KR102835008B1 - Non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence - Google Patents

Abstract

The present invention relates to a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence, and more specifically, to a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence, the system comprising: an imaging device which irradiates crossed light sources having different wavelengths to a body part and captures reflected light reflected from the body epidermis by the irradiated crossed light sources to generate a sequence image; a lesion progression analysis unit which analyzes lesion progression for diagnosing and predicting pressure ulcers based on non-contact data of the sequence image generated by the imaging device; a display which visually displays the pressure ulcer progression status and predicted situation diagnosed and predicted by the lesion progression analysis unit; and a database which includes a pressure ulcer diagnosis DB and a pressure physics engine DB so as to enable diagnosis based on standardized data on the grade of pressure ulcers.
According to the non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence proposed in the present invention, the system comprises an imaging device which irradiates crossed light sources having different wavelengths to a body part, captures reflected light reflected from the body epidermis by the irradiated crossed light sources, and generates a sequence image; a lesion progression analysis unit which analyzes the lesion progression for diagnosing and predicting pressure ulcers based on non-contact data of the sequence image generated by the imaging device; a display which visually displays the diagnosed and predicted pressure ulcer progression status and predicted situation through the lesion progression analysis unit; and a database which includes a pressure ulcer diagnosis DB and a pressure physics engine DB to enable diagnosis based on standardized data on the grade of pressure ulcers, thereby enabling diagnosis of the progression status of pressure ulcers and prediction of their course by measuring the blood flow rate in the epidermis using an image sequence in which the epidermis is captured using crossed light sources.
In addition, according to the non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence of the present invention, a pressure ulcer site is diagnosed non-contactly using an image sequence captured using a crossed light source and a camera, and the progression degree and future course of the pressure ulcer can be predicted using a big data technique, thereby objectively detecting skin abnormalities caused by the occurrence of pressure ulcers by visualizing blood flow using the difference in light absorption wavelengths of oxidized hemoglobin and hemoglobin, diagnosing pressure ulcers based on standardized data on the grade of pressure ulcers, and visualizing the predicted site where pressure ulcers may progress and the predicted matters for preventing them, thereby providing a composite solution.
In addition, according to the non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence of the present invention, by analyzing the shape of the pressure ulcer lesion and analyzing the shape and direction of pressure applied to the wounded area, it is possible to predict pressure ulcers that may occur in the future and project them onto the patient's affected area, and analyze pressure factors that must be removed to prevent pressure ulcers so as to provide a guideline for changing body position.

Description

{NON-CONTACT DATA-BASED PRESSURE ULCER DIAGNOSIS AND PREDICTION SYSTEM USING VISUAL INTELLIGENCE}

The present invention relates to a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence, and more specifically, to a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence that can predict the progression and progress of pressure ulcers by measuring the amount of blood flow in the epidermis using an image sequence of the epidermis captured using a crossed light source.

In general, pressure ulcers (bedsores) are a condition in which continuous pressure is applied to a part of the body while sitting or lying in one position, resulting in impaired blood circulation (ischemic state) in that part, causing damage to the subcutaneous tissue (ulcers) in that part. These bedsores often occur in unconscious patients, patients with brain or spinal nerve damage who cannot move, critically ill patients, and patients with general weakness. Bedsores are usually pressure necrosis that occurs in areas that directly touch the floor when a critically ill patient has been lying in bed for a long time. When there is no movement for a long time, the skin over the protruding bone is pressed, and blood circulation is cut off, causing the skin to die and rot due to lack of oxygen, resulting in bedsores.

Also, bedsores can occur anywhere, but they often occur in areas that receive a lot of pressure, such as the back of the head, shoulders, buttocks (sacrum, ischium), greater trochanter (when lying on your side), heels, and bony prominences such as the shins. In the early stages of bedsores, the skin in the area receiving pressure turns red, then an ulcer-like sore develops on the skin in that area, and further, skin necrosis progresses. In other words, stage 1 bedsores represent a condition in which continuous epidermal rash and inflammation occur in normal skin, stage 2 bedsores represent a superficial ulcer state in which only the dermis layer is destroyed, and stage 3-4 bedsores represent a state in which the entire skin layer, including the skin and subcutaneous fat, is destroyed.

