2021
DOI: 10.1038/s41598-021-84593-z
|View full text |Cite
|
Sign up to set email alerts
|

AI-based localization and classification of skin disease with erythema

Abstract: Although computer-aided diagnosis (CAD) is used to improve the quality of diagnosis in various medical fields such as mammography and colonography, it is not used in dermatology, where noninvasive screening tests are performed only with the naked eye, and avoidable inaccuracies may exist. This study shows that CAD may also be a viable option in dermatology by presenting a novel method to sequentially combine accurate segmentation and classification models. Given an image of the skin, we decompose the image to … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
14
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
4
2
1
1

Relationship

0
8

Authors

Journals

citations
Cited by 20 publications
(14 citation statements)
references
References 19 publications
0
14
0
Order By: Relevance
“…Deep learning-based approaches have recently gained popularity in skin pigment analysis, especially in the segmentation of erythema and hyperpigmented areas [12][13][14]. Regression analysis is a commonly used technique that a neural network architecture learns the relationship between input images and ground truths to predict desired outputs.…”
Section: Introductionmentioning
confidence: 99%
“…Deep learning-based approaches have recently gained popularity in skin pigment analysis, especially in the segmentation of erythema and hyperpigmented areas [12][13][14]. Regression analysis is a commonly used technique that a neural network architecture learns the relationship between input images and ground truths to predict desired outputs.…”
Section: Introductionmentioning
confidence: 99%
“…Furthermore, unlike existing techniques based on expert judgement grading [4], the developed system can provide not only quantified values, such as the number of diagnostic points, average size, total area, and intensity of the entire facial skin, but also visual information, such as the appearance of the location. Providing such diagnostic results requires a lot of time and effort in preparing image training data; however, it is very useful for researchers because it can provide more quantified and visualized information [1,2]. These advantages were also demonstrated in another study of changes in facial skin characteristics with age in Korean women.…”
Section: Discussionmentioning
confidence: 99%
“…The use of skin images for diagnostic purposes has been widespread in the fields of medicine and skin research. However, recent advances in artificial intelligence (AI) learning technologies, particularly deep learning, have accelerated the development of diagnostic tools using skin images [1,2]. Previous diagnostic methods relied on simple feature extraction methods, which made it difficult to account for complex changes in skin characteristics.…”
Section: Introductionmentioning
confidence: 99%
“…It also helps in reaching a representation. That is a powerful feature because it helps identify the objects in the image regardless of where they are [29], [30]. Adding fully connected layer is an inexpensive way to teach non-linear components of these properties.…”
Section: Algorithmmentioning
confidence: 99%