A Non-Invasive Interpretable Diagnosis of Melanoma Skin Cancer Using Deep Learning and Ensemble Stacking of Machine Learning Models

Alfi, Iftiaz A.; Rahman, Md. Mahfuzur; Shorfuzzaman, Mohammad; Nazir, Amril

doi:10.3390/diagnostics12030726

Cited by 54 publications

(26 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Some have used concept-based techniques [21]. Interpretability via visualization techniques [22][23][24] such as heatmaps is another method that has been used in research in this domain. A novel CNN based pipeline is describe in [23] for Melanoma identification with the WSIs (Whole-slide images) along with a heat map-based interpretability approach.…”

Section: Irregular Dots and Globulesmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Label Prototype Based Interpretable Machine Learning for Melanoma Detection

Hussaindeen¹,

Iqbal²,

Ambegoda³

2022

IJASIS

View full text Add to dashboard Cite

Skin cancer is the most common of all cancers that exist in the world and melanoma is the deadliest among all the skin cancers. It is found that melanoma roughly kills a person every hour somewhere in the world. Considering the severity of the disease, significant effort goes into minimizing delays in the process of diagnosing melanoma. There are several approaches based on Machine Learning (ML) that can assist dermatologists in melanoma detection. However, many experts hesitate to trust ML systems due to their black-box nature, despite the accuracy of their performance. This highlights the need for applications that facilitate not only accurate classifications but also the ability to justify such decisions. In this work, we propose a prototypebased interpretable melanoma detector that uses the Seven Point Checklist, a well-known criterion used for the detection of melanoma. Prototypes provide the justification behind the decisions suggested by the ML model in a way of showing similar cases that are already known. In addition to identifying the dermoscopic features listed in the seven-point checklist, our work aims to provide reasoning that is similar to the ones used by the dermatologists in clinical practice for each decision made by the model. F1-Score has been used as the main performance metric in evaluating the model performance and that of the best performing class was 0.87. Furthermore, we show comparisons of our approach with Local Interpretable Model-Agnostic Explanations (LIME), a popular approach for interpretability for deep learning models.

show abstract

Section: Irregular Dots and Globulesmentioning

confidence: 99%

“…The model has achieved an AUROC of 0.962 according to the evaluation results. Another recent work is discussed in [24] that has followed an ensemble approach for the diagnosis of Melanoma in skin lesions. This approach also has used heat maps as visual explanations for their classifications.…”

Section: Irregular Dots and Globulesmentioning

confidence: 99%

Multi-Label Prototype Based Interpretable Machine Learning for Melanoma Detection

Hussaindeen¹,

Iqbal²,

Ambegoda³

2022

IJASIS

View full text Add to dashboard Cite

show abstract

“…Melanoma is multicolored, asymmetric and contrasting [4]. The ABCD rule is the classic method used by dermatologists [5].…”

Section: Introductionmentioning

confidence: 99%

U-Net-RCB7: Image Segmentation Algorithm

AKYEL¹

2023

Politeknik Dergisi

View full text Add to dashboard Cite

The incidence of skin cancer is increasing. Early detection of cases of skin cancer is vital for treatment. Recently, computerized methods have been widely used in cancer diagnosis. These methods have important advantages such as no human error, short diagnosis time, and low cost. We can segment skin cancer images using deep learning and image processing. Properly segmented images can help doctors predict the type of skin cancer. However, skin images can contain noise such as hair. These noises affect the accuracy of segmentation. In our study, we created a noise dataset. It contains 3000 images and masks. We performed noise removal and lesion segmentation by utilizing the ISIC and PH2. We have developed a new deep learning model called U-Net-RCB7. U-Net-RCB7 contains EfficientNetB7 as the encoder and ResNetC before the last layer. This paper uses a modified U-Net model. Images were divided into 36 layers to prevent loss of pixel values in the images. As a result, noise removal and lesion segmentation were 96% and 98.36% successful, respectively.

show abstract

“…Dermoscopy, or less commonly confocal microscopy, plays a vital role in the in vivo visualisation of lesioned features, allowing for more accuracy in risk stratification in the dermatological field. In various studies, AI-based algorithms exceed clinical performance to detect disease in medical imaging [ 8 , 13 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

An Efficient Deep Learning-Based Skin Cancer Classifier for an Imbalanced Dataset

et al. 2022

View full text Add to dashboard Cite

Efficient skin cancer detection using images is a challenging task in the healthcare domain. In today’s medical practices, skin cancer detection is a time-consuming procedure that may lead to a patient’s death in later stages. The diagnosis of skin cancer at an earlier stage is crucial for the success rate of complete cure. The efficient detection of skin cancer is a challenging task. Therefore, the numbers of skilful dermatologists around the globe are not enough to deal with today’s healthcare. The huge difference between data from various healthcare sector classes leads to data imbalance problems. Due to data imbalance issues, deep learning models are often trained on one class more than others. This study proposes a novel deep learning-based skin cancer detector using an imbalanced dataset. Data augmentation was used to balance various skin cancer classes to overcome the data imbalance. The Skin Cancer MNIST: HAM10000 dataset was employed, which consists of seven classes of skin lesions. Deep learning models are widely used in disease diagnosis through images. Deep learning-based models (AlexNet, InceptionV3, and RegNetY-320) were employed to classify skin cancer. The proposed framework was also tuned with various combinations of hyperparameters. The results show that RegNetY-320 outperformed InceptionV3 and AlexNet in terms of the accuracy, F1-score, and receiver operating characteristic (ROC) curve both on the imbalanced and balanced datasets. The performance of the proposed framework was better than that of conventional methods. The accuracy, F1-score, and ROC curve value obtained with the proposed framework were 91%, 88.1%, and 0.95, which were significantly better than those of the state-of-the-art method, which achieved 85%, 69.3%, and 0.90, respectively. Our proposed framework may assist in disease identification, which could save lives, reduce unnecessary biopsies, and reduce costs for patients, dermatologists, and healthcare professionals.

show abstract

A Non-Invasive Interpretable Diagnosis of Melanoma Skin Cancer Using Deep Learning and Ensemble Stacking of Machine Learning Models

Cited by 54 publications

References 24 publications

Multi-Label Prototype Based Interpretable Machine Learning for Melanoma Detection

Multi-Label Prototype Based Interpretable Machine Learning for Melanoma Detection

U-Net-RCB7: Image Segmentation Algorithm

An Efficient Deep Learning-Based Skin Cancer Classifier for an Imbalanced Dataset

Contact Info

Product

Resources

About