In addition, bedsores are not damaged to the same extent as the entire affected area, but rather appear in different degrees in different parts. In other words, even in the same affected area, they show various forms from stage 1 to stage 4 depending on the area. Therefore, even in the same affected area, appropriate treatment and treatment methods are required. In particular, bedsores appear in various sizes and degrees, so accurate diagnosis and application of appropriate treatment are necessary.

In this way, although there are clear criteria for diagnosing bedsores, most of the diagnoses are determined by confirmation according to the risk assessment system by the visual inspection of a doctor. This is a diagnosis based on experience, and may show different diagnosis contents depending on the diagnosing person. The conventional method that relies on the experience of the diagnosing person has a high probability of misdiagnosis, so a diagnosis system based on standardized data on the grade of bedsores is needed. In addition, there was a problem that bedsores could continuously recur if only primary treatment such as dressing was performed without complex treatment such as changing the position or correcting the posture. Korean Patent Publication No. 10-2012-0072701 is disclosed as a prior art document.

The present invention has been proposed to solve the above problems of the existing proposed methods, and the purpose of the present invention is to provide a pressure ulcer diagnosis and prediction system based on non-contact data using visual intelligence, which comprises an imaging device which irradiates a body part with crossed light sources having different wavelengths, captures reflected light reflected from the body epidermis by the irradiated crossed light sources, and generates a sequence image, a lesion progression analysis unit which analyzes the lesion progression for diagnosing and predicting pressure ulcers based on non-contact data of the sequence image generated by the imaging device, a display which visually displays the pressure ulcer progression status and predicted situation diagnosed and predicted through the lesion progression analysis unit, and a database which has a pressure ulcer diagnosis DB and a pressure physics engine DB to enable diagnosis based on standardized data on the grade of pressure ulcers, thereby measuring the blood flow in the epidermis using an image sequence in which the epidermis is captured using crossed light sources, thereby diagnosing the progression status of pressure ulcers and predicting their progress.

In addition, the present invention provides a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence, which non-contactly diagnoses a pressure ulcer site using an image sequence captured using a crossed light source and a camera, and predicts the progression degree and future course of the pressure ulcer using a big data technique, thereby objectively detecting skin abnormalities caused by the occurrence of pressure ulcers by visualizing blood flow using the difference in the light absorption wavelengths of oxidized hemoglobin and hemoglobin, diagnosing pressure ulcers based on standardized data on the grade of pressure ulcers, and visualizing the predicted site where pressure ulcers may progress and the predicted items for preventing them, thereby providing a composite solution. Another purpose of the present invention is to provide a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence, which non-contactly diagnoses a pressure ulcer site using an image sequence captured using a crossed light source and a camera, and predicts the progression degree and future course of the pressure ulcer using a big data technique, and objectively detects skin abnormalities caused by the occurrence of pressure ulcers by visualizing blood flow, and diagnoses pressure ulcers based on standardized data on the grade of pressure ulcers, and provides a composite solution by visualizing the predicted site where pressure ulcers may progress and the predicted items for preventing them.

In addition, the present invention analyzes the shape of a pressure ulcer lesion, analyzes the shape and direction of pressure applied to the wound site, thereby predicting pressure ulcers that may occur in the future and projecting them onto the patient's affected area, and analyzes pressure factors that must be removed to prevent pressure ulcers and provides a guideline for changing patient position, thereby providing another purpose of the present invention is to provide a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence.

In order to achieve the above-mentioned purpose, a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to the features of the present invention is provided.

A non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence.

An imaging device that irradiates a body part with crossed light sources having different wavelengths and captures reflected light reflected from the body's epidermis to generate a sequence image;

A lesion progression analysis unit that analyzes the lesion progression for diagnosis and prediction of pressure ulcers based on non-contact data of sequence images generated from the above imaging device;

A display that visually displays the diagnosed and predicted pressure ulcer progression status and predicted situation through the above lesion progression analysis section; and

Its structural features include a database including a pressure ulcer diagnosis DB and a pressure physics engine DB to enable diagnosis based on standardized data on the grade of pressure ulcers.

