Dysphagia rehabilitation training device and control method thereof

Abstract

The present application relates to the technical field of rehabilitation equipment, and particularly relates to a dysphagia rehabilitation training device and a control method thereof. The device comprises: a gas pump and a training assembly connected through a gas pipe, wherein the training assembly comprises a plurality of air bags; a first pressure sensor is arranged on the surface of the training assembly, the first pressure sensor is used for collecting a first pressure, and the first pressure represents the tongue pressure of a user; a second pressure sensor is used for collecting a second pressure, and the second pressure represents the pressure in the air bag; a solenoid valve is used for controlling the flow and on-off of the gas in the gas pipe; and a controller is connected with the gas pump, the solenoid valve, the first pressure sensor and the second pressure sensor, and is used for realizing the control of the pressure in any air bag. The device can realize the partition training of the tongue of the user, and improves the rehabilitation effect.

Description

A dysphagia rehabilitation training device and its control method Technical Field

[0001] This invention relates to the field of rehabilitation equipment technology. More specifically, this invention relates to a rehabilitation training device for swallowing disorders and its control method. Background Technology

[0002] Dysphagia is a syndrome characterized by difficulties in the safe and effective transport of food to the stomach due to structural and / or functional impairment of organs such as the jaw, lips, tongue, soft palate, pharynx, and esophagus during oral preparation, oral transport, and pharyngeal and esophageal stages. Dysphagia can affect food intake and nutrient absorption, and can also lead to aspiration pneumonia caused by food aspiration into the trachea, which can be life-threatening in severe cases.

[0003] However, most existing rehabilitation training devices are based on manual operation, lack automation and intelligence, and cannot accurately measure the pressure of different parts of the tongue (such as the front, middle and back of the tongue). Furthermore, users find it difficult to conduct targeted training for weak areas, resulting in mediocre training effects.

[0004] Therefore, how to assist users in targeted training based on the patient's tongue muscle capabilities is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] To address the aforementioned technical problem of the inability to provide targeted training based on the patient's tongue muscle capabilities, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a dysphagia rehabilitation training device, comprising: an air pump and a training component connected via a trachea, the training component including multiple air bladders; a first pressure sensor disposed on the surface of the training component, the first pressure sensor being used to collect a first pressure, the first pressure representing the pressure of the user's tongue; a second pressure sensor, the second pressure sensor being used to collect a second pressure, the second pressure representing the pressure inside the air bladders; a solenoid valve, the solenoid valve being used to control the flow rate and on / off state of gas in the trachea; and a controller, the controller being connected to the air pump, the solenoid valve, the first pressure sensor, and the second pressure sensor, for controlling the pressure inside any one of the air bladders.

[0007] Furthermore, the controller includes a correction module and a display module, the correction module being connected to the display module, wherein the correction module is used to correct the effect of saliva on the tongue pressure, and the display module is used to display the user's pressure heat map.

[0008] Furthermore, the device also includes an alarm connected to the controller.

[0009] Furthermore, the alarm is a buzzer or a speaker.

[0010] Furthermore, the first pressure sensor is a flexible matrix sensor or a capacitive sensor.

[0011] In a second aspect, the present invention provides a control method for a dysphagia rehabilitation training device based on any one of the first aspects, the dysphagia rehabilitation training device further comprising a second pressure sensor for collecting pressure within a target airbag, the method comprising: acquiring a first pressure collected by a first pressure sensor during user training and a second pressure collected by a second pressure sensor during user training; in response to a training mode of self-feedback training, controlling an air pump and a solenoid valve to operate based on the difference between the first pressure and the second pressure, so that the pressure value within the target airbag reaches a set value.

[0012] Furthermore, the method also includes: in response to the training mode being active training, determining the target pressure based on the ultimate tongue pressure, controlling the solenoid valve and the air pump to make the pressure inside the target airbag reach the target pressure, wherein the target pressure is positively correlated with the ultimate tongue pressure, and the ultimate tongue pressure represents the maximum pressure value when the user forcefully presses their tongue against the target airbag.

[0013] Furthermore, the method also includes: in response to the training mode being passive training, adjusting the pressure value inside the target airbag at a preset speed based on the extreme tongue pressure.

