KR102543555B1 - Ai breast cancer diagnosis apparatus, and breast cancer self-diagnosis method using the same - Google Patents

Abstract

The present invention relates to an apparatus for diagnosing breast cancer by oneself without the assistance of doctors by receiving the assistance of AI. The breast cancer diagnosis apparatus and a self-diagnosis method using the same can remove a measurement deviation on a thermal image according to a surrounding temperature or a body excitatory state of a patient by specifying a location of a lesion while gradually decreasing a breast surface temperature of the patient, induce a right posture required for the measurement through augmented reality that delivers a feedback control command with an object image to the patient for breast cancer self-diagnosis through a display panel of a smart mirror, and allow the patient to self-diagnose breast cancer by an AI neural network trained by the thermal image.

Description

Artificial intelligence breast cancer diagnosis device and breast cancer self-diagnosis method using the same

The present invention is a device for self-diagnosing breast cancer by oneself without the aid of a doctor with the help of artificial intelligence, and more specifically, by using a smart mirror equipped with a thermal imaging camera, a patient is provided with the necessary information for measurement by augmented reality. Thermal imaging while changing the temperature of the patient's breast by the breast temperature control device to induce the correct posture, display the measurement result, and remove the patient's breast temperature fluctuation component according to the ambient temperature or the patient's body excitement state The present invention relates to an artificial intelligence breast cancer diagnosis device that allows a patient to self-diagnose the risk of breast cancer by obtaining an image and analyzing it by a pre-learned artificial intelligence neural network, and a breast cancer self-diagnosis method using the same.

Recently, along with medical equipment manufacturing technology, digital image processing technology has been applied to the field of clinical diagnosis, and many advances in radiology have been made.

In particular, ultrasound diagnosis is harmless to the human body as it can avoid harmful radiation exposure compared to CT or X-ray medical equipment. It is characterized by low cost and low cost. In particular, since images can be obtained in real time, there is an advantage in that the movement state of organs can be observed in real time.

This ultrasound diagnostic technology is widely used in breast cancer screening along with mammography using X-rays.

However, in breast ultrasound diagnosis, when the scanning posture of an ultrasound scanner is not correct, it is impossible to obtain a high-quality ultrasound image due to low resolution of the ultrasound image and severe noise. Moreover, since the ultrasound scanner needs to vary the contact pressure and inclination for each affected area, it is virtually difficult for non-skilled experts to handle the ultrasound scanner, making it difficult to perform self-examination at home.

Moreover, breast ultrasound images alone are inaccurate in determining breast cancer, so breast cancer is diagnosed in parallel with mammography.

On the other hand, scientists have found that the skin temperature of the affected area rises in case of breast cancer, but the breast cancer examination using the thermal imaging method based on skin temperature measurement is affected by the ambient temperature and the temperature deviation according to the patient's body's thermal excitement. It can involve many false positives and measurement deviations, making it difficult to apply as a standard test method. However, thermal imaging is still attracting attention from many scientists, medical device industry and patients because it measures in a non-contact way.

Breast cancer is a disease with a high cure rate if detected early, and as the importance of self-diagnosis for breast cancer has been highlighted worldwide, active research is being conducted on a diagnosis method using a thermal imaging camera. However, self-diagnosis using a thermal imaging camera is There are many problems in securing its reliability and accuracy due to the large measurement deviation according to the ambient temperature and the thermal excitability of the patient's body.
(Patent Document 1) KR 10-1936059 B1

In the prior art, there are many fluctuations in the patient's body temperature depending on the ambient temperature or the patient's body excitement state, resulting in many difficulties and errors in accurate breast cancer examination using thermal images. For example, if you have breast cancer, the skin temperature of the relevant lesion area rises, but in summer or in a state of physical excitement, this lesion area is not clearly revealed on the thermal image, or the skin temperature itself becomes hot, so it is mistakenly diagnosed as breast cancer. There are concerns.

The present invention is to solve the problems of the prior art described above, and to increase the accuracy of diagnosis and self-diagnosis by using thermal images obtained while changing the temperature of the patient's breast by a breast temperature control device for breast cancer examination It is an object of the present invention to provide an artificial intelligence breast cancer diagnosis device capable of diagnosing breast cancer.

For example, while the patient normally looks at the smart mirror, a thermal image is obtained by a thermal imaging camera installed on the smart mirror, and the breast temperature control device analyzes the thermal image to determine the ambient temperature according to the ambient temperature or the patient's body excitement state. is gradually lowered, and in the meantime, the thermal imaging camera obtains thermal images of the breast area at each temperature level, and analyzes them by the pre-learned thermal image artificial intelligence neural network to determine the patient's breast cancer symptoms, and if breast cancer is suspected , Smart Mirror is to provide an artificial intelligence breast cancer diagnosis device capable of self-diagnosis of breast cancer by an artificial intelligence neural network that recommends remote diagnosis with a specialist doctor to a patient through an internet network and a breast cancer self-diagnosis method using the same The purpose.

That is, when it is determined that the ambient temperature is too hot or the patient is in an excited state, the breast temperature control device can lower the surface temperature of the patient's breast while lowering the ambient temperature step by step, and accordingly, the breast lesion ( Infrared radiation (thermal radiation) emitted from the Lesion area is clearly exposed on the thermal image, enabling more precise breast cancer diagnosis.

The artificial intelligence breast cancer diagnosis device of the present invention not only enables patients to frequently perform breast cancer examinations comfortably at home without exposure to radiation, but also significantly improves the reliability of breast cancer examinations by securing big data on patients according to the increase in the number of self-examinations. can

However, the technical problem to be achieved by the embodiments of the present application is not limited to the technical problems described above, and other technical problems may exist.

As a technical means for achieving the above technical problem, the first aspect of the present application is a smart mirror; a thermal imaging camera for obtaining a thermal image by detecting thermal radiation emitted from the patient's body; and a breast temperature control device for varying the breast surface temperature to minimize a measurement deviation of the thermal image of the breast region according to the ambient temperature and the patient's body arousal state. It provides an artificial intelligence breast cancer diagnosis device comprising a.

According to one embodiment of the present invention, the breast temperature control device may include: a breast temperature distribution calculator that analyzes a thermal image of a breast region and calculates an ambient temperature or an excited state of the patient's body; a cooling fan for providing cool air to the patient's breast; and a remote control transmitter configured to control the temperature of the cooling fan step by step according to the calculation result of the breast temperature distribution calculator to adjust the surface temperature of the breast. It may include, but is not limited thereto.

