical properties of the treatment light produced, e.g. adjusting a set intensity level of the light. This can be adjusted according to e.g. the body part and / or age.

[0155] Hence, in some embodiments, the operating parameter comprises an intensity level of the pulsed light produced by the device.

[0156] Figure 14 An example processing flow is shown according to an example implementation, wherein the body treatment device is an Intense Pulsed Light (IPL) hair removal device, and wherein the operating parameter comprises an intensity level of the pulsed light generated by the device.

[0157] The workflow comprises a user applying 152 the device to a body part; the optical arrangement acquiring 154 an optical pattern or image representation of a venous pattern of the body part; extracting 156 a venous pattern mask from the representation; classifying 158 the body part to which the venous pattern belongs by applying a trained classifier to the venous pattern mask, such as a trained neural network; classifying 160 the age of the user by applying a trained classifier to the venous pattern mask (e.g. a trained neural network), and generating 162 control instructions for configuring the intensity level of the light generated by the device according to the classification, e.g. based on a lookup table.

[0158] With respect to age dependency, the intensity level can be adjusted according to the age of the user, as older users have more fragile skin than younger users. Hence, for older users the intensity level can be set lower.

[0159] The optical sensing system 30 also has utility for obtaining performance or monitoring information about one or more usage / treatment sessions. In particular, according to a set of advantageous embodiments, the optical sensing system comprises functionality for tracking body coverage by the body treatment device by uniquely identifying skin patches being scanned by using a classifier, and by aggregating skin patch detections over one or more treatment sessions to track overall body coverage by the device.

[0160] As a non-limiting example, when using an optical hair removal device, such as an IPL device, for achieving optimal results, it is best to apply the device evenly and in a regular non-overlapping pattern to the body. Any gaps in the device application will result in poorer hair removal. At the same time, multiple overlapping of the same area can result in overexposure.

[0161] It would therefore be advantageous to provide functionality to guide uniform application. This also applies to any other body treatment device that requires uniform consistent application.

[0162] According to the presently described set of embodiments, detection of superficial vein patterns is therefore used to track device coverage on the body.

[0163] The same concept also allows reliable and accurate collection of exposure statistics, which is beneficial for guiding optimal use of the device as well as for helping usage-life based replacement of consumables, e.g. heads of the device.

[0164] Figure 16 Components of an example system according to one or more embodiments in which coverage tracking is performed are outlined.

[0165] The system can be identical to the system of any of the embodiments of Figure 4 , Figure 5 or Figure 6 except for at least the following additional features.

[0166] In this set of embodiments, the previously discussed classifier module 34 is adapted to generate a classification comprising an identification of a unique skin patch on which the vein pattern is located.

[0167] Furthermore, the system 30 also comprises a coverage tracking module 46 comprising one or more processors adapted to, during a body treatment session: receive an indicator of each detected unique skin patch; and store a record of each skin patch detection in a memory module (38) to compile a history of body treatment sessions.

[0168] In this set of embodiments, the treatment-related information generated by the processing module can comprise the body treatment session history described above. Additionally or alternatively, the treatment-related information can comprise analysis information derived from the body treatment session history.

[0169] For example, in some embodiments, the processing module 36 can be adapted to identify from the treatment session history any unique skin patches for which a plurality of detections is recorded in the session history, and wherein the treatment-related information comprises an indication of each unique skin patch for which a plurality of detections is recorded. This provides an indication of skin areas that have been covered more than once in treatment sessions. It is valuable to detect such areas because covering an area more than once can cause problems caused by overexposure.

[0170] In addition to or instead of detecting skin patches that have been covered more than once in a given session, it is also valuable to detect skin patches that have not been covered at all. To this end, a data structure such as a database can be provided that records a target set of unique skin patches that are to be covered in each treatment session of at least a given session type, and wherein the skin patches recorded in the treatment session history are compared to the target set of patches to identify patches that have not been covered in the treatment session.