[0014] Furthermore, after acquiring the first pressure, the method also includes: calibrating the first pressure based on Kalman filtering.

[0015] Furthermore, the method also includes: if the first pressure is detected to remain unchanged within a preset time, triggering an alarm indicating an abnormal situation.

[0016] The beneficial effects of this invention are as follows: The device of this invention can accurately measure tongue pressure and also enables zoned training of the user's tongue. Furthermore, it provides multiple training modes to meet the training needs of different users, improving training effectiveness. Especially in the self-feedback mode, it can dynamically adjust the training intensity according to the user's real-time situation, ensuring safe and effective training. Furthermore, calibrating the collected tongue pressure avoids signal drift caused by saliva, thereby improving the accuracy of tongue pressure measurement and evaluation. Furthermore, it can promptly trigger an alarm in case of abnormal situations during training, such as tongue cramps, thus ensuring the safety of the user's training. Attached Figure Description

[0017] Figure 1 is a schematic structural block diagram illustrating a dysphagia rehabilitation training device according to an embodiment of the present invention;

[0018] Figure 2 is a schematic structural block diagram illustrating a dysphagia rehabilitation training device according to another embodiment of the present invention;

[0019] Figure 3 is a flowchart schematically illustrating a control method for a dysphagia rehabilitation training device according to an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0022] Swallowing disorders affect food intake and nutrient absorption, and can also lead to aspiration pneumonia caused by food being aspirated into the trachea, which can be life-threatening in severe cases. Most existing swallowing disorder training devices can only train the entire tongue. Since the degree of impairment varies across different parts of the tongue, training the entire tongue fails to effectively train weaker areas, thus prolonging training time. Furthermore, existing devices have poor versatility, failing to meet the training needs of different users and unable to adjust the training intensity in real time according to the user's actual situation.

[0023] In this regard, in a first aspect, the present invention provides a swallowing disorder rehabilitation training device, as shown in FIG1. ​​The swallowing disorder rehabilitation training device of the present invention includes: a training component, a trachea, an air pump, a solenoid valve, a first pressure sensor, a second pressure sensor, and a controller. The device can be used for active and passive training of the tongue / tongue muscles to assist the user in restoring swallowing function.

[0024] The training component is connected to an air pump via a trachea. Specifically, the training component includes multiple independent air bladders, which can be made of food-grade silicone. This training component, composed of these air bladders, can cover the front, middle, and back of the tongue, thereby enabling the acquisition of complete tongue muscle pressure (tongue pressure / tongue area pressure). Furthermore, by allowing users to select one or more air bladders according to their needs, they can achieve zoned and targeted training of the tongue muscles, avoiding the limitation of existing devices that can only train the entire tongue, thus improving the user's training effect.

[0025] It should be noted that the number of airbags corresponds to the number of tracheas. In one embodiment, the training component consists of four airbags, and therefore, four tracheas are correspondingly provided (as shown in Figure 2). Specifically, one end of each of these four tracheas is connected to a corresponding airbag, and the other end is connected to the same main trachea (one-piece molding, i.e., one trachea has four branch tracheas) before being connected to the air pump. In another embodiment, these four tracheas may not be connected to the main trachea, i.e., each is connected to the air pump individually. It should be noted that the tracheas connected to one end of the four airbags in Figure 2 actually contain four separate tracheas, thereby achieving control of the pressure inside different airbags.

[0026] Furthermore, each air tube is equipped with a corresponding solenoid valve (i.e., the number of air bladders matches the number of solenoid valves). This solenoid valve controls the flow and on / off state of the gas within the air tube, thereby enabling individual control of the pressure within any single air bladder. By controlling the pressure within any air bladder through the solenoid valves, the training needs of different users can be met. Specifically, those with weaker tongue muscle strength can reduce the pressure within the corresponding air bladder, while those with stronger tongue muscle strength can increase the pressure within the air bladder. In optional embodiments, those skilled in the art can select a suitable flow controller based on actual needs.