According to one embodiment of the present application, the smart mirror further includes a thermal breast cancer detection unit, and the thermal breast cancer detection unit accumulates pixel-by-pixel thermal images of the breast region captured a predetermined number of times during a predetermined period of time. a breast thermal image mapper for generating a breast thermal image composed of average values of sums; a cutoff adjustment unit for generating a thermal hot spot image consisting of breast cancer hot spots exhibiting a temperature value equal to or greater than a predetermined value in the breast thermal image; and a breast cancer hot spot memory for storing the thermal hot spot image or the breast cancer hot spot information. It may include, but is not limited thereto.

According to one embodiment of the present application, the smart mirror further includes a thermal imaging breast cancer detection unit, and the thermal imaging breast cancer detection unit includes a thermal image of a patient's breast region obtained while lowering the ambient temperature step by step by the breast temperature control device. a breast thermal image mapper for generating a breast thermal image as an average value after taking pixel-by-pixel additions for the images; a cutoff adjustment unit for generating a thermal hot spot image consisting of breast cancer hot spots exhibiting a temperature value equal to or greater than a predetermined value in the breast thermal image; and a breast cancer hot spot memory for storing the thermal hot spot image or the breast cancer hot spot information. It may include, but is not limited thereto.

According to one embodiment of the present application, the smart mirror may additionally include a hot spot guider that informs the breast cancer hot spot area and location in augmented reality by overlapping a mirror on which the patient's image is projected through a virtual object image. , but is not limited thereto.

According to one embodiment of the present application, the smart mirror further includes a body posture interaction unit, and the body posture interaction unit includes: a body navigator that creates a body map including a body outline and a boundary line distinguishing body organs from a patient's body image; a body posture correction request unit for displaying a body outline image for posture fitting on the smart mirror through a virtual object image to provide a body posture to be taken by a patient during breast cancer diagnosis; a body fitting check unit that calculates a degree of fitting between the patient's body outline image and the body outline image for posture fitting and gives feedback to the patient; and a patient authentication unit that confirms who the patient is. It may include, but is not limited thereto.

According to one embodiment of the present application, the smart mirror may further include a standing position information providing means for calculating optimal position information advantageous to breast cancer diagnosis and providing the information to the patient, but is not limited thereto.

According to one embodiment of the present application, the smart mirror observes the analysis result of the thermal image artificial intelligence neural network according to the periodic follow-up examination by the thermal imaging camera, the size and number of breast cancer hot spots, and informs the patient of the degree of risk Alternatively, a breast cancer tracking management unit notifying the patient of the next breast cancer inspection schedule may be included, but is not limited thereto.

In addition, the second aspect of the present application is a breast cancer self-diagnosis method performed by the artificial intelligence breast cancer diagnosis device according to the first aspect of the present application, wherein the breast cancer self-diagnosis method is an optimal breast cancer diagnosis advantageous to breast cancer diagnosis by means of providing standing position information providing location information of the patient; obtaining a thermal image of the breast while gradually lowering the temperature of the surface of the breast by a breast temperature control device; finding a breast cancer hot spot from a breast thermal image; If a breast cancer hot spot is found, requesting a patient to self-examine the breast cancer hot spot area; During self-examination, superimposing breast cancer hot spots discovered on the patient's image reflected in the mirror with augmented reality and displaying them on the smart mirror; and observing changes in the size of breast cancer hot spots and the number of breast cancer hot spots as a result of the analysis of the thermal image artificial intelligence neural network, and notifying the patient of a breast cancer risk or a schedule for the next breast cancer examination; It provides a breast cancer self-diagnosis method comprising a.

The above-described problem solving means are merely exemplary and should not be construed as intended to limit the present disclosure. In addition to the exemplary embodiments described above, additional embodiments may exist in the drawings and detailed description of the invention.

According to the above-described problem solving means of the present application, the measurement deviation of the breast thermal image according to the ambient temperature or the patient's body excitement state is removed by the breast temperature control device,

Artificial intelligence breast cancer that allows patients to self-diagnose breast cancer by inducing the correct posture required for measurement to patients self-diagnosing breast cancer through augmented reality displayed on the smart mirror, and by artificial intelligence neural networks pre-learned by breast thermal images A diagnostic device and a self-examination method using the same are provided.

However, the effects obtainable herein are not limited to the effects described above, and other effects may exist.

1 is a schematic diagram of a breast cancer diagnosis device according to an embodiment of the present application.
2 is an operating principle of a breast cancer diagnosis apparatus according to an embodiment of the present application.
3A and 3B are examples of a body navigator for obtaining a body map from a body image in the apparatus for diagnosing breast cancer according to one embodiment of the present invention.
Figure 4a is an implementation of a standing position information providing unit for providing standing position information on an optimal standing position advantageous for breast cancer diagnosis to a patient when the patient stands in front of a smart mirror using a breast cancer diagnosis apparatus according to an embodiment of the present invention Yes.
4B is an embodiment of a means for providing standing position information having a means for determining the position of a patient's foot print in the apparatus for diagnosing breast cancer according to an embodiment of the present invention.
5A and 5B are embodiments in which a self-examination is performed by overlapping a virtual object image required for breast cancer diagnosis on a display panel on a patient's own image reflected in the mirror using a smart mirror.

Hereinafter, embodiments of the present application will be described in detail so that those skilled in the art can easily practice with reference to the accompanying drawings.

However, the present disclosure may be implemented in many different forms and is not limited to the embodiments described herein. And in order to clearly describe the present application in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the present specification, when a part is said to be “connected” to another part, it is not only “directly connected”, but also “electrically connected” or “indirectly connected” with another element in between. "Including cases where

Throughout the present specification, when a member is referred to as being “on,” “above,” “on top of,” “below,” “below,” or “below” another member, this means that a member is located in relation to another member. This includes not only the case of contact but also the case of another member between the two members.

Throughout the present specification, when a certain component is said to "include", it means that it may further include other components without excluding other components unless otherwise stated.

As used herein, the terms "about," "substantially," and the like are used at or approximating that number when manufacturing and material tolerances inherent in the stated meaning are given, and are intended to assist in the understanding of this disclosure. Accurate or absolute figures are used to prevent undue exploitation by unscrupulous infringers of the stated disclosure. In addition, throughout the present specification, “steps of” or “steps of” do not mean “steps for”.

Throughout the present specification, the term "combination thereof" included in the expression of the Markush form means one or more mixtures or combinations selected from the group consisting of the components described in the expression of the Markush form, and the components It means including one or more selected from the group consisting of.

Throughout this specification, reference to "A and/or B" means "A or B, or A and B".

Throughout this specification, patient is used interchangeably with self-diagnosis.

