Effective Skin Cancer Diagnosis Through Federated Learning and Deep Convolutional Neural Networks

Mabrook S. Al-Rakhami,Salman A. AlQahtani,Abdulaziz Alawwad

DOI: https://doi.org/10.1080/08839514.2024.2364145

IF: 2.777

2024-06-08

Applied Artificial Intelligence

Abstract:Skin cancer is a prevalent type of cancer that affects millions of people globally. However, detecting it can be a challenging task, even for specialized dermatologists. Early detection is crucial for successful treatment, and deep learning techniques, particularly deep convolutional neural networks (DCNNs), have shown tremendous potential in this area. However, achieving high accuracy results requires large volumes of data for training these DCNNs. Since medical organizations and institutions, individually, do not usually have such amounts of information available, and due to the current regulations regarding intellectual property and privacy of medical patient data, it is difficult to share data in a direct way. The primary objective of this work is to overcome this issue through a federated learning approach. We created a privacy-preserving and accurate skin cancer classification system that can assist dermatologists and specialists in making informed patient care decisions. The federated learning DCNNs architecture uses a combination of convolutional and pooling layers to extract relevant features from skin lesion images. It also includes a fully connected layer for classification. To evaluate the proposed architecture, we tested it on three datasets of varying complexity and size. The results demonstrate the applicability of the proposed solution and its efficiency for skin cancer classification.

computer science, artificial intelligence,engineering, electrical & electronic

What problem does this paper attempt to address?

### Problems the Paper Aims to Solve

The main goal of this paper is to develop a privacy-preserving and accurate skin cancer classification system to assist dermatologists in making informed decisions regarding patient care. Specifically, the study aims to achieve this through federated learning methods and deep convolutional neural networks (DCNN).

#### Background and Challenges
- **Importance of Skin Cancer Diagnosis**: Skin cancer is a common type of cancer globally, affecting millions of people each year. Early detection is crucial for successful treatment, but even professional dermatologists may face challenges in diagnosis.
- **Data Sharing Dilemma**: Traditional deep learning methods require a large amount of training data to achieve high accuracy, but individual medical institutions usually do not have sufficient data. Additionally, due to intellectual property and patient privacy regulations, directly sharing data becomes difficult.

#### Methods and Contributions
- **Federated