Preferably, the imaging device comprises:

A light source unit that irradiates a body part with crossed light sources having different wavelengths to utilize the optical principle that oxidized hemoglobin and hemoglobin absorb different light wavelengths; and

The above light source may be configured to include a camera section that captures reflected light reflected from the body's epidermis by the crossed light source irradiated from the above light source section to generate a sequence image.

More preferably, the light source unit,

In order to utilize the optical principle that oxidized hemoglobin and hemoglobin absorb different light wavelengths, two types of light sources are configured to irradiate crossed light sources with different wavelengths to the body part, and the two types of light sources can be configured with multiple LEDs that irradiate light of 765 nm and light of 880 nm.

Even more preferably, the light source unit,

It is composed of a plurality of LEDs that irradiate 765 nm light and 880 nm light as two types of light sources, and can be configured with a structure in which two or more concentric circles are alternately arranged on an LED control body.

Even more preferably, the camera part,

It can be combined by being positioned in a hole formed at the center of the light source portion, which is alternately arranged in two or more concentric circles on the LED control body.

Preferably, the lesion progression analysis unit,

With the trained artificial intelligence model, it can function to diagnose pressure ulcers using big data-based deep learning on pressure ulcer progression, predict areas where pressure ulcers may progress, and provide guidelines that analyze pressure factors that need to be removed to prevent pressure ulcers.

More preferably, the lesion progression analysis unit,

It can function as a composite solution by measuring the distribution of hemoglobin in the blood using visual intelligence to diagnose the location of pressure ulcers on the skin, predicting changes according to the progression of pressure ulcers through deep learning, and then visualizing the predicted pressure ulcer progression status and predictions.

Even more preferably, the lesion progression analysis unit,

A feature detection unit that detects abnormal features of blood flow in the epidermis from non-contact data of sequence images generated by the imaging device;

A feature extraction unit that extracts features of abnormal areas of blood flow in the epidermis detected from the above feature detection unit;

A wound prediction unit that predicts the progression of a pressure ulcer based on the abnormal area features of the blood flow in the epidermis extracted through the above feature extraction unit; and

It can be configured to include an AI diagnosis engine (124) that diagnoses and determines bedsores based on predicted information on the progress of bedsores predicted through the above wound prediction unit and standardized data on the grade of bedsores learned and stored in advance in a database.

According to the non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence proposed in the present invention, the system comprises an imaging device which irradiates crossed light sources having different wavelengths to a body part, captures reflected light reflected from the body epidermis by the irradiated crossed light sources, and generates a sequence image; a lesion progression analysis unit which analyzes the lesion progression for diagnosing and predicting pressure ulcers based on non-contact data of the sequence image generated by the imaging device; a display which visually displays the diagnosed and predicted pressure ulcer progression status and predicted situation through the lesion progression analysis unit; and a database which includes a pressure ulcer diagnosis DB and a pressure physics engine DB to enable diagnosis based on standardized data on the grade of pressure ulcers, thereby enabling diagnosis of the progression status of pressure ulcers and prediction of their course by measuring the blood flow rate in the epidermis using an image sequence in which the epidermis is captured using crossed light sources.

In addition, according to the non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence of the present invention, a pressure ulcer site is diagnosed non-contactly using an image sequence captured using a crossed light source and a camera, and the progression degree and future course of the pressure ulcer can be predicted using a big data technique, thereby objectively detecting skin abnormalities caused by the occurrence of pressure ulcers by visualizing blood flow using the difference in light absorption wavelengths of oxidized hemoglobin and hemoglobin, diagnosing pressure ulcers based on standardized data on the grade of pressure ulcers, and visualizing the predicted site where pressure ulcers may progress and the predicted matters for preventing them, thereby providing a composite solution.

In addition, according to the non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence of the present invention, by analyzing the shape of the pressure ulcer lesion and analyzing the shape and direction of pressure applied to the wounded area, it is possible to predict pressure ulcers that may occur in the future and project them onto the patient's affected area, and analyze pressure factors that must be removed to prevent pressure ulcers so as to provide a guideline for changing body position.