Throughout the present specification, the body outline image for posture fitting is used interchangeably with the image for posture fitting.

Throughout this specification, the thermal imaging camera mode refers to a period in which the thermal imaging camera operates to obtain a corresponding thermal image from the patient's breast.

Throughout the present specification, the self-examination mode refers to a period in which a patient examines his/her own breast while looking at the smart mirror and touching the breast by a breast self-examination method with the help of virtual object images and audio information provided from the smart mirror.

Throughout the present specification, the breast self-examination method includes the patient's own lumps and pains, lumps, nipple secretions, nipple depression, breast wrinkles, nipple eczema, changes in breast skin, changes in breast size, and changes in position of the nipples. It includes the process of checking the self-inspection items one by one.

The breast self-examination method may include a process of answering a question through a digital Automatic Response Service (ARS) provided by a smart mirror.

For example, a digital ARS can ask a patient if they have nipple discharge, and the patient can answer "yes" or "no".

Q&A of the digital ARS of the present invention can be performed by a voice message, a Q&A sentence provided as an object image on a display panel, and an answer selection click window by touch.

Throughout this specification, self-diagnosis includes a thermal imaging camera mode and a self-exam mode.

Hereinafter, an artificial intelligence breast cancer diagnosis device according to the present invention and a method for self-diagnosing breast cancer using the same will be described in detail with reference to embodiments, examples, and drawings. However, the present application is not limited to these embodiments and examples and drawings.

As a technical means for achieving the above technical problem, the first aspect of the present application is a smart mirror; a thermal imaging camera for obtaining a thermal image by detecting thermal radiation emitted from the patient's body; and a breast temperature control device for varying the breast surface temperature to minimize a measurement deviation of the thermal image of the breast region according to the ambient temperature and the patient's body arousal state. It provides an artificial intelligence breast cancer diagnosis device comprising a.

The smart mirror in the present invention is a display device in which a mirror and a touch display panel are combined, and is manufactured in the form of attaching a mirror film to a touch display panel. A smart mirror has the shape of a mirror, but displays an image of an object that provides various functions such as time, weather, date, medical guide service for self-diagnosis, etc. It is a mirror to pursue the convenience of self-diagnosis.

In addition to these functions, the smart mirror, for use in the beauty industry, when you select a color you want, superimposes on yourself to show your hair and lips dyed in that color, or to show you what you want before buying clothes It is possible to provide an Augmented Reality function that shows fashion superimposed on one's own appearance.

The present invention has been made to solve the problems of the prior art, and the breast cancer diagnosis device of the present invention includes a smart mirror, and the smart mirror detects thermal radiation emitted from a patient's body to generate a two-dimensional thermal image. A thermal imaging camera for obtaining a body image of a patient, a plurality of image sensors for obtaining a body image of a patient, a speaker and a display panel for transmitting feedback control commands to the patient or providing information and instructions necessary for breast cancer diagnosis to the patient, and a breast cancer image from the thermal image. A breast thermal image mapper for obtaining a breast thermal image showing a temperature distribution of the breast, and a thermal image artificial intelligence neural network trained in advance by deep learning using breast cancer thermal images for learning. By analyzing the thermal image, it is characterized in that the risk of breast cancer of the patient is automatically determined.

The wireless transmission/reception connection of the present invention is preferably made by Wi-Fi, Bluetooth or Internet of Things connection.

The display panel of the smart mirror of the present invention is characterized by showing the progress of the patient performing self-diagnosis by overlapping a virtual object image necessary for breast cancer diagnosis on the patient's own image reflected in the mirror .

Preferably, the breast thermal image mapper generates a breast thermal image composed of an average value of an accumulated sum of pixels of thermal images of the breast area captured a predetermined number of times during a predetermined period.

By using the average value of the thermal image of the breast region acquired over a predetermined number of times, the reliability of the breast cancer examination is much better than the result of the thermal image obtained only by the examination once a year.

Hereinafter, an image composed of image pixels (breast cancer hot spots) exhibiting a temperature value higher than a predetermined value in a breast thermal image is referred to as a thermal image hot spot image, and an area of corresponding pixels is referred to as a breast cancer hot spot area.

The breast cancer hot spot area obtained in the thermal imaging camera mode refers to an area with a higher temperature than other areas, and this area corresponds to a breast cancer suspected area.

Therefore, the apparatus for diagnosing breast cancer according to the present invention provides a means to increase the reliability of breast cancer examination and to detect breast cancer at an early stage by re-examining the breast cancer hot spot area obtained by the thermal imaging camera mode through the self-examination mode. .

In addition, the smart mirror is provided with a hot spot guider that informs the breast cancer hot spot area and location obtained in the thermal imaging camera mode during the self-examination mode by overlapping it with the mirror on which the patient's own image is projected through an object image. It is characterized by allowing interactive diagnosis with the patient.

For example, during the self-examination mode, the patient can check for lumps or lumps while looking in a mirror at his/her own breast area corresponding to the breast cancer hotspot area indicated by the hot spot guider and touching it by himself through the breast self-examination method.

In addition, the smart mirror is a body navigator that creates a body map including body outlines and boundary lines that distinguish the patient's face, breast, arm, abdomen, leg, foot, and other body parts on the patient's body image. (body navigator) may be further provided. Accordingly, it is possible to obtain a breast outline that distinguishes the breast from other body organs on the body map obtained by the body navigator.

The body map is deep learning by semantic segmentation labeled body images with different colors for face, breast, arm, belly, leg, foot and other body parts. can be obtained by artificial intelligence neural networks.

From the semantic segmented body image, boundary lines and body outlines for distinguishing body organs can be obtained. For example, the boundary between the semantic segmented belly and breast provides a breast boundary line (breast outline).

The semantic segmentation is an artificial intelligence neural network that classifies in units of pixels where a specific object is included in a given "image" and divides it from other objects.

Another aspect of the body navigator of the present invention is based on body images marked with body outline images including boundary lines dividing body organs including face, breast, arm, abdomen, leg, and foot. With a trained artificial intelligence neural network, a body map may be obtained using the artificial intelligence neural network for a patient's body image given from an image sensor.

Another aspect of the body navigator of the present invention is a body map in which boundary lines dividing major body organs are arranged (included) in consideration of medical and physical arrangement correlations on a body outline image characterized in that it is implemented by

The body outline image may be obtained by obtaining a differential image between an image when the patient is not in front of the smart mirror and an image when the patient is in front of the smart mirror and taking an edge component thereof.

