A Novel Fuzzy Multilayer Perceptron (F-MLP) for the Detection of Irregularity in Skin Lesion Border Using Dermoscopic Images

Abder-Rahman Ali, Jingpeng Li, Summrina Kanwal, Guang Yang, Amir Hussain, Sally J. O’SheaFrontiers in Medicine 2020 · DOI

Abstract

Skin lesion border irregularity, representing the "B" feature in the ABCD rule, is a critical factor in melanoma diagnosis. However, its assessment often relies on subjective clinical judgment. While Artificial Neural Networks (ANNs) have advanced supervised learning, they typically fail to account for input ambiguity during training, which can degrade performance. This paper proposes a Fuzzy Multilayer Perceptron (F-MLP) that incorporates input ambiguity to refine the learning process. We introduce a fuzzy gradient descent optimization function and a type-II fuzzy sigmoid activation function. The F-MLP was applied to predict border irregularity after automated segmentation and feature extraction. Results show that the F-MLP outperforms standard neural networks and several state-of-the-art classifiers, though it requires more training time.

Introduction

The diagnosis of melanoma frequently utilizes the ABCD rule: Asymmetry, Border irregularity, Color variegation, and Diameter. Among these, border irregularity is a primary indicator of malignancy. Despite its importance, clinical assessment is notoriously subjective, leading to significant inter-observer variability. Automated, objective methods are therefore essential for reliable diagnosis.

Standard multilayer perceptrons (MLPs) treat all training samples with equal certainty. In medical imaging, however, many samples are ambiguous—occupying a "gray area" between classes. When a standard MLP updates its weights based on these uncertain inputs, it may learn misleading patterns. To address this, we propose a fuzzy-logic-enhanced MLP (F-MLP) designed to handle the inherent uncertainty in dermoscopic images.

Methodology

Image Processing Pipeline

The proposed system follows a multi-stage approach to move from raw pixels to a diagnostic prediction:

  1. Segmentation: Lesions are first segmented from the skin using a U-Net architecture.
  2. Edge Detection: The lesion border is extracted using FuzzEdge, a fuzzy-logic-based edge detector previously developed by the authors.
  3. Feature Extraction: Two sophisticated descriptors quantify the border's irregularity:
  4. Fractal Dimension: Measures the complexity and "raggedness" of the border.
  5. Convexity: The ratio of the lesion's perimeter to its convex hull perimeter.

The F-MLP Architecture

The core innovation is the F-MLP, which integrates fuzzy logic into the neural network's learning and activation mechanisms to dampen the influence of ambiguous data.

The proposed F-MLP architecture incorporating fuzzy clustering and type-II fuzzy activation functions.

Fuzzy Gradient Descent

We modified the standard gradient descent to include an "ambiguity parameter." Using Fuzzy C-Means, we calculate membership degrees ($u$) for each input. The weight update is weighted by the difference in membership degrees ($|u_1 - u_2|^2$). If a sample is highly ambiguous (e.g., $u_1 \approx u_2 \approx 0.5$), its influence on the weight update is minimized, preventing the network from over-adjusting to uncertain data.

Type-II Fuzzy Sigmoid Activation

To better model uncertainty, we replaced the standard sigmoid with a type-II fuzzy sigmoid activation function. Unlike traditional functions that produce a single "crisp" value, type-II fuzzy sets define a "footprint of uncertainty" (FOU) between upper and lower membership boundaries, mirroring the uncertainty inherent in clinical judgment.

Illustration of the transition from a type-I fuzzy set to a type-II fuzzy set, showing the footprint of uncertainty.

Experimental Results

The model was evaluated using the ISIC 2018 dataset. After data augmentation to balance the "regular" and "irregular" border classes, the dataset comprised 310 images.

The F-MLP achieved an accuracy of 95.2%, significantly higher than the 91.9% achieved by a standard MLP. Comparative analysis showed the F-MLP also outperformed Support Vector Machines (87.1%), Logistic Regression (87.1%), and K-Nearest Neighbors (93.5%). While Random Forests achieved a slightly higher accuracy (96.8%) in some tests, the F-MLP demonstrated superior robustness in handling ambiguous samples and required fewer training iterations to reach peak performance.

Conclusion

The proposed F-MLP provides a robust framework for objectively quantifying skin lesion border irregularity. By explicitly modeling ambiguity through fuzzy gradient descent and type-II activation functions, the network achieves high diagnostic accuracy. While the computational cost of training is higher than standard models, the gain in reliability is significant. Future work will explore "Fuzzy Deep Learning" by integrating these concepts directly into convolutional layers.

Summary updated

Abstract

Skin lesion border irregularity, which represents the B feature in the ABCD rule, is considered one of the most significant factors in melanoma diagnosis. Since signs that clinicians rely on in melanoma diagnosis involve subjective judgment including visual signs such as border irregularity, this deems it necessary to develop an objective approach to finding border irregularity. Increased research in neural networks has been carried out in recent years mainly driven by the advances of deep learning. Artificial neural networks (ANNs) or multilayer perceptrons have been shown to perform well in supervised learning tasks. However, such networks usually don't incorporate information pertaining the ambiguity of the inputs when training the network, which in turn could affect how the weights are being updated in the learning process and eventually degrading the performance of the network when applied on test data. In this paper, we propose a fuzzy multilayer perceptron (F-MLP) that takes the ambiguity of the inputs into consideration and subsequently reduces the effects of ambiguous inputs on the learning process. A new optimization function, the fuzzy gradient descent, has been proposed to reflect those changes. Moreover, a type-II fuzzy sigmoid activation function has also been proposed which enables finding the range of performance the fuzzy neural network is able to attain. The fuzzy neural network was used to predict the skin lesion border irregularity, where the lesion was firstly segmented from the skin, the lesion border extracted, border irregularity measured using a proposed measure vector, and using the extracted border irregularity measures to train the neural network. The proposed approach outperformed most of the state-of-the-art classification methods in general and its standard neural network counterpart in particular. However, the proposed fuzzy neural network was more time-consuming when training the network.
BibTeX
@inproceedings{ali2020novel,
  author = {Abder-Rahman Ali and Jingpeng Li and Summrina Kanwal and Guang Yang and Amir Hussain and Sally J. O’Shea},
  title = {{A Novel Fuzzy Multilayer Perceptron (F-MLP) for the Detection of Irregularity in Skin Lesion Border Using Dermoscopic Images}},
  booktitle = {Frontiers in Medicine},
  year = {2020},
  doi = {10.3389/fmed.2020.00297},
}
Figure