[0027] The first pressure sensor, encapsulated in food-grade silicone, can be disposed on either the outer or inner surface of the training component. It is used to collect the pressure exerted on the training component, specifically the pressure from the user's tongue. Furthermore, the first pressure sensor has multiple pressure sampling points to achieve complete collection of the user's tongue pressure. In one embodiment, the first pressure sensor can be a flexible matrix sensor. In alternative embodiments, those skilled in the art can select a suitable sensor, such as a capacitive sensor, based on actual needs.

[0028] The second pressure sensor is used to collect the pressure inside the airbag, i.e., the degree of inflation or stiffness of the airbag, so that the airbag can be adjusted to a target pressure value based on the current pressure value. In an optional embodiment, the number of second pressure sensors can be set to be the same as the number of airbags.

[0029] Furthermore, the air pump, solenoid valve, first pressure sensor, and second pressure sensor are all connected to the controller, thereby enabling the control of the pressure inside the airbag and the acquisition, processing, and display of pressure data (including data collected by the first pressure sensor and data collected by the second pressure sensor).

[0030] In one embodiment, the controller includes a signal conversion module, a correction module, and a display module connected in sequence. The signal conversion module converts analog signals into digital signals, and the correction module filters and corrects the acquired tongue pressure data. In one embodiment, the correction module uses Kalman filtering to address signal drift caused by saliva, thereby improving the reliability and accuracy of the obtained tongue pressure data. The specific processing procedure is described below and will not be repeated here.

[0031] Furthermore, the display module shows a pressure heat map of the user's tongue. Through this pressure heat map, the user can clearly understand their tongue muscle strength, allowing them to use the device for selective and focused training based on their individual abilities, thereby improving rehabilitation outcomes.

[0032] In addition, the controller also includes a power module and a Bluetooth module. The power module is used to supply power to the air pump, solenoid valve, etc., and the Bluetooth module is used for communication.

[0033] In one embodiment, the training device of the present invention further includes an alarm connected to a controller. Specifically, the alarm can be a speaker, a buzzer, an audible and visual alarm, or an LED, etc. During training, when the user's tongue pressure is detected to remain unchanged for a period of time, an alarm is triggered. The alarm method can be a speaker broadcast, a continuous buzzer, a red LED light, or a flashing red LED, etc., thereby enabling medical personnel to confirm whether the patient has experienced an unexpected situation (such as tongue cramps), thus ensuring the user's safety during training.

[0034] In one embodiment, gold nanowire electrodes can be used to ensure the long-term stability of the pressure sensor, photolithography can be used to fabricate interlayer interconnects, and micro lithium batteries can be used for wireless power supply.

[0035] When using the device, the training component is placed under the tongue, and the user presses their tongue firmly against it. At this time, the first pressure sensor collects the pressure data from different parts of the tongue and transmits it to the controller. The controller processes this pressure data and displays it as a pressure heatmap. The user can then use this heatmap to select active training, self-feedback training, or passive training on the controller. Furthermore, the user selects the airbags to use based on their needs, and the device deflates any unused airbags. During training, the user can monitor the effectiveness of their training in real time through the pressure heatmap.

[0036] When a user needs to perform active training, they first select the target airbag based on their needs. Then, they manually set the pressure of the target airbag or use the system-recommended target pressure (calculated based on the ultimate tongue pressure). The device then controls the air pump and solenoid valve to bring the target airbag to the target pressure and deflates the other airbags (excluding the target airbag). The user then performs active training based on the target airbag. By selecting the target airbag according to actual needs, different areas of the tongue can be trained individually, thereby improving training effectiveness and shortening rehabilitation time.

[0037] When users need to perform self-feedback training, they first select the airbag to use. Then, the system dynamically adjusts the pressure value of the target airbag through an air pump and a solenoid valve based on the pressure data during the assessment (maximum tongue pressure) and the real-time tongue pressure during training. By dynamically adjusting the pressure value inside the target airbag according to the user's actual situation, it ensures that the user is always training at an appropriate intensity, avoiding the problem of tongue fatigue after a period of training while still training at a high intensity, thus ensuring the reliability and safety of the training.