In this case, the body map is the body outline image based on the medical and physical arrangement situation between the body organs (face, arm, breast, abdomen, leg, foot, etc.) and the statistic body organ arrangement method for the size ratio between the body organs. It can be obtained by placing the boundary lines dividing the body organs on the image.

The statistic body organ arrangement method selects the location of body organs according to medical statistical data based on the patient's body information including the patient's sex, race, age, height, weight, and waist circumference entered during patient registration It is preferred to do

The patient authentication is preferably performed by any one method selected from face recognition, fingerprint recognition, voice recognition, and ID authentication methods registered at the time of patient registration.

The smart mirror of the present invention may further include a face recognition unit for recognizing a corresponding face by comparing an image of a face on the body map with a face database of a patient registered in advance.

The smart mirror is characterized by having a body posture correction request unit that provides the patient with a body posture to be taken by the patient through an object image so as to facilitate diagnosis of breast cancer of the patient.

For example, the body posture correction requesting unit displays an image of an object by using the image of the patient with both arms raised as a body outline image for posture fitting in order to guide the patient to raise both arms when it is necessary to check the patient's armpits. It is preferable to calculate the degree of fitting between the body outline image and the body outline image for posture fitting and to give feedback to the patient.

The body outline image for posture fitting of the body posture correction requesting unit is preferably a body outline image in a standing posture while the patient is looking straight ahead toward the smart mirror.

The degree of fitting between the patient's body outline image and the body outline image for posture fitting is preferably visually fed back to the patient through an object image means.

The smart mirror is characterized in that it further includes a standing position information providing unit for calculating and providing optimal standing position information advantageous to breast cancer diagnosis during the thermal imaging camera mode to the patient.

For breast cancer examination with a thermal imaging camera, it is necessary for the patient to stand in an optimal position considering the field of view (FOV) and focal length of the thermal imaging camera.

The optimal standing position information refers to an optimal place for a patient to stand, that is, a standing place, in which a patient should stand in front of a smart mirror for breast cancer examination using a thermal imaging camera.

The standing position information providing means measures the height of the patient by an image sensor,

It is preferred to display a body outline image for a standing position or posture fitting advantageous for breast cancer diagnosis through a virtual object image on a display panel.

Another aspect of the standing position information providing means is to mark a standing place on the ground as a footprint mark by a laser beam scanning means.

For example, the laser beam scanning means may provide a laser footprint pattern in the shape of a footprint on the floor of a standing place.

The standing position information providing means of the present invention preferably provides the patient with a standing position in the same position as used in the thermal imaging camera mode during the self-examination mode.

In addition, the standing position information providing unit is characterized in that it further includes a patient footprint positioning unit for checking whether the sole of the patient's foot is within the range of the laser footprint pattern by the image sensor.

For example, the patient's footprint positioning unit finds a foot on a body map, checks the presence or absence of a laser footprint pattern in the corresponding area and the degree of fitting with the foot to determine how well the patient fits and aligns the foot in the standing position. can

In addition, it is preferable to provide audio or visual guidance to the patient to fit the patient's foot within the range of the laser footprint pattern by means of the speaker and display panel of the smart mirror.

Also, it is preferable that the laser footprint pattern blinks when the patient's foot stays outside the range of the laser footprint pattern.

The display panel of the smart mirror of the present invention shows the progress of the patient performing self-diagnosis by augmented reality that overlaps the image of a virtual object necessary for breast cancer diagnosis on the patient's image reflected in the mirror. to be

The virtual object image according to the embodiment of the present invention is preferably one or more object images selected from among a virtual body organ, a footprint outline for fitting, and a breast cancer hot spot.

The virtual body organs may include a breast outline and a body outline of the patient.

The breast cancer hot spot area and location obtained during the thermal imaging camera mode are calculated based on the breast outline, and it is preferable to store the calculated breast cancer hot spot area and location in a breast cancer hot spot memory.

During the self-examination mode, the hot spot guider superimposes the breast cancer hot spot on the patient's breast image reflected in the mirror in augmented reality and displays it on the display panel, so that the patient can intensively re-examine the breast cancer hot spot area. can induce

For example, during the self-examination mode, the hot spot guider superimposes a breast cancer hot spot obtained during the thermal imaging camera mode with a mirror image of the patient's breast and displays it on the display panel in augmented reality, so that the patient can view the breast cancer hot spot in the breast self-examination method. may lead to re-examination.

To this end, the hot spot guider aligns the breast cancer hot spot area and position obtained in the thermal imaging camera mode based on the breast outline during the self-examination mode, superimposes it on the patient's breast image reflected in the mirror through the object image, and displays it in augmented reality. characterized by

In the present invention, it is preferred that the breast cancer thermal images for learning for the thermal image artificial intelligence neural network are composed of labeled thermal images for each grade of breast cancer. For example, breast cancer thermal images for learning can be labeled by dividing them into normal, mild, moderate, severe, or benign, malignant, and normal grades for each grade of breast cancer, and the trained thermal image artificial intelligence neural network can be labeled from the thermal image camera. It is characterized in that the breast cancer risk is calculated by analyzing the obtained patient's breast thermal image and dividing it into grades.

In another embodiment of the breast cancer thermal images for training in the present invention, a breast row for training in which tumors, masses, and micro-calcification clusters are labeled with semantic segmentation in different colors After pre-supervised learning by image images, semantic segmentation is performed on the patient's breast thermal image obtained from the thermal imaging camera to obtain a semantically segmented thermal image for the tumor, mass and microcalcification cluster ;

It is characterized in that the risk of breast cancer is calculated according to the degree of a tumor, mass or calcification cluster found in the semantic segmented thermal image.

In the present invention, the breast cancer thermal images used in the thermal image artificial intelligence neural network are the average value of the cumulative sum for each pixel of the thermal images of the breast region captured more than once during a predetermined period by the breast thermal image mapper. It may be a breast thermal image composed of .

In addition, while the breast cancer thermal images used in the thermal image artificial intelligence neural network in the present invention are lowered in stages by the breast temperature controller, the breast thermal image mapper at each temperature stage for a predetermined period. It may be a breast thermal image composed of an average value of a pixel-by-pixel cumulative sum of thermal images of the breast region captured one or more times.

In addition, the smart mirror performs a periodic thermal image tracking test, and informs the patient of the risk of breast cancer according to the analysis result of the thermal image artificial intelligence neural network, the size of breast cancer hot spots, and the change in the number of breast cancer hot spots, or the next breast cancer thermal imaging camera examination. It is preferred to further include a breast cancer tracking management unit informing the patient of the schedule.