FIG. 1 is a diagram illustrating the configuration of a non-contact data-based bedsore diagnosis and prediction system using visual intelligence according to one embodiment of the present invention as a functional block.
FIG. 2 is a drawing illustrating the configuration of an imaging device of a non-contact data-based bedsore diagnosis and prediction system using visual intelligence according to an embodiment of the present invention as a functional block.
FIG. 3 is a diagram illustrating the configuration of a lesion progression analysis unit of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to one embodiment of the present invention, as a functional block.
FIG. 4 is a diagram illustrating the configuration of a system schematic diagram for explaining the concept of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to one embodiment of the present invention.
FIG. 5 is a diagram schematically illustrating the configuration of an imaging device of a non-contact data-based bedsore diagnosis and prediction system using visual intelligence according to one embodiment of the present invention.
FIG. 6 is a diagram illustrating a configuration of a schematic diagram of a non-contact pressure ulcer measurement using visual intelligence of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to one embodiment of the present invention.
FIG. 7 is a diagram illustrating the expected progression of a pressure ulcer predicted based on the current condition of the pressure ulcer in a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to one embodiment of the present invention.

Hereinafter, with reference to the attached drawings, preferred embodiments will be described in detail so that those with ordinary skill in the art can easily practice the present invention. However, when describing preferred embodiments of the present invention in detail, if it is determined that a specific description of a related known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. In addition, the same reference numerals are used throughout the drawings for parts that perform similar functions and actions.

In addition, throughout the specification, when a part is said to be ‘connected’ to another part, this includes not only cases where it is ‘directly connected’ but also cases where it is ‘indirectly connected’ with other elements in between. Also, ‘including’ a component means that it may include other components, rather than excluding other components, unless otherwise specifically stated.

FIG. 1 is a diagram illustrating the configuration of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to an embodiment of the present invention as a functional block, FIG. 2 is a diagram illustrating the configuration of an imaging device of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to an embodiment of the present invention as a functional block, and FIG. 3 is a diagram illustrating the configuration of a lesion progression analysis unit of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to an embodiment of the present invention as a functional block. As illustrated in FIGS. 1 to 3, a non-contact data-based pressure ulcer diagnosis and prediction system (100) using visual intelligence according to an embodiment of the present invention may include an imaging device (110) that irradiates crossed light sources having different wavelengths to a body part and captures reflected light reflected from the body epidermis by the irradiated crossed light sources to generate a sequence image, a lesion progression analysis unit (120) that analyzes lesion progression for diagnosing and predicting pressure ulcers based on non-contact data of the sequence image generated by the imaging device (110), a display (130) that visually displays the diagnosed and predicted pressure ulcer progression status and predicted situation through the lesion progression analysis unit (120), and a database (140) having a pressure ulcer diagnosis DB and a pressure physics engine DB so that diagnosis based on standardized data on the grade of pressure ulcers is possible. Hereinafter, a specific configuration of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to an embodiment of the present invention will be described with reference to the attached drawings.

FIG. 4 is a diagram illustrating a configuration of a system schematic diagram for explaining the concept of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to an embodiment of the present invention, FIG. 5 is a diagram illustrating a schematic configuration of an imaging device of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to an embodiment of the present invention, FIG. 6 is a diagram illustrating a configuration of a schematic diagram of non-contact pressure ulcer measurement using visual intelligence of a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to an embodiment of the present invention, and FIG. 7 is a diagram illustrating an expected progression of a pressure ulcer predicted based on a current pressure ulcer state in a non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to an embodiment of the present invention.

The imaging device (110) is configured to irradiate a body part with crossed light sources having different wavelengths, and capture reflected light reflected from the body epidermis by the irradiated crossed light sources to generate a sequence image. As illustrated in FIGS. 2, 5, and 6, the imaging device (110) may include a light source unit (111) that irradiates a body part with crossed light sources having different wavelengths in order to utilize the optical principle that oxidized hemoglobin and hemoglobin each absorb different light wavelengths, and a camera unit (112) that captures reflected light reflected from the body epidermis by the crossed light sources irradiated from the light source unit (111) to generate a sequence image.