[0171] To this end, in some embodiments, the processing module 36 can be constituted by an analysis module. The analysis module can be adapted to retrieve from the memory module a body treatment session history of at least one body treatment session. The analysis module can also be adapted to access a data structure recorded in a data store that records a set of known or assumed unique skin patches included by the user (i.e. a target set of skin patches of the user). In this case, the treatment-related information can comprise coverage feedback indicating coverage and / or non-coverage of different unique skin patches during the treatment session.

[0172] The above-described data structure recorded in the data store can comprise a body map comprising an indicator of a body location of each patch in the set of unique skin patches.

[0173] In some embodiments, the coverage feedback can comprise a graphical representation of the body map comprising a visual indication of coverage and / or non-coverage of each unique skin patch, and optionally wherein the processing module is adapted to communicate the coverage feedback to the user interface or the mobile computing device.

[0174] A function can be provided for implementing a calibration procedure for determining a set of known or assumed unique skin patches (i.e. a target set of skin patches of the user) comprised by the user. The procedure can comprise the following instructions: applying the optical arrangement of the vein visualisation subsystem on a plurality of different skin patch locations, and wherein the subsystem acquires a light pattern or image of superficial veins at each location. Thereby, a reference vein pattern mask is acquired for each defined location, and these reference vein pattern masks can form a set of known skin patches comprised by the user. This can be used to form a body map, as each vein pattern acquired in the visualisation procedure can be associated with a known location.

[0175] In a set of advantageous embodiments, the process related information (e.g. coverage information) is transmitted to an auxiliary device comprising a user interface for presenting information to the user, or can be transmitted to a server (e.g. a cloud based server) for facilitating access to the information from the portable computing device.

[0176] According to a set of advantageous embodiments, the process related information is received by a software application installed on the mobile computing device, and wherein the software application controls the user interface of the mobile computing device to provide a representation of the information.

[0177] For example, the software application can present a visual representation of the aforementioned body map showing the coverage and / or non-coverage of each unique skin patch. The software application can allow the user to display coverage information for a selected one of a plurality of different treatment sessions, recording a treatment history for a plurality of different treatment sessions, e.g. corresponding to different days. The software application can generate recommendations for the user based on the coverage information, e.g. guiding the user to provide more uniform and complete coverage.

[0178] To implement, the optical arrangement can preferably be integrated in the body treatment device. The preferred embodiments can use the above described array of parallel light sources and optical detector array, effectively providing a parallel light injection band and a scanning band. Additionally, it can be advantageous to incorporate the optional proximity sensor discussed previously for measuring the proximity distance of the optical detector to the skin surface, and to perform normalization to a uniform size for each scanned patch. By using this configuration, when pulling the device along a certain area, the vein visualisation subsystem will scan the vascular pattern and other possible markers in the underlying tissue, and these can be passed to the pattern extraction module and / or classification module.

[0179] According to some embodiments, a respective pair of light injection and scanning bands can be comprised at both ends of the operating treatment head of the device, e.g. at both ends of the head, such that scanning can be performed before and after exposure to the area. This can make the system less sensitive to the direction in which the user moves the head.

[0180] According to one or more embodiments, additional functionality can be provided for analyzing the light pattern (e.g., image) detected by the optical detector and / or analyzing the vein pattern representation (e.g., mask) extracted from the detected light pattern with a diagnostic analysis module. In particular, the inventors have recognized that the light information acquired by the optical detector during the course of operation can effectively serve as an auxiliary function for skin health or skin pathology assessment. For example, in some embodiments, the optical sensing system further comprises a diagnostic analysis module, and wherein the diagnostic analysis module is adapted to receive the pattern of light detected by the optical detector and / or the vein pattern mask extracted from the detected pattern of light, and to determine one or more diagnostic indicators or generate one or more diagnostic classifications. As one example, the diagnostic analysis module can be adapted to process the extracted vein pattern mask to derive a varicose vein diagnostic classification, e.g., presence / absence and / or severity. As another example, the diagnostic analysis module can be adapted to receive the light pattern (e.g., image) detected by the optical detector and detect or classify skin lesions. For example, mole detection and / or classification can be performed.