The artificial intelligence neural network of the present invention preferably uses semantic segmentation, a convolutional neural network (CNN) or a recurrent neural network (RNN).

In the present invention, the artificial neural network is a neural network that allows deep learning learning, and includes a convolution layer, a pooling layer, a ReLu layer, a Transpose convolution layer, and Unpooling layer, 1x1 convolution layer, skip connection, Global Average Pooling (GAP) layer, Fully Connected layer, SVM (support Vector Machine), LSTM (long short term memory), Atrous convolution ( Atrous Convolution), Atrous Spatial Pyramid Pooling, Separable Convolution, and Bilinear Upsampling. Preferably, the artificial intelligence neural network further includes an operation unit for batch normalization operation in front of the ReLu layer.

1 and 2 are schematic diagrams and operating principles of a breast cancer diagnosis device according to an embodiment of the present disclosure, respectively.

Referring to FIGS. 1 and 2 , the breast cancer diagnosis apparatus 600 using the smart mirror 700 according to the present invention is a thermal imaging camera mode while the patient looks at his or her own image reflected in the smart mirror 700, a thermal imaging camera 200 for obtaining a thermal image made of a two-dimensional image of changes in infrared radiation emitted according to the distribution of the surface temperature of the affected area; A plurality of image sensors (50a, 50b) for obtaining a body image of the patient; A speaker that checks the patient's posture information based on the body image and delivers a feedback control command to the patient in order to derive an optimized standard posture for breast thermal imaging camera examination or informs the patient of guidance and instructions for breast cancer examination ( 60) and display panel 20b; a breast temperature control device (53) for providing thermal images of the breast at each temperature level to the thermal imaging camera (200) while gradually lowering the temperature of the breast region according to the ambient temperature or the patient's body excitement state; a thermal breast cancer detection unit 52 that detects a breast cancer hot spot, which is a breast cancer suspected region, from the breast thermal image; Body posture interaction, which creates a body map in which virtual body organs are arranged on the body image, or guides the body posture to be taken by the patient during breast cancer diagnosis through the object image and the speaker 60 and gives feedback to the patient. ) part 51; a virtual object imaging unit 88 generating a virtual object image on the display panel 20b; and a pre-trained thermal image artificial neural network 41 for analyzing the breast thermal image to determine signs of breast cancer. The smart mirror 700 automatically determines the patient's risk of breast cancer by using the pre-learned thermal image artificial intelligence neural network 41 with breast thermal image information obtained from the patient. .

The smart mirror 700 of the present invention is a display in which a mirror 20a and a display panel 20b are combined, and a virtual object image necessary for breast cancer examination is displayed on the patient's own image reflected in the mirror 20a. ), it can be shown in augmented reality to patients undergoing self-diagnosis.

The thermal breast cancer detection unit 52 determines a breast thermal image mapper 80 for obtaining a breast thermal image from a thermal image and a cutoff value determined by the cutoff value adjusting unit 82 on the breast thermal image. and a breast cancer hot spot memory 84 for storing breast cancer hot spots composed of pixels having a temperature value greater than the value.

Preferably, the breast cancer hot spot memory 84 stores breast cancer hot spot information or a thermal hot spot image itself.

The breast cancer hot spot information may include center coordinates (positions) of breast cancer hot spots, regions of breast cancer hot spots, and image pixel values thereof.

The breast cancer hot spot area and center coordinates (position) obtained during the thermal imaging camera mode are preferably calculated based on the breast outline.

The breast thermal image may be generated by the breast thermal image mapper 80 as an average value of a cumulative sum for each pixel of thermal images of the breast region captured a predetermined number of times (eg, 6 times or more for 6 months) for a predetermined period of time. To this end, the breast thermal image mapper 80 arranges the thermal images of the breast region on the basis of the breast outline in a two-dimensional space, performs pixel-by-pixel addition between these thermal images, and calculates the breast as an average value. It is preferred to obtain thermal images.

Another aspect of the breast thermal image is to obtain thermal images of the patient's breast region while lowering the ambient temperature step by step by the breast temperature controller 53, and then by the breast thermal image mapper 80. A breast thermal image may be generated with an average value after pixel-by-pixel addition is performed on the thermal images of the breast region.

The breast cancer hot spot area obtained in the thermal imaging camera mode refers to an area with a higher temperature than other areas, and the breast cancer hot spot area applies a predetermined cutoff value selected by the cutoff value adjusting unit 82 to the breast thermal image. By application, it may be composed of pixels having a value greater than or equal to the cutoff value.

In addition, if the patient additionally conducts a re-examination on the breast cancer hot spot area obtained in the thermal imaging camera mode through the self-examination mode, the breast cancer examination can be performed more precisely, increasing the accuracy of the examination and detecting breast cancer at an early stage. .

According to one embodiment of the present application, the breast temperature control device 53 is for removing a measurement deviation of the thermal image of the breast region by adjusting the ambient temperature, and analyzes the thermal image of the patient's breast region to remove the measurement deviation of the thermal image of the breast region. a breast temperature distribution calculator 81 that calculates the temperature or the excited state of the patient's body; a cooling fan (500) for providing cool air to the patient's breast; and a remote controller transmitter 19 for controlling the temperature of the surface of the breast by controlling the temperature of the cooling fan 500 step by step according to the calculation result of the breast temperature distribution calculator 81, but is not limited thereto.

That is, when it is determined by the breast temperature distribution calculator 81 that the ambient temperature is too hot or the patient is in a state of body excitement, the breast temperature control device 53 lowers the ambient temperature step by step, By lowering the surface temperature of the breast, the effect of the ambient temperature or the patient's thermal excitement is minimized, so that the infrared radiation emitted from the lesion area is more clearly revealed as a breast cancer hot spot area.

The cooling fan 500 may include a remote control receiving unit 18 for receiving a control command of the cooling fan 500 from the remote control transmitting unit 19 and a cooling fan opening 17 through which cold air is emitted.

The breast temperature distribution calculator 81 takes pixels of a thermal image in the patient's body region except for the breast region to determine whether the ambient temperature is too hot or the patient's body excitement state, and the average value of these pixels is a predetermined cutoff value. It is preferably done by judging whether or not it exceeds. If the predetermined cut-off value is exceeded, it is determined that the ambient temperature is too hot or the patient's body is excited. ) is sent to.

Therefore, while gradually lowering the ambient temperature, thermal images of the patient's breast region are obtained, and then, by the breast thermal image mapper 80, pixel-by-pixel additions of the thermal images of the breast region are taken, and the average value of the thermal images is obtained. You can create visual images.

