Detection of malignant melanoma in H&amp;E-stained images using deep learning techniques

Alheejawi, Salah; Berendt, Richard; Jha, Naresh; Maity, Santi P.; Mandal, Mrinal

doi:10.1016/j.tice.2021.101659

“…Skin-related disorders can be classified into permanent and temporary. Temporary disorders include (1) acne, (2) contact dermatitis, (3) cold sore, (4) keratosis pilaris, (5) blister, (6) hives, (7) sunburn, (8) actinic keratosis, (9) carbuncle, (10) latex allergy, (11) cellulitis, (12) measles, (13) chickenpox, and ( 14) impetigo. Permanent disorders can be divided into skin cancer and skin diseases (i.e., not cancer).…”

Section: Skin Disorders Taxonomymentioning

confidence: 99%

“…Alheejawi et al [6] suggested a DL-based technique to segment regions of melanoma. Results obtained using a small dataset of melanoma images showed that the suggested approach could perform the segmentation with a dice coefficient around 85%.…”

Section: Deep Learning-based Approachesmentioning

confidence: 99%

Skin cancer diagnosis based on deep transfer learning and sparrow search algorithm

Balaha

¹

,

Hassan

²

2022

Neural Comput & Applic

View full text Add to dashboard Cite

Skin cancer affects the lives of millions of people every year, as it is considered the most popular form of cancer. In the USA alone, approximately three and a half million people are diagnosed with skin cancer annually. The survival rate diminishes steeply as the skin cancer progresses. Despite this, it is an expensive and difficult procedure to discover this cancer type in the early stages. In this study, a threshold-based automatic approach for skin cancer detection, classification, and segmentation utilizing a meta-heuristic optimizer named sparrow search algorithm (SpaSA) is proposed. Five U-Net models (i.e., U-Net, U-Net++, Attention U-Net, V-net, and Swin U-Net) with different configurations are utilized to perform the segmentation process. Besides this, the meta-heuristic SpaSA optimizer is used to perform the optimization of the hyperparameters using eight pre-trained CNN models (i.e., VGG16, VGG19, MobileNet, MobileNetV2, MobileNetV3Large, MobileNetV3Small, NASNetMobile, and NASNetLarge). The dataset is gathered from five public sources in which two types of datasets are generated (i.e., 2-classes and 10-classes). For the segmentation, concerning the “skin cancer segmentation and classification” dataset, the best reported scores by U-Net++ with DenseNet201 as a backbone architecture are 0.104, $$94.16\%$$ 94.16 % , $$91.39\%$$ 91.39 % , $$99.03\%$$ 99.03 % , $$96.08\%$$ 96.08 % , $$96.41\%$$ 96.41 % , $$77.19\%$$ 77.19 % , $$75.47\%$$ 75.47 % in terms of loss, accuracy, F1-score, AUC, IoU, dice, hinge, and squared hinge, respectively, while for the “PH2” dataset, the best reported scores by the Attention U-Net with DenseNet201 as backbone architecture are 0.137, $$94.75\%$$ 94.75 % , $$92.65\%$$ 92.65 % , $$92.56\%$$ 92.56 % , $$92.74\%$$ 92.74 % , $$96.20\%$$ 96.20 % , $$86.30\%$$ 86.30 % , $$92.65\%$$ 92.65 % , $$69.28\%$$ 69.28 % , and $$68.04\%$$ 68.04 % in terms of loss, accuracy, F1-score, precision, sensitivity, specificity, IoU, dice, hinge, and squared hinge, respectively. For the “ISIC 2019 and 2020 Melanoma” dataset, the best reported overall accuracy from the applied CNN experiments is $$98.27\%$$ 98.27 % by the MobileNet pre-trained model. Similarly, for the “Melanoma Classification (HAM10K)” dataset, the best reported overall accuracy from the applied CNN experiments is $$98.83\%$$ 98.83 % by the MobileNet pre-trained model. For the “skin diseases image” dataset, the best reported overall accuracy from the applied CNN experiments is $$85.87\%$$ 85.87 % by the MobileNetV2 pre-trained model. After computing the results, the suggested approach is compared with 13 related studies.

show abstract

“…In dermatopathology, deep learning approaches have been applied for the segmentation of both melanocytic [ 21 ] and non-melanoma lesions, including basal cell carcinoma (BCC) [ 22 ]. A U-Net architecture automatically segmenting the epidermis in histological images has also been implemented [ 23 ].…”