[0181] In some embodiments, the information derived by the diagnostic analysis module can be used to further configure one or more operational parameters of the device. For example, the treatment module can be adapted to adjust the intensity of the applied treatment light in response to detection of a skin lesion by the diagnostic analysis module falling within any of a predefined set of categories. For example, in the case of detection of a mole, the intensity can be reduced.

[0182] Additionally or alternatively, the information derived by the diagnostic analysis module can be used to provide additional feedback to the user or health care provider, e.g., through a user interface integrated in the body treatment device or optical sensing system, or through a secondary computing device such as an application installed on a mobile computing device. For example, the treatment module can be adapted to generate treatment recommendation feedback for the user depending on the type and / or severity of the detected skin lesion. For example, in response to detection of a skin lesion, a recommendation can be generated indicating a treatment depending on the detected lesion category, e.g., application of a topical medication such as a cream. In this context, “skin lesion” includes any abnormal pathology of the skin, including wounds, rashes, moles, inflammation, abscesses, etc.

[0183] As to implementation, the diagnostic analysis module can employ machine learning algorithms to generate diagnostic information from the light pattern and / or extracted vein mask, or can employ more classical types of machine vision algorithms, e.g., image segmentation, feature extraction, or pattern recognition. As to the option of using machine learning algorithms, one possibility is to train an artificial neural network (ANN), e.g., a convolutional neural network, to generate a diagnostic classification based on an input light pattern (or image) or extracted vein pattern mask. In some examples, there can be separate ANN models trained for detecting or classifying each different type of skin pathology of interest.

[0184] Embodiments of the optical sensing system described in this document can be applied to a variety of different use cases. While the examples discussed above primarily relate to use with body treatment devices, the same inventive concept can be applied substantially to any handheld device that is applied to a body during use, e.g. where positioning on the surface of the body can be useful. This can include for example handheld medical scanning devices such as ultrasound probes, or other body applied measurement devices. Another possible field of application is in hair counting, where it can be used to perform positioning of a skin patch covered by each of a series of acquired images, e.g. relative to a complete head map.

[0185] The various embodiments described above use a trained artificial neural network. The structure of an artificial neural network (or simply, neural network) is inspired by the human brain. A neural network comprises a plurality of layers, each layer comprising a plurality of neurons. Each neuron comprises a mathematical operation. In particular, each neuron can comprise a different weighted combination of a single type of transformation (e.g. the same type of transformation, a sigmoidal transformation, etc. but with different weights). In the process of processing input data, the mathematical operation of each neuron is performed on the input data to produce a numerical output, and the output of each layer in the neural network is sequentially fed to the next layer. The last layer provides the output.

[0186] Methods of training machine learning algorithms are well known. Typically, such a method comprises obtaining a training data set comprising training input data entries and corresponding training output data entries. An initialized machine learning algorithm is applied to each input data entry to generate a predicted output data entry. An error between the predicted output data entry and the corresponding training output data entry is used to modify the machine learning algorithm. This process can be repeated until the error converges, and the predicted output data entry is sufficiently similar to the training output data entry (e.g. ±1%). This is commonly referred to as a supervised learning technique.

[0187] For example, the weights of the mathematical operation of each neuron can be modified until the error converges. Known methods of modifying a neural network include gradient descent, backpropagation algorithms, etc.

[0188] The above-described embodiments of the application employ a processing device. The processing device can be a single processor or multiple processors. It can be located in a single housing, structure, or unit, or it can be distributed among multiple different housings, structures, or units. Accordingly, references to a processing device adapted or configured to perform a particular step or task can correspond to a single processing component or multiple processing components acting in concert. Those skilled in the art will understand how to implement such a distributed processing device. The processing device includes a communications module or input / output for receiving data and outputting data to other components.