When bedsores occur, skin tissue necrosis occurs, which causes abnormalities in blood flow and oxygen saturation not only in the area where the ulcers occur but also in the surrounding area. In addition, since bedsores are a representative symptom of skin necrosis due to pressure, in order to prevent recurrence, not only the area where the ulcers occur should be treated, but also the pressure applied should be fundamentally removed. According to an embodiment of the present invention, a bedsore diagnosis and prediction system (100) measures blood flow in the epidermis using a cross-light source, detects abnormalities in blood flow, and diagnoses the progression of bedsores and predicts their course.

In addition, the light source unit (111) is configured with two types of light sources for irradiating cross light sources with different wavelengths to the body part in order to utilize the optical principle that oxidized hemoglobin and hemoglobin each absorb different light wavelengths. The two types of light sources may be configured with multiple LEDs that irradiate 765 nm light and 880 nm light.

In addition, the light source unit (111) is composed of a plurality of LEDs that irradiate 765 nm light and 880 nm light as two types of light sources, and may be configured with a structure in which two or more concentric circles are alternately arranged on the LED control body, as shown in FIGS. 5 and 6, respectively.

In addition, the camera unit (112) can be positioned and combined in a hole formed at the center of two or more light source units (111) alternately arranged in concentric circles on the LED control body.

In this way, the imaging device (110) is configured to include a light source unit (111) and a camera unit (112), and LEDs of different wavelengths of the light source unit (111) are arranged on concentric circles of the LED control body, a camera lens hole is formed at the center of the LED control body, and the camera lens hole can be configured as a hole having a size of 32 mm or more. That is, as shown in FIGS. 5 and 6, the LED control body can be configured to have two types of cross light source LEDs arranged alternately on two concentric circles with different diameters so that two types of light can be uniformly irradiated to a body part. In addition, a hole is formed at the center of the concentric circles where the LEDs of the two cross light sources are arranged so that the lens of the camera unit (112) is positioned, and the lens of the camera unit (112) is inserted into the hole at the center so that reflected light can be measured without being biased to one side. In addition, the lens of the camera unit (112) is positioned in the hole formed in the LED control body so that it can be combined to enable photography of body parts, but can be configured to be detachable so that it can be detached.

In addition, the camera unit (112) can measure the reflected light reflected from the epidermis by the cross light source irradiated from the light source unit (111). In addition to a general visible light camera, the camera unit (112) can be implemented with various types of cameras such as a thermal imaging camera, a depth camera, and an infrared camera. In addition, the camera unit (112) can be easily held in hand and can diagnose bedsores of patients with limited movement by irradiating light on a body part of a bedsore patient and measuring the reflected light.

The lesion progression analysis unit (120) is a configuration that analyzes the lesion progression for diagnosing and predicting pressure ulcers based on non-contact data of sequence images generated from the imaging device (110). This lesion progression analysis unit (120) can function to provide a guideline that analyzes the diagnosis of pressure ulcers, predicted areas where pressure ulcers may progress, and pressure factors that must be removed to prevent pressure ulcers using deep learning based on big data on pressure ulcer progression as an artificial intelligence model that has completed learning. Here, the lesion progression analysis unit (120) can function to measure the blood flow of the body's epidermis using the optical principle that oxidized hemoglobin and hemoglobin each absorb different light wavelengths, detect abnormal areas of the epidermis through the measured blood flow, and predict and visualize the progress using a statistical method and ensemble learning, thereby providing a treatment solution suitable for the patient.

In addition, the lesion progression analysis unit (120) can function to measure the distribution of hemoglobin in the blood using visual intelligence to diagnose the site of pressure ulcer occurrence on the skin, predict changes according to the progression of pressure ulcers through deep learning, and then visualize the predicted pressure ulcer progression status and predictions to provide a composite solution.