The hot spot guider 86 according to an embodiment of the present invention reads breast cancer hot spots obtained in the thermal imaging camera mode during the self-examination mode from the breast cancer hot spot memory 84, and uses the virtual object imaging unit 88 to display the breast cancer hot spots to the patient. It is characterized in that the appearance of is displayed on the display panel (20b) as an augmented reality superimposed on the reflected mirror (20a).

For example, during the self-examination mode, the patient sees an overlapping image between the breast cancer hot spot area indicated by the hot spot guider 86 and the breast area of the patient reflected in the mirror, and touches the breast cancer hot spot area by himself to check for lumps or lumps. A precise self-diagnosis is possible.

In this case, preferably, the hot spot guider 86 displays breast hot spot areas where self-examination is completed in a blue color, and breast hot spot areas where self-examination is not completed are displayed in a red color. do.

Therefore, the patient can perform an interactive breast cancer self-examination based on augmented reality while receiving help from the smart mirror 700 .

The body posture interaction unit 51 according to an embodiment of the present invention finds body outlines from the patient's body image, and includes the patient's face, breast, arm, stomach, legs, and feet. a body navigator 90 that creates a body map by creating boundary lines that distinguish body organs such as a foot; a body posture correction request unit 92 that displays a body outline image for posture fitting on the smart mirror through a virtual object image in order to provide a body posture to be taken by a patient during breast cancer diagnosis; and a body fitting check unit 95 that calculates a degree of fitting between the body outline image of the patient and the body outline image for posture fitting and gives feedback to the patient.

In the present invention, the determination of the fitting degree is calculated using any one of Sum of Squared Difference (SSD), Sum of Absolute Difference (SAD), K-nearest neighbor algorithm (KNN), and Normalized Cross Correlation (NCC) techniques It is preferred to do

In addition, the smart mirror 700 is provided with a patient authentication unit 91 that recognizes who the corresponding face is by comparing the image of the face on the body map with the patient's face database registered in advance, and the control unit 70 ) is preferred to activate the thermal imaging camera mode and self-examination mode only for the face-recognized patient.

Another aspect of the patient authentication unit 91 is to authenticate the patient by inputting a fingerprint or ID number, and the control unit 70 preferably activates the thermal imaging camera mode and self-examination mode only for the authenticated patient. do.

Preferably, the image sensors 50a and 50b are installed through openings (not shown) prepared on the display panel 20b and the mirror 20a.

In addition, the breast cancer diagnosis apparatus according to the present disclosure may include a power supply unit 55 for supplying electricity to each part of the smart mirror 700 .

The breast cancer tracking management unit 72 according to an embodiment of the present invention performs a periodic thermal image tracking test, and tracks and manages the analysis result of the thermal image artificial intelligence neural network, the size of breast cancer hot spots, and the trend of changes in the number of breast cancer hot spots.

In addition, the breast cancer tracking management unit 72 may collect breast cancer risk information of a patient calculated during a thermal imaging camera mode or a self-examination mode, and notify the patient of the patient's breast cancer risk level and a schedule for the next breast cancer examination. The breast cancer tracking management unit 72 drives and controls the speaker 60 and the display panel 20b through the controller 70 to notify the patient of the breast cancer examination schedule or breast cancer risk determination result through object images or text transmission. Preferred.

In addition, the breast cancer tracking and management unit 72 uses the thermal image artificial intelligence neural network 41 to select regions suspected of being tumors, masses, or calcification clusters in the thermal imaging camera mode as breast cancer hot spots in the breast cancer hot spot memory 84. can be added to.

In addition, the smart mirror 700 may further include a communication means (not shown) for providing Wi-Fi, Bluetooth connection, wired/wireless Internet, and Internet of Things.

In addition, the control unit 70 uses a virtual object imaging unit 88 and a speaker to provide feedback information to the patient according to the operations of the breast cancer tracking management unit 72, patient authentication unit 91, and body fitting check unit 95. It performs the function of controlling (60).

In addition, the smart mirror 700 may include a smart mirror frame 700a surrounding an edge of the smart mirror.

3A and 3B are examples of a body navigator 90 for obtaining a body map 33 from a body image 31 in the device for diagnosing breast cancer according to an embodiment of the present invention.

Referring to FIG. 3A , a body image labeled with body organs including a face 36a, a breast 36b, an arm 36c, an armpit 36d, a stomach 36e, a leg 36f, and a foot 37g The body navigator 90 is implemented by obtaining the body map 33 by the artificial intelligence neural network learned by (31).

For example, when it is necessary to check the patient's armpit 36d during breast cancer diagnosis, the body navigator 90 checks the body map obtained by semantic segmentation of the patient's body image 31 to locate the armpit 36d. can figure it out

Also, since the breast region 36b can be identified on the body map 33, only thermal image pixels of the breast region can be separately extracted from the thermal image.

It is preferred that the body map 33 of the present invention includes boundary lines distinguishing body organs.

The body organs are characterized in that they include the patient's breast outline and body outline.

3B is another embodiment of the body navigator 90, which obtains a body outline image 35 from a body image 31 and provides medical and physical information between body organs (face, arm, breast, abdomen, leg, foot). A body map 33 can be obtained by marking the body organs on the body outline image 35 according to a body organ layout method that is statistically based on the arrangement situation and the size ratio between the body organs.

In another aspect of FIG. 3B, a body outline image 35 is obtained from the body image 31, and the body outline images 34 including boundary lines dividing body organs are labeled by images. An embodiment of obtaining a body map 33 by a learned artificial intelligence neural network is shown.

The body outline image 35 of FIG. 3B learns the artificial intelligence neural network by images labeled with the body outline image for the given body image 31, and then the patient's body image 31 given from the image sensor. ), the body outline image 35 can be obtained using the learned artificial intelligence neural network.

Another aspect of the body outline image 35 of FIG. 3B is to obtain a differential image through differential operation between an image when the patient is not in front of the smart mirror 700 and an image when the patient is present, and the edge component thereof By taking the body outline image 35 can be obtained.

FIG. 4A is a standing position information providing means (not shown) for providing the patient 77 with location information about an optimal standing place that is advantageous for breast cancer diagnosis when the patient 77 stands in front of the smart mirror 700. shows an example of

The standing position information providing means measures the patient's height using the image sensors 50a and 50b to calculate a standing position that is advantageous for breast cancer diagnosis, and uses the laser beam irradiation means 30a and 30b to obtain a laser footprint pattern 99a. ) can be achieved by laser beam illumination at the corresponding floor location.

For example, a laser footprint pattern 99a in the shape of a footprint is formed on the floor where the smart mirror 700 is installed by the laser beam irradiation means 30a and 30b to visually provide footprint location information to the patient 77. .