[0189] The one or more processors of the processing device can be implemented in numerous ways, with software and / or hardware, to perform the various functions required. Typically, the processor(s) will employ one or more microprocessors that can be programmed using software (e.g., microcode) to perform the required functions. The processor(s) can be implemented as a combination of dedicated hardware to perform some functions and one or more programmed microprocessors and associated circuitry to perform other functions.

[0190] Examples of circuitry that can be employed in various embodiments of the application include, but are not limited to, conventional microprocessors, application specific integrated circuits (ASICs), and field-programmable gate arrays (FPGAs).

[0191] In various implementations, the processor(s) can be associated with one or more storage media such as volatile and non-volatile computer memory such as RAM, PROM, EPROM, and EEPROM. The storage media can be encoded with one or more programs that, when executed on one or more processors and / or controllers, perform the required functions. Various storage media can be fixed within a processor or controller or can be transportable, such that the one or more programs stored thereon can be loaded into a processor.

[0192] Variations of the disclosed embodiments can become apparent to those of ordinary skill in the art upon reading the foregoing description of the disclosed embodiments with reference to the drawings. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite articles "a" or "an" do not exclude a plurality. Means-plus-function or step-plus-function clauses can be used to describe features implemented with well-known components or structures. Such clauses should be construed to cover the described structure as well as equivalents thereof that operate in the same manner to accomplish the same function.

[0193] A single processor or other unit can implement the functions of several of the items recited in the claims.

[0194] The mere fact that measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0195] A computer program can be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state storage medium supplied together with or as part of other hardware, but can also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems.

[0196] If the term "adapted to" is used in the claims or specification it should be noted that the term "adapted to" is intended to be equivalent to the term "configured to".

[0197] The use of any of the following terms in the claims or specification should not be interpreted as limiting the scope:

Claims

1. An optical sensing system (30) for a body treatment device, comprising: a vein visualization subsystem (40) for integration in the body treatment device for detecting a pattern of superficial veins of a user, the vein visualization subsystem (40) comprising an optical arrangement (50), the optical arrangement (50) comprising: a light source (52) for directing light into surface tissue of the user during use of the body treatment device, and an optical detector (54) for detecting a pattern of light reflected from the surface tissue in which a vein pattern is represented; a classification module (34) for application to an output from the vein visualization subsystem to provide one or more classifications of a biological structure or region to which a detected vein pattern is attributed; and a processing module (36) adapted to use an output from the classification module for generating treatment-related information for at least one detected vein pattern.

2. The system of claim 1, wherein the one or more classifications generated by the classification module comprise: an identification of a body part to which the vein pattern is attributed; a classification of one or more demographic features of the individual to which the vein pattern is attributed, such as age and / or gender; and / or an identification of a unique skin patch in which the vein pattern is located.

3. The system of any one of claims 1 to 2, wherein the vein visualization subsystem further comprises a pattern extraction module for extracting a representation of a vein pattern from the detected pattern of light; and wherein the classification module is adapted to operate on the representation of the extracted vein pattern, and optionally wherein the representation of the vein pattern comprises a vein pattern mask.

4. The system of claim 3, wherein the representation of the vein pattern comprises a vein pattern mask, and wherein the pattern extraction module is a trained neural network trained to extract the vein pattern mask from a detected light pattern.

5. The system of any one of claims 1 to 4, wherein the classification module comprises a trained neural network, and optionally wherein the trained neural network is a network trained using training data comprising vein pattern masks annotated with ground truth information indicative of one or more of: a body part to which the vein pattern is attributed, an age of the individual to which the vein pattern is attributed, a gender of the individual to which the vein pattern is attributed, and a unique skin patch identifier of the skin patch to which the vein pattern is attributed.