In addition, the lesion progression analysis unit (120) may be configured to include a feature detection unit (121) that detects abnormal features of intradermal blood flow from non-contact data of sequence images generated from an imaging device (110), a feature extraction unit (122) that extracts abnormal features of intradermal blood flow detected by the feature detection unit (121), a wound prediction unit (123) that predicts the progression status of a pressure ulcer based on the abnormal features of intradermal blood flow extracted by the feature extraction unit (122), and an AI diagnosis engine (124) that diagnoses and determines a pressure ulcer based on prediction information of the progression status of the pressure ulcer predicted by the wound prediction unit (123) and standardized data on the grade of the pressure ulcer that is learned in advance and stored in a database (140).

The display (130) is configured to visually display the diagnosed and predicted bedsore progression status and predicted situation through the lesion progression analysis unit (120). As illustrated in FIG. 7, this display (130) can measure the distribution amount of hemoglobin in the blood using visual intelligence to diagnose the site of bedsore occurrence on the skin, predict changes according to bedsore progression using deep learning technology, and provide the predicted bedsore progression status and predicted items as visuals to the user. That is, FIG. 7 sequentially displays the expected progression status of bedsores predicted based on the current bedsore status.

The database (140) is configured to have a bedsore diagnosis DB and a pressure physics engine DB to enable diagnosis based on standardized data on the grade of bedsores. This database (140) includes a bedsore diagnosis database for resolving diagnosis based on standardized data on the grade of bedsores, and a database of pressure physics status information according to bedsores. Here, pre-learned bedsore diagnosis data and pressure physics data of the database (140) can be used as standardized data on the grade of bedsores.

In this way, a non-contact data-based pressure ulcer diagnosis and prediction system (100) using visual intelligence, which includes an imaging device (110) that irradiates a body part with crossed light sources having different wavelengths, captures reflected light reflected from the body epidermis by the irradiated crossed light sources, and generates a sequence image, a lesion progression analysis unit (120) that analyzes the lesion progression for diagnosing and predicting pressure ulcers based on non-contact data of the sequence image generated by the imaging device (110), a display (130) that visually displays the diagnosed and predicted pressure ulcer progression status and predicted situation through the lesion progression analysis unit (120), and a database (140) having a pressure ulcer diagnosis DB and a pressure physics engine DB to enable diagnosis based on standardized data on the grade of pressure ulcers, can diagnose a pressure ulcer site based on non-contact data of visual intelligence using image sequences captured on the epidermis using crossed light sources, predicts the degree of progression and future course of pressure ulcers using big data techniques, and visualizes the predicted items of pressure ulcers to provide a composite solution. there is.

As described above, the non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence according to an embodiment of the present invention comprises an imaging device that irradiates crossed light sources having different wavelengths to a body part and captures reflected light reflected from the body epidermis by the irradiated crossed light sources to generate a sequence image, a lesion progression analysis unit that analyzes the lesion progression for diagnosing and predicting pressure ulcers based on non-contact data of the sequence image generated by the imaging device, a display that visually displays the diagnosed and predicted pressure ulcer progression status and predicted situation through the lesion progression analysis unit, and a database having a pressure ulcer diagnosis DB and a pressure physics engine DB that enables diagnosis based on standardized data on the grade of pressure ulcers, thereby enabling diagnosis of the progression status of pressure ulcers and prediction of the progress by measuring the blood flow in the epidermis using an image sequence captured using crossed light sources, and also enables diagnosis of pressure ulcer sites in a non-contact manner using an image sequence captured using crossed light sources and a camera, and prediction of the degree of progression and future progress of pressure ulcers using big data techniques. By doing so, it is possible to objectively detect skin abnormalities caused by bedsores by visualizing blood flow using the difference in the optical absorption wavelength of oxidized hemoglobin and hemoglobin, diagnose bedsores based on standardized data on the grade of bedsores, and provide a composite solution by visualizing the predicted areas where bedsores may progress and the predicted measures to prevent them. In particular, by analyzing the shape of the bedsore lesion and analyzing the shape and direction of the pressure applied to the wound area, it is possible to predict possible bedsores in the future and project them onto the patient's affected area, and analyze the pressure factors that need to be removed to prevent bedsores so that a guideline for changing the patient's position can be provided.