Section: Introductionmentioning

confidence: 99%

Evaluation of a Deep Learning Approach to Differentiate Bowen’s Disease and Seborrheic Keratosis

Jansen

¹

,

Baguer

²

,

Duschner

³

et al. 2022

Cancers

View full text Add to dashboard Cite

Background: Some of the most common cutaneous neoplasms are Bowen’s disease and seborrheic keratosis, a malignant and a benign proliferation, respectively. These entities represent a significant fraction of a dermatopathologists’ workload, and in some cases, histological differentiation may be challenging. The potential of deep learning networks to distinguish these diseases is assessed. Methods: In total, 1935 whole-slide images from three institutions were scanned on two different slide scanners. A U-Net-based segmentation deep learning algorithm was trained on data from one of the centers to differentiate Bowen’s disease, seborrheic keratosis, and normal tissue, learning from annotations performed by dermatopathologists. Optimal thresholds for the class distinction of diagnoses were extracted and assessed on a test set with data from all three institutions. Results: We aimed to diagnose Bowen’s diseases with the highest sensitivity. A good performance was observed across all three centers, underlining the model’s robustness. In one of the centers, the distinction between Bowen’s disease and all other diagnoses was achieved with an AUC of 0.9858 and a sensitivity of 0.9511. Seborrheic keratosis was detected with an AUC of 0.9764 and a sensitivity of 0.9394. Nevertheless, distinguishing irritated seborrheic keratosis from Bowen’s disease remained challenging. Conclusions: Bowen’s disease and seborrheic keratosis could be correctly identified by the evaluated deep learning model on test sets from three different centers, two of which were not involved in training, and AUC scores > 0.97 were obtained. The method proved robust to changes in the staining solution and scanner model. We believe this demonstrates that deep learning algorithms can aid in clinical routine; however, the results should be confirmed by qualified histopathologists.

show abstract

“…Nevertheless, the main characteristic of the medical images datasets is that they are limited in size compared to ImageNet; Which led to the use of these state -of-the-art models as a pre-trained feature extractor [5]. Researchers have used the transfer learning models to classify X-ray images to identify COVID-19 infection [6]- [12] or pneumonia in general [5], [13]- [15], Cardiomegaly [16], osteoarthritis [17], Breast cancer [18], skin cancer [19], [20], tuberculosis detection [21], and disease-free chest [22]. Popular state-of-the-art models such as NASNet [23], ResNet101/152 [24], InceptionResNetV2 [25], and Xception [26] have been used as transfer learning models [12], [13], [18], [27] based on the hypothesis that the ImageNet features can be generalized.…”

Section: Introductionmentioning

confidence: 99%

Pneumonia binary classification using multi-scale feature classification network on chest x-ray images

Mandeel

¹

,

Awad

²

,

Naji³

2022

IJ-AI

View full text Add to dashboard Cite

<span lang="EN-US">According to the world health organization, pneumonia was the cause for 14% of all deaths of children under 5 years old. A computer-aided diagnosis (CADx) system can help the radiologist in the detection of pneumonia in chest radiographs by serving as a second opinion. The typical CADx is based on transfer learning which is done by transferring the learning of feature extraction from one task with plenty of available data to a related task with a scarcity of data. This approach has two limitations which are first, blocking the transferred model from extracting the features that are singular to the new dataset as well as the inability to reduce the complexity of the original model. To address these drawbacks, we proposed a convolutional neural network (CNN) model with low complexity and three paths for feature extraction. The proposed model extracts three different types of features and concatenates them into one feature that provides a good representation for the classes. The proposed model was evaluated on a publicly available dataset. The results showed outperformance by the proposed model compared to the transfer learning models with recall 0.912±0.039, precision 0.942±0.029, F-beta score 0.93, and Cohen’s kappa score 0.740±0.008. </span>

show abstract

Detection of malignant melanoma in H&E-stained images using deep learning techniques

Cited by 17 publications

References 13 publications

Skin cancer diagnosis based on deep transfer learning and sparrow search algorithm

Skin cancer diagnosis based on deep transfer learning and sparrow search algorithm

Evaluation of a Deep Learning Approach to Differentiate Bowen’s Disease and Seborrheic Keratosis

Pneumonia binary classification using multi-scale feature classification network on chest x-ray images

Contact Info

Product

Resources

About