Another aspect of the standing position information providing means is to display the body outline image 32 for posture fitting or the footprint outline 99b for fitting as an object image on the display panel 20b through the virtual object image unit 88. It can be done.

The body outline image 32 for posture fitting is a virtual body outline image displayed on the display panel 20b using the standing position as the origin coordinate. At this time, the body fitting check unit 95 calculates the degree of fitting between the body outline image 35 based on the coordinates at which the patient is currently standing and the body outline image 32 for posture fitting, and feeds back to the patient through the control unit 70. Preferred.

The controller 70 can simultaneously display the body outline image 32 for posture fitting and the body outline image 35 as virtual objects on the display panel 20b so that the patient 77 can visually feel the degree of fitting. there is. Also, the controller 70 may guide the patient 77 to a fitting posture through the speaker 60 .

Preferably, the standing position information providing means provides the patient with the same standing position as used in the thermal imaging camera mode even during the self-examination mode.

Figure 4a is a standing position for providing standing position information about an optimal standing place favorable for breast cancer diagnosis to a patient when a patient 77 stands in front of a smart mirror 700 using a breast cancer diagnosis apparatus according to an embodiment of the present invention. This is an embodiment of the location information providing unit.

Specifically, during the self-examination mode when the patient 77 stands in front of the smart mirror 700, the area and location of the breast cancer hot spot 84a obtained in the thermal imaging camera mode are superimposed on the patient's breast reflected in the mirror through an object image. Show examples of augmented reality.

4B is an embodiment of a means for providing standing position information having a means for determining the position of a patient's foot print in the apparatus for diagnosing breast cancer according to an embodiment of the present invention.

Specifically, a body fitting check unit 95 for checking whether the patient's sole 99c is within the range of the laser footprint pattern 99a by the image sensors 50a and 50b, that is, how well the fitting is. It is another embodiment of the standing position information providing means using.

Referring to FIG. 4B, reference numeral 99b denotes a footprint outline 99b for fitting, which is a virtual object image for a laser footprint pattern 99a generated by laser irradiation or a footprint shape at a desired standing position displayed on a display panel. It is an image of a virtual object.

Also, reference numeral 99c denotes a patient's sole pattern 99c, which is a virtual object image of the patient's foot 36g.

The distance difference between the actual patient's foot 36g and the laser footprint pattern 99a can be calculated by the image sensors 50a and 50b providing stereo vision, and fitting this distance difference through the virtual object image unit 88. It can be expressed on the display panel 20b with the dragon footprint outline 99b and the patient's sole pattern 99c.

It is preferred that the body fitting check unit 95 calculates the degree of fitting between the footprint outline 99b for fitting and the patient's sole pattern 99c and gives feedback to the patient through the control unit 70. A fitting guide is provided to the patient by virtual object images on the speaker 60 and the display panel 20b.

Also, when the patient's foot 36g stays outside the range of the laser footprint pattern 99a, it is preferable that the laser footprint pattern 99a blinks.

The control unit 70 according to the present invention is authenticated and at the same time, the patient's body outline image 35 is converted into the body outline image 32 for posture fitting and the thermal imaging camera mode and self-examination mode only for the patient in which the fitting is properly performed. It is preferred to activate

5A and 5B show a self-exam by overlapping a virtual object image required for breast cancer diagnosis on a display panel 20b on the patient 77's own image reflected in the mirror 20a using the smart mirror 700. It is an embodiment in progress.

Specifically, FIG. 5A is an embodiment of re-examining the breast cancer hot spot 84a obtained in the thermal imaging camera mode by a breast self-examination method, and the location of the breast cancer hot spot 84a displayed on the display panel 20b. Perform a breast self-examination while recognizing The body outline image 32 includes a breast outline 32b and a body outline 32a.

By showing the position of the breast cancer hot spot 84a obtained in the thermal imaging camera mode to the patient as an object image, the patient 77 can easily identify the region to be self-examined.

During the self-examination mode, the hot spot guider 86 superimposes the area and location of the breast cancer hot spot 84a obtained in the thermal imaging camera mode on the patient's breast reflected in the mirror through an object image based on the breast outline 32b. show in augmented reality.

5B is another embodiment in which the breast cancer hot spot 84a obtained in the thermal imaging camera mode is re-examined with the help of the self-examination helper 85. It is a virtual object image that helps the patient perform a breast self-examination by showing hand gestures and movement directions to the patient on the display panel 20b.

In addition to virtual object images, the self-examination helper 85 may accompany an interactive process of questioning and answering by digital ARS (Automatic Response Service) provided by the smart mirror.

For example, digital ARS asks the patient during a self-exam, "Do you have a lump at this location?" The question is asked, and the patient can answer "yes" or "no", and the answer is stored by the breast cancer tracking management unit 72 and can be used to determine the patient's breast cancer risk.

Another aspect of the question and answer of the digital ARS can be achieved by a question and answer sentence provided as an object image on a display panel along with a voice message and a click window for selecting an answer by touch.

The hot spot guider 86 displays the breast hot spot area 84b that has been re-examined in a blue color through the virtual object image, and the breast hot spot area 84c that has not been re-examined is a red color. It is preferred to indicate

Hereinafter, based on the details described above, the operation flow of the present application will be briefly reviewed. Although not shown in the drawing, the breast cancer self-diagnosis method may be performed by the artificially intelligent breast cancer diagnosis device described above. Hereinafter, for convenience of description, it will be described as being performed by an artificial intelligence breast cancer diagnosis device.

In step S100, it is possible to provide optimal positional information advantageous to breast cancer diagnosis to the patient by means of the standing positional information providing means.

In step S101, a thermal image of the breast portion may be obtained while lowering the surface temperature of the patient's breast in each temperature step.

In step S102, a step of analyzing the breast thermal image for each temperature step by using a pre-learned thermal image artificial intelligence neural network may be included.

In step S103, when a breast cancer hot spot is discovered by the thermal imaging camera, the patient may be requested to self-examine the breast cancer hot spot area using a breast self-examination method.

In step S104, during the self-diagnosis, the discovered breast cancer hot spot may be superimposed on the patient reflected in the mirror in augmented reality and displayed on the smart mirror.

In step S105, the breast cancer tracking management unit may notify the patient of a breast cancer risk level or a next breast cancer examination schedule by observing changes in the size and number of breast cancer hot spots as a result of the analysis of the thermal image artificial intelligence neural network.

In step S106, the patient's breast cancer symptoms are determined, and if breast cancer is suspected, remote diagnosis of breast cancer can be performed with the specialist doctor through the smart mirror connected to the specialist doctor through communication means.