The present invention described above can be variously modified or applied by those skilled in the art, and the scope of the technical idea according to the present invention should be defined by the following patent claims.

100: Bedsore diagnosis and prediction system according to one embodiment of the present invention
110: Imaging Device
111: Light source
112: Camera Department
120: Lesion progression analysis unit
121: Feature detection unit
122: Feature Extraction Unit
123: Wound Prediction Department
124: AI Diagnostic Engine
130: Display
140: Database (DB)

Claims (8)

A non-contact data-based pressure ulcer diagnosis and prediction system (100) using visual intelligence,
An imaging device (110) that irradiates crossed light sources with different wavelengths to a body part and captures reflected light reflected from the body epidermis from the irradiated crossed light sources to generate a sequence image;
A lesion progression analysis unit (120) that analyzes the lesion progression for diagnosis and prediction of pressure ulcers based on non-contact data of sequence images generated from the above imaging device (110);
A display (130) that visually displays the diagnosis and predicted bedsore progression status and predicted situation through the above lesion progression analysis unit (120); and
A database (140) including a pressure ulcer diagnosis DB and a pressure physics engine DB to enable diagnosis based on standardized data on the grade of pressure ulcers,
The above imaging device (110) is
In order to utilize the optical principle that oxidized hemoglobin and hemoglobin absorb different light wavelengths, it is configured to include a light source unit (111) that irradiates crossed light sources with different wavelengths to a body part, and a camera unit (112) that captures reflected light reflected from the body epidermis by the crossed light sources irradiated from the light source unit (111) to generate a sequence image.
The above light source (111) is
In order to utilize the optical principle that oxidized hemoglobin and hemoglobin absorb different light wavelengths, it is composed of two types of light sources to irradiate cross light sources with different wavelengths to the body parts, and the two types of light sources are composed of multiple LEDs that irradiate light of 765 nm and light of 880 nm.
The above light source (111) is
It is composed of multiple LEDs that irradiate 765 nm light and 880 nm light as two types of light sources, and is structured so that two or more concentric circles are alternately arranged on the LED control body.
The above camera part (112)
It is positioned and connected in a hole formed at the center of the light source portion (111) that is alternately arranged in two or more concentric circles on the LED control body,
The above lesion progression analysis unit (120)
It comprises a feature detection unit (121) for detecting abnormal features of blood flow in the epidermis from non-contact data of a sequence image generated by the imaging device (110), a feature extraction unit (122) for extracting features of abnormal features of blood flow in the epidermis detected by the feature detection unit (121), a wound prediction unit (123) for predicting the progression of a pressure ulcer based on the features of abnormal features of blood flow in the epidermis extracted by the feature extraction unit (122), and an AI diagnosis engine (124) for diagnosing and judging a pressure ulcer based on prediction information of the progression of the pressure ulcer predicted by the wound prediction unit (123) and standardized data on the grade of the pressure ulcer that is learned and stored in advance in a database (140).
The above display (130) is
A non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence, characterized in that the above-mentioned lesion progression analysis unit (120) measures the amount of hemoglobin distribution in the blood using visual intelligence to diagnose the site of pressure ulcer occurrence on the skin, and provides the user with a visual representation of the predicted pressure ulcer progression status and predictions using deep learning technology according to the changes in the progression of the pressure ulcer, while sequentially displaying the expected progression status of the pressure ulcer predicted based on the current pressure ulcer status.
delete delete delete delete In the first paragraph, the lesion progression analysis unit (120)
A non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence, characterized by using a trained artificial intelligence model to diagnose pressure ulcers using big data-based deep learning on pressure ulcer progression, and to provide guidelines that analyze predicted areas where pressure ulcers may progress and pressure factors that need to be removed to prevent pressure ulcers.
In the 6th paragraph, the lesion progression analysis unit (120)
A non-contact data-based pressure ulcer diagnosis and prediction system using visual intelligence, characterized by measuring the distribution of hemoglobin in the blood using visual intelligence to diagnose the site of pressure ulcer occurrence on the skin, predicting changes according to the progression of pressure ulcers through deep learning, and then visualizing the predicted pressure ulcer progression status and predictions to provide a composite solution.
delete