KR-20260063489-A - Skin cancer diagnosis and skin health care system using AI

Abstract

The present invention relates to a skin cancer diagnosis and skin health management system utilizing AI. More specifically, the invention relates to a skin cancer diagnosis and skin health management system utilizing AI that acquires a user's skin image, diagnoses the presence or absence of skin cancer and skin health through a skin image diagnosis unit, and designs lifestyle habits and skincare routines based on the diagnosis results obtained through a medical history unit.

Inventors

• 정자연

Assignees

• 정자연

Dates

Publication Date

20260507

Application Date

20241030

Claims (3)

1. An integrated skin management server capable of diagnosing skin issues and managing skin health through skin images and surveys; User terminal that transmits and receives to and from the skin integrated management server via a network including Regarding the integrated skin management server A user data management unit capable of storing and managing data collected from a user terminal, A skin image diagnosis unit that captures a skin image to acquire a lesion image, and diagnoses the presence of skin cancer and skin health through the acquired user's skin image and questions, A medical history department that provides a skin type questionnaire to a user terminal to present the diagnostic results obtained, diagnoses lifestyle items based on the diagnostic results, and designs a skincare routine for correction accordingly, and A notification unit capable of sending notifications to a user terminal regarding execution status and urging execution AI-based skin cancer diagnosis and skin health management system including
2. In paragraph 1 The skin image diagnosis department Skin cancer diagnosis data construction unit that collects, refines, and processes data to determine the presence or absence of skin cancer, An image acquisition unit for acquiring an image of a skin lesion area, A skin lesion diagnostic unit that determines whether there is skin cancer from a skin lesion image, and Skin health diagnostic unit that evaluates pores, wrinkles, pigmentation, and skin age from skin images AI-based skin cancer diagnosis and skin health management system including
3. In paragraph 1 The medical history department A lifestyle diagnosis unit that provides a survey on lifestyle habits to a user terminal and generates a report on the survey results, A skin type diagnosis unit that calculates diagnostic results obtained from a survey and informs the appropriate skin type, A cosmetics recommendation unit that recommends cosmetics based on skin type and lifestyle habits, and Routine Design Department providing personalized routines for lifestyle correction AI-based skin cancer diagnosis and skin health management system including

Description

Skin cancer diagnosis and skin health care system using AI The present invention relates to a skin cancer diagnosis and skin health management system utilizing AI. More specifically, the invention relates to a skin cancer diagnosis and skin health management system utilizing AI that captures a user's skin image, diagnoses the presence or absence of skin cancer and skin health through a skin image diagnosis unit, and designs lifestyle habits and skincare routines based on the diagnosis results obtained through a medical history unit. With increasing interest in slow aging, there is growing attention not only on healthy diets and habits but also on skincare. Since using high-quality cosmetics can have a negative impact on the skin due to unsuitable products or poor lifestyle habits, there is significant interest in finding products that suit one's skin type. Furthermore, as interest in skin health grows, so does interest in skin health management; in particular, there is growing concern regarding skin cancer, which is on the rise every year. Skin cancer is a type of cancer that originates in the skin, the outermost layer of the human body, and includes basal cell carcinoma, squamous cell carcinoma, and melanoma. The number of skin cancer patients is increasing annually, and because skin cancer typically causes no pain or itching, it is difficult for the general public to clearly distinguish between skin cancer and moles. For an accurate diagnosis, a skin biopsy is required, which involves removing a portion of the lesion for examination under local anesthesia. Due to the complex diagnostic procedures and the difficulty in distinguishing between the two, patients often fail to seek examination or overlook the condition, missing the optimal time for treatment. As such, people are facing difficulties in skin care, such as damaging their skin health by not using appropriate cosmetics or missing a skin cancer. Therefore, there is an urgent need for a system that can diagnose the skin to determine the presence of cancer when lesions occur and recommend customized cosmetics suitable for the user's skin type. [Prior Art] KR Registered Patent Publication 10-2036043 (Published Oct. 24, 2019) FIG. 1 is a configuration diagram of a skin cancer diagnosis and health management system utilizing AI according to an embodiment of the present invention. FIG. 2 is a block diagram of a skin image diagnostic unit according to an embodiment of the present invention. FIG. 3 is a block diagram illustrating a skin cancer diagnosis data construction unit according to an embodiment of the present invention. FIG. 4 is a diagram showing the class attributes of metadata collected by the data collection unit of the skin cancer diagnosis data construction unit according to an embodiment of the present invention. FIG. 5 is a drawing illustrating labeling performed in a data labeling unit according to an embodiment of the present invention when the boundary line between normal skin and a lesion can be clearly distinguished. FIG. 6 is a drawing illustrating labeling in a data labeling unit according to an embodiment of the present invention when a lesion is obscured by hair but the lesion boundary line is distinguishable. FIG. 7 is a drawing illustrating labeling performed in a data labeling unit according to an embodiment of the present invention when dark and light colored parts exist simultaneously among a lesion. FIG. 8 is a drawing illustrating labeling in a data labeling unit according to an embodiment of the present invention when a distinguishable boundary line is formed outside the lesion due to a skin microscope lens, which is a diagnostic instrument. FIG. 9 is a drawing illustrating labeling in a data labeling unit according to an embodiment of the present invention when a boundary line is formed outside the lesion due to a healed scar. FIG. 10 is a drawing illustrating the process of labeling when there are multiple lesions within an image in a data labeling unit according to an embodiment of the present invention. FIG. 11 is a drawing illustrating an example of a case where labeling that does not meet processing standards is performed in a data inspection unit according to an embodiment of the present invention. FIG. 12 is a drawing illustrating another example of a case where labeling is performed that does not meet the processing standards according to an embodiment of the present invention. FIG. 13 is a drawing illustrating another example of a case where labeling is performed that does not meet the processing standards according to an embodiment of the present invention. FIG. 14 is a drawing illustrating a case according to an embodiment of the present invention where the extent of the lesion is wide and indistinguishable from normal skin. FIG. 15 is a drawing illustrating that labeling is not performed in a data inspection unit according to an embodiment of the present invention when the image resolution is low and appears blu