Cancer-Net SCa: tailored deep neural network designs for detection of skin cancer from dermoscopy images

Lee, James Ren Hou; Pavlova, Maya; Famouri, Mahmoud; Wong, Alexander

doi:10.1186/s12880-022-00871-w

Cited by 16 publications

(6 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tailored deep neural network architecture has been proposed in [39]. The authors have implemented data balancing associated with data augmentation including random rotations, shifts, illumination correction, and contrast enhancement in order to enhance image quality and the generalization ability of their proposed model.…”

Section: Literature Reviewmentioning

confidence: 99%

MSDSC: A Multistage Deep Learning-based Skin Cancer Classifier

Hassan¹,

Mahar²,

Fouad³

2023

Preprint

View full text Add to dashboard Cite

One of the most frequent types of cancer in the globe is skin cancer which is a complicated public health issue. A sample of tissue from the skin lesion confirms the diagnosis of skin cancer. However, before making a definitive diagnosis, professionals detect specific signs that can be considered as early diagnosis. An early skin cancer diagnosis is prone to mistakes due to specialists’ lack of knowledge and similarities with other illnesses. This study presents a multistage deep learning-based skin cancer classifier (MSDSC) to give early detection of melanoma and non-melanoma skin lesions. The first stage is the pre-processing of samples to remove the hair surrounding the skin lesion. Since the dataset is unbalanced, the second stage was creating synthetic images using the Generative Adversarial Networks (GAN) model. This stage is followed by an attention-based U-Net model that gener- ates masks for regions of interest to eliminate the background. Lastly, two different types of EfficientNet are trained to classify skin lesions using segmented images. The proposed framework was tested using the International Skin Imaging Collaboration dataset (ISIC) and the results showed that it outperformed the state-of-the-art studies achieving an accuracy of 0.96, F1-score of 0.91, recall of 0.95, and precision of 0.88.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

MSDSC: A Multistage Deep Learning-based Skin Cancer Classifier

Hassan¹,

Mahar²,

Fouad³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Here, we consider ResNet101 [77] as backbone for DeepLabV3+, ResNet101 is a very popular CNN that obtains residual functions by referencing block inputs (for a complete list of CNN structures please refer to [78]). It is pre-trained on the VOC segmentation dataset and then tuned using the parameters specified on the github page 3 . We adopted the same parameters to prevent overfitting (i.e.…”

Section: Deep Learning For Semantic Image Segmentationmentioning

confidence: 99%

A Standardized Approach for Skin Detection: Analysis of the Literature and Case-Studies

Nanni¹,

Loreggia²,

Lumini³

et al. 2022

Preprint

View full text Add to dashboard Cite

Skin detection, the process of distinguishing between skin and non-skin regions in a digital image, is widely used in a variety of applications ranging from hand gesture analysis to body part tracking to facial recognition. Skin detection is a challenging problem that has received a lot of attention from experts and proposals from the research community in the context of intelligent systems, but the lack of common benchmarks and unified testing protocols has hampered fairness among approaches. Comparisons are very difficult. Recently, the success of deep neural networks has had a major impact on the field of image segmentation detection, resulting in various successful models to date. In this work, we survey the most recent research in this field and propose fair comparisons between approaches using several different datasets. The main contributions of this work are: (i) a comprehensive literature review of approaches to skin color detection and a comparison of approaches that may help researchers and practitioners choose the best method for their application; (ii) a comprehensive list of datasets that report ground truth for skin detection; (iii) a framework for evaluating and combining different skin detection approaches. Moreover, we proposed an ensemble of convolutional neural networks and transformers that obtains state of the art performance. All the code is made publicly available at https://github.com/LorisNanni

show abstract

“…In addition, the association of dermoscopy with some other technologies adds new development possibilities, such as super high magnification optical dermoscopy [ 6 ], teledermoscopy [ 7 ], ex vivo dermoscopy [ 8 ] and artificial intelligence-assisted diagnosis [ 9 ].…”

Section: Introductionmentioning

confidence: 99%

Status of Dermoscopy in Chile: First National Study for Dermatologists

Morales Etcheberry,

González-Coloma,

Alonso-Traviesa

et al. 2024

Dermatology Practical &Amp; Conceptual

View full text Add to dashboard Cite

Introduction: Scientific evidence supports dermoscopy as an essential tool in dermatological diagnosis. Objectives: The objective is to know the factors that influence its use in Chilean dermatologists. Materials and Methods: Analytical cross-sectional study. An adapted version of the survey was submitted from the pan-European study by Forsea et.al. to members of the Chilean Society of Dermatology, between September and December 2020. Analysis using descriptive statistics and multivariate analysis with ordinal logistic regression looking for factors associated with greater use of. Results: 198 responses, mean age 46.3 years and 14.6 years on average practicing as dermatologists. 61.6% trained in dermoscopy during their residency. 98% use a dermatoscope. More than 80% consider dermoscopy useful for the diagnosis of melanomas, follow-up of melanocytic lesions, and diagnosis of pigmented and non-pigmented tumors. Between 50% and 70% consider it useful for monitoring non-melanocytic lesions, nail and hair pathologies. Greater confidence when evaluating pigmented and non-pigmented tumors and capillary pathology. Adjusting for age, sex, confidence, and education, participation in teaching was associated with greater use of dermoscopy in non-pigmented and pigmented tumors, and capillary pathology. Conclusions: Percentage of participation in the survey and training in dermoscopy higher than in the reference study, recognizing the usefulness of dermoscopy for the diagnosis and follow-up of tumor pathologies. Participating in teaching is a strong independent factor that is associated with a greater use of dermoscopy in Chile. Dermoscopy is positioned as a tool widely used by Chilean dermatologists in their daily practice.

show abstract

Cancer-Net SCa: tailored deep neural network designs for detection of skin cancer from dermoscopy images

Cited by 16 publications

References 66 publications

MSDSC: A Multistage Deep Learning-based Skin Cancer Classifier

MSDSC: A Multistage Deep Learning-based Skin Cancer Classifier

A Standardized Approach for Skin Detection: Analysis of the Literature and Case-Studies

Status of Dermoscopy in Chile: First National Study for Dermatologists

Contact Info

Product

Resources

About