The above description of the present application is for illustrative purposes, and those skilled in the art will understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present application. Therefore, the embodiments described above should be understood as illustrative in all respects and not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present application is indicated by the following claims rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and equivalent concepts thereof should be construed as being included in the scope of the present application.

17 Cooler opening
18 remote control receiver
19 Remote control transmitter
20a mirror
20b display panel
30a, 30b image sensors (laser beam irradiation means)
31 body image
32 Body outline image for posture fitting
32a body outline
32b Breast outline
33 body maps
35 body outline images
36a face 36b breast 36c arm 36d armpit 36e belly
36f legs 36g feet
41 Artificial Neural Networks for Thermal Imaging
50a, 50b image sensors
51 Body Posture Interaction
52 Thermal Imaging Breast Cancer Detector
53 breast thermostat
55 power supply
60 speaker
70 Control
72 Breast Cancer Tracking Department
77 patients
80 Breast Thermal Image Mapper
81 Breast temperature distribution calculator
82 cut-off control unit
84 Breast Cancer Hot Spot Memory
84a Breast Cancer Hot Spots
84b Breast hot spot area with reexamination completed
84c Breast hot spot areas not completed reexamination
85 Self-Test Helper
86 Hot Spot Guider
88 virtual object imaging unit
90 body navigator
91 Patient Certification Department
92 Body Posture Correction Request Unit
95 body fitting check
99a laser footprint pattern
Footprint outline for 99b fitting
99c patient plantar pattern
200 thermal imaging camera
500 cold fan
600 breast cancer diagnosis device
700 smart mirror
700a smart mirror frame

Claims (9)

smart mirror;
a thermal imaging camera for obtaining a thermal image by detecting thermal radiation emitted from the patient's body; and
a breast temperature control device that varies a breast surface temperature to minimize a measurement deviation of a thermal image of a breast region according to an ambient temperature or a patient's body arousal;
including,
In the artificial intelligence breast cancer diagnosis device,
The smart mirror further includes a thermal imaging breast cancer detection unit,
The thermal imaging breast cancer detection unit,
a breast thermal image mapper for generating a breast thermal image as an average value after taking pixel-by-pixel additions for thermal images of the patient's breast region obtained while the ambient temperature is lowered step by step by the breast temperature controller;
a cutoff adjustment unit for generating a thermal hot spot image consisting of breast cancer hot spots exhibiting a temperature value equal to or greater than a predetermined value in the breast thermal image; and
a breast cancer hot spot memory for storing the thermal hot spot image or the breast cancer hot spot information;
that includes,
The artificial intelligence breast cancer diagnosis device further includes a thermal image artificial intelligence neural network,
The thermal image artificial intelligence neural network automatically determines the patient's breast cancer risk through the breast thermal image generated by the breast thermal image mapper,
Artificial intelligence breast cancer diagnosis device.
According to claim 1,
The breast temperature control device,
a breast temperature distribution calculator that analyzes the thermal image of the breast area and calculates the ambient temperature or the patient's body excitement state;
a cooling fan for providing cool air to the patient's breast; and
a remote controller transmitting unit controlling the temperature of the breast surface by controlling the temperature of the cooling fan step by step according to the calculation result of the breast temperature distribution calculator;
Which includes,
Artificial intelligence breast cancer diagnosis device.
According to claim 1,
The smart mirror further includes a thermal imaging breast cancer detection unit,
The thermal imaging breast cancer detection unit,
a breast thermal image mapper configured to generate a breast thermal image composed of an average value of a pixel-by-pixel cumulative sum of thermal images of a breast region captured at least a predetermined number of times during a predetermined period of time;
a cutoff adjustment unit for generating a thermal hot spot image consisting of breast cancer hot spots exhibiting a temperature value equal to or greater than a predetermined value in the breast thermal image; and
a breast cancer hot spot memory for storing the thermal hot spot image and the breast cancer hot spot information;
Which includes,
Artificial intelligence breast cancer diagnosis device.
delete According to claim 1,
The smart mirror further includes a hot spot guider that informs the breast cancer hot spot area and location in augmented reality by overlapping a mirror on which the patient's appearance is projected through a virtual object image,
Artificial intelligence breast cancer diagnosis device.
According to claim 1,
The smart mirror further includes a body posture interaction unit,
The body posture interaction unit,
a body navigator that creates a body map including body outlines and boundary lines distinguishing body organs from the patient's body image;
a body posture correction request unit for displaying a body outline image for posture fitting on the smart mirror through a virtual object image to provide a body posture to be taken by a patient during breast cancer diagnosis;
a body fitting check unit that calculates a degree of fitting between the patient's body outline image and the body outline image for posture fitting and gives feedback to the patient; and
a patient authentication unit that verifies who the patient is;
Which includes,
Artificial intelligence breast cancer diagnosis device.
According to claim 1,
The smart mirror further includes a standing position information providing means for calculating and providing optimal position information advantageous to breast cancer diagnosis to a patient,
Artificial intelligence breast cancer diagnosis device.
According to claim 1,
The smart mirror,
As a result of the analysis of the thermal image artificial intelligence neural network according to the periodic follow-up examination by the thermal imaging camera, the size and number of breast cancer hot spots are observed to inform the patient of the risk level or to inform the patient of the schedule for the next breast cancer thermal imaging camera examination. An artificial intelligent breast cancer diagnosis device that further includes a breast cancer tracking management unit.
In the breast cancer self-diagnosis method performed by the artificial intelligence breast cancer diagnosis device according to any one of claims 1 to 3 and 5 to 8,
The breast cancer self-diagnosis method,
providing optimum location information advantageous to breast cancer diagnosis to a patient by means of a standing location information providing means;
obtaining a thermal image of the breast by a thermal imaging camera while gradually lowering the temperature of the breast surface step by step by the breast temperature controller;
finding a breast cancer hot spot from a breast thermal image by a thermal breast cancer search unit;
if a breast cancer hot spot is found, requesting a self-examination of the breast cancer hot spot area from the patient through a speaker and a display panel;
superimposing breast cancer hot spots discovered during self-examination by a hot spot guider on the patient's image reflected in the mirror in augmented reality and displaying them on the smart mirror; and
Informing the patient of the breast cancer risk level or the next breast cancer examination schedule by observing changes in the size of breast cancer hot spots and the number of breast cancer hot spots as a result of analysis of the thermal image artificial intelligence neural network by the breast cancer tracking management unit; Which includes,
How to self-diagnose breast cancer.

Images