6. The system of any one of claims 1 to 5, wherein the treatment-related information comprises electronic control instructions for adjusting an operating parameter of the body treatment device in dependence on the classification, and optionally wherein: the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises a selection between available treatment modes, with a first mode being a shaving mode and a second mode being an optical hair removal mode; or the operating parameter comprises The body treatment device is a shaving device, and the operating parameters include the cutting speed / rate of the cutting mechanism of the shaving device; or The body treatment device is an intense pulsed light (IPL) hair removal device, and the operating parameters include the intensity level of the pulsed light generated by the device.

7. The system (30) according to any one of the preceding claims, wherein The classifier module (34) is adapted to generate a classification that includes an identifier of the unique skin patch on which the vein pattern is located, and The system also includes a coverage tracking module (46), which includes one or more processors adapted to: Receive an indicator for each unique skin patch detected; and Records of each skin patch detection are stored in the memory module (38) to edit the body treatment history.

8. The system of claim 7, wherein the processing module is adapted to identify any unique skin patch from the treatment history in which multiple tests were recorded, and wherein the treatment-related information includes an indication of each unique skin patch in which multiple tests were recorded.

9. The system according to claim 7 or 8, wherein the processing module comprises an analysis module. The analysis module is adapted to retrieve the history of the body treatment for at least one body treatment from the memory module; The analysis module is also adapted to access data structures recorded in a data storage system, the data storage system comprising a set of known or assumed unique skin patches included by the user; and The treatment-related information includes coverage feedback, which indicates the coverage and / or non-coverage of the unique skin patch during the treatment course.

10. The system according to any one of the preceding claims, wherein the vein visualization subsystem further comprises: A proximity sensor for detecting the proximity distance between the optical detector and the user's surface tissue; as well as A patch normalization module is used to apply a positive or negative amplification to the detected light pattern, the size of which depends on the sensed proximity distance, in order to normalize each detected light pattern size to a standardized size.

11. A body processing device comprising the optical sensing system as described in any of the preceding claims.

12. An optical sensing method for use during the application of a body processing device to a user's body, the method comprising: Detecting the vein pattern of a user's superficial veins, wherein the detection includes: During use of the body treatment device, light is directed into the user's surface tissues; and In response to the guidance of the light, the pattern of light reflected from the surface tissue is detected; The application classification module provides a classification of one or more biological structures or regions to which the vein pattern represented in the detected pattern of light belongs; and For at least one detected vein pattern, information related to processing is generated using the output from the classification module.

13. The method of claim 12, wherein the classification module comprises a trained neural network, and wherein the method comprises a training process for training the neural network, the training process comprising training the network using training data, the training data comprising a vein pattern mask annotated with baseline truth information indicating one or more of the following: The body part to which the vein pattern belongs. The age of the individual to whom the vein pattern belongs. The sex of the individual to which the vein pattern belongs, and The vein pattern is a unique skin block identifier for the skin block to which it belongs.

14. A computer program product comprising machine-executable instructions, said machine-executable instructions being configured, when executed by a processor, to cause the processor to: The detection of vein patterns in a user's superficial veins is controlled as follows: communicate with a light source to control the light source to direct light into surface tissue of the user during use of the body treatment device; as well as Communicating with an optical detector to control the optical detector to detect the pattern of light reflected from the surface tissue in response to the guidance of the light; The application classification module provides a classification of one or more biological structures or regions to which the vein patterns represented in the detected patterns of light belong; as well as For at least one detected vein pattern, information related to processing is generated using the output from the classification module.

15. A method for training an artificial neural network, comprising: Receive training data, wherein the training data includes multiple training data entries, each training data entry including a vein pattern mask, the vein pattern mask being annotated with reference truth information indicating the biological structure or region to which the vein pattern represented in the vein pattern mask belongs; as well as The artificial neural network is trained using the training data.

Images