[0038] When a user needs passive training, the solenoid valve and air pump are controlled based on the user's maximum tongue pressure during the assessment, causing the airbag to inflate and deflate at a specific rate. Understandably, the pressure of the target airbag continuously changes during training, thereby inducing up and down movement of the user's tongue. This gradually improves the range of tongue movement and strengthens tongue muscles, making it suitable for users with severe tongue impairments and for relaxation training. Furthermore, the inflation and deflation rate of the target airbag can be manually set to meet the needs of different users.

[0039] In summary, the training device of this invention not only targets different parts of the tongue for training, improving the rehabilitation effect, but also allows for the selection of different training methods (active training, self-feedback training, and passive training), thus meeting the training needs of different users and demonstrating strong versatility. Furthermore, by incorporating an alarm, abnormal situations during training can be detected promptly, ensuring user safety.

[0040] In a second aspect, the present invention provides a control method based on the dysphagia rehabilitation training device described in the first aspect, as shown in FIG3. The method of the present invention includes:

[0041] S101, Collect the user's maximum tongue pressure, the first pressure and the second pressure during the training process.

[0042] Specifically, the user selects the assessment mode on the controller, then places the training component in their mouth and presses their tongue forcefully against it. At this point, the first pressure sensor collects data. In one embodiment, the maximum pressure value collected at each pressure sampling point can be taken as the user's limit tongue pressure and displayed on the screen as a pressure heatmap. Furthermore, the user can also view the maximum, minimum, and average pressure values ​​for the tongue regions corresponding to different airbags. By displaying the pressure heatmap, the user can intuitively understand the obstacles in different parts of their tongue and then focus their training on the weaker areas.

[0043] In one embodiment, normal tongue pressure can also be collected. Specifically, the user places their tongue on the training component with normal force in a relaxed state; the pressure collected at this time is the normal tongue pressure. After collecting the normal tongue pressure, the system can display the user's normal tongue pressure, the tongue pressure of a normal person (of the same age and gender as the user), and the difference between the two, thereby enabling the user to clearly understand their rehabilitation progress and goals. Furthermore, if historical data exists, the system can display the most recent normal tongue pressure readings, allowing the user to clearly understand their training effectiveness.

[0044] It should be noted that the airbag is fully inflated when collecting extreme and normal tongue pressure data. Furthermore, the first pressure represents the pressure value collected by the first pressure sensor when the user presses their tongue against the target airbag, and the second pressure represents the pressure value inside the target airbag collected by the second pressure sensor.

[0045] In one embodiment, after acquiring tongue pressure (including extreme tongue pressure, normal tongue pressure, and tongue pressure during training), the method further includes: calibrating the tongue pressure based on Kalman filtering (implemented by a correction module).

[0046] First, make a prediction, including x. k =Fx k-1 State prediction and P k-1 =FP k-1 F T +Q error covariance prediction, x k The filtered output at the current calibration time, x k-1 This is the filtered output at the previous calibration time. F is the transition matrix, used to describe the trend of signal change; Q is the process noise, used to reflect the uncertainty of the model, such as the drift rate.

[0047] Then, measurements are updated, including the Kalman gain K. k =P k - H T HP K - H T +R)-1 and state correction x k =x k - +K k (z k -Hx k - H is the observation matrix of the sensor model, R is the observation noise of the short-term noise of the first pressure sensor, and z k This is the reading of the first pressure sensor at the current calibration time.

[0048] Furthermore, the observed value at the calibration time is used as the true value z. cal And force reset the current state x of the Kalman filter. K This is to eliminate the accumulated drift error from the previous calibration time. Further, the residual (∈ c =z cal -Hx K - If the noise level is greater than a preset threshold, it indicates severe drift. In this case, the process noise Q is increased or the observation noise R is decreased to make the filter more reliant on the calibration data. Finally, the drift model parameters are estimated. In one embodiment, the drift rate can be fitted using the calibration data, and the state equation can be updated:

[0049] d k =d k-1 +βΔt+w k ;

[0050] In the formula, d k For the drift term at the current calibration time, d k-1 This represents the drift term from the previous calibration time, where Δt is the difference between the current calibration time and the previous calibration time (in this embodiment, it is 1 hour), and w k β is a constant, and β is the drift rate, which can be updated by least squares or Gaussian estimation.

[0051] By calibrating tongue pressure, the influence of saliva on signal drift can be avoided, thereby improving the accuracy and reliability of tongue pressure measurement and evaluation.

[0052] S102. Control the pressure value inside the target airbag according to the training mode selected by the user.

[0053] Specifically, users can select a training mode according to their actual needs. In this embodiment, the training modes include active training, self-feedback training, and passive training. Active training refers to a fixed inflation level of the target airbag, with the user providing the power during training; self-feedback training refers to a change in the inflation level of the target airbag, with the user providing the power during training; and passive training refers to a change in the inflation level of the target airbag, with the airbag providing the power during training.

[0054] In active training mode, the user selects the airbags to be used (one or more target airbags) and sets the airbag pressure value according to their needs. Specifically, the solenoid valve and air pump are controlled to inflate and deflate the target airbags to bring the pressure inside the target airbags to the specified value, and unused airbags are deflated. Alternatively, the user can use the target pressure of the target airbags automatically generated by the system. In one embodiment, the target pressure automatically generated by the system can be determined based on the ultimate tongue pressure, and the target pressure is positively correlated with the ultimate tongue pressure. Specifically, the maximum pressure value (ultimate tongue pressure) collected at each pressure sampling point corresponding to the target airbag is determined, and then the average value is calculated. A certain percentage (e.g., 85%, 90%) of this average value is used as the target pressure, and then the target airbag is inflated and deflated to bring the pressure inside the target airbags to the target pressure.

[0055] By setting the target pressure based on the user's maximum tongue pressure, effective training can be ensured, avoiding the problem of fatigue caused by excessive pressure setting and insufficient training and stimulation of the tongue muscles caused by insufficient pressure setting. Furthermore, it allows training of any part of the tongue, avoiding the problem of traditional methods that can only train the entire tongue, thus improving the rehabilitation effect.

[0056] In self-feedback mode, the user selects the airbag to use based on their needs. The system then adjusts the pressure value inside the target airbag according to the difference between the first and second pressures during training, ensuring the target airbag reaches a set value that is positively correlated with the first pressure. Specifically, the first pressure is multiplied by an adjustment coefficient to obtain a reference tongue pressure (set value). The system then determines whether the reference tongue pressure equals the second pressure. If it does, the target airbag is not inflated or deflated. If not, the target airbag is inflated or deflated to ensure the pressure inside the target airbag matches the reference tongue pressure. That is, if the reference tongue pressure is greater than the second pressure, the air pump and solenoid valve are controlled to inflate the target airbag until the pressure inside the target airbag matches the reference tongue pressure; if the reference tongue pressure is less than the second pressure, the air pump and solenoid valve are controlled to deflate the target airbag until the pressure inside the target airbag matches the reference tongue pressure.

[0057] The adjustment coefficient is greater than 1, and in one embodiment, it can be 1.2. Furthermore, the adjustment cycle for the pressure value inside the target airbag is once every 50 ms. In optional embodiments, those skilled in the art can set the adjustment coefficient and adjustment cycle according to actual needs.

[0058] By dynamically adjusting the pressure of the target airbag according to the user's actual situation, it ensures that the user always trains within an appropriate intensity, adapting to the user's needs and avoiding the problem of continuing to train at a high intensity after fatigue, thus improving the safety and reliability of training. Simultaneously, by setting the airbag pressure according to the user's actual situation, it ensures that the user's weak areas receive sufficient and targeted training, achieving zoned training of the tongue and improving the rehabilitation effect of the user's weak areas.

[0059] In passive training mode, the pressure value inside the target airbag is adjusted at a preset speed according to the maximum tongue pressure. Specifically, during training, the air pump and solenoid valve are controlled to inflate the target airbag from empty to the maximum tongue pressure or the preset value, and then the air pump and solenoid valve are controlled to deflate the target airbag from the maximum tongue pressure or the preset value back to empty. This process is repeated, thereby enabling users with greater difficulty to move their tongues, gradually increasing the range of tongue movement and strengthening tongue muscles.

[0060] In optional embodiments, other training modes can also be set, such as progressive training and random training. In progressive training, the pressure range consisting of normal tongue pressure and maximum tongue pressure is divided into a certain number of segments (e.g., 5 segments), and then a specified training duration is performed for each segment. After the training is completed, the next segment is trained. In one embodiment, training can start from the maximum tongue pressure to avoid premature fatigue. Random training, on the other hand, randomly uses a specific segment for training.

[0061] By setting different training modes for different users, we can meet the training needs of different users, ensure that users receive sufficient training, and improve the training effect.

[0062] Furthermore, the method of the present invention also includes: if the first pressure remains unchanged within a preset time, triggering an alarm indicating an abnormal situation, and simultaneously deflating the target airbag. It should be noted that if the user selects multiple airbags for training, as long as the tongue pressure corresponding to any one airbag remains unchanged within the preset time, the system immediately issues a control command to activate the alarm, alerting relevant personnel to a potential abnormal situation (e.g., tongue cramps). By monitoring whether the user's tongue pressure remains unchanged in real time, abnormal situations can be handled promptly, thereby improving the safety of training.

[0063] Furthermore, the method of the present invention also includes: generating a training report in response to the end of training, the training report containing detailed training data of the user, such as normal tongue pressure, maximum tongue pressure, and the pressure value of the target airbag. Through this training report, the user can clearly understand their training status and training effect, thereby providing a reference for adjustment and optimization of the next training session.

[0064] In the description of this specification, "multiple" means at least two, such as two, three or more, unless otherwise explicitly specified. Furthermore, the steps described above are for clarity only; in implementation, they can be combined into one step or some steps can be broken down into multiple steps, as long as they include the same logical relationships.

[0065] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A rehabilitation training device for swallowing disorders, characterized in that, include: An air pump and a training component are connected via a trachea. The training component includes multiple airbags. A first pressure sensor is provided on the surface of the training component to collect a first pressure, which represents the pressure applied by the user to the surface of the training component by the tongue. A second pressure sensor is provided to collect a second pressure, which represents the pressure inside the airbags. A solenoid valve is used to control the flow and on / off of gas in the air tube connecting the air pump and the training component; a controller is connected to the air pump, the solenoid valve, the first pressure sensor, and the second pressure sensor to control the pressure in any one of the airbags; the controller includes a correction module and a display module, the correction module being connected to the display module, wherein the correction module is used to correct the effect of saliva on the tongue pressure, and the display module is used to display the user's pressure heatmap.

2. The dysphagia rehabilitation training device according to claim 1, characterized in that, It also includes an alarm that is connected to the controller.

3. The dysphagia rehabilitation training device according to claim 2, characterized in that, The alarm is a buzzer or a speaker.

4. The dysphagia rehabilitation training device according to claim 1, characterized in that, The first pressure sensor is a flexible matrix sensor or a capacitive sensor.

5. A control method for the dysphagia rehabilitation training device according to claim 1, characterized in that, The method includes: acquiring a first pressure during user training collected by a first pressure sensor, and a second pressure during user training collected by a second pressure sensor; in response to a self-feedback training mode, controlling an air pump and a solenoid valve to operate based on the difference between the first pressure and the second pressure, so that the pressure value inside the target airbag reaches a set value.

6. The control method for the dysphagia rehabilitation training device according to claim 5, characterized in that, Also includes: In response to the training mode being active training, the target pressure is determined based on the ultimate tongue pressure, and the solenoid valve and air pump are controlled to make the pressure inside the target airbag reach the target pressure. The target pressure is positively correlated with the ultimate tongue pressure, which represents the maximum pressure value when the user forcefully presses their tongue against the target airbag.

7. The control method for the dysphagia rehabilitation training device according to claim 6, characterized in that, Also includes: In response to the training mode being passive training, the pressure value inside the target airbag is adjusted at a preset speed based on the extreme tongue pressure.

8. The control method for the dysphagia rehabilitation training device according to claim 5, characterized in that, After acquiring the first pressure, the method further includes calibrating the first pressure based on Kalman filtering.

9. The control method for the dysphagia rehabilitation training device according to claim 5, characterized in that, Also includes: If the first pressure remains unchanged within a preset time, an alarm for an abnormal situation is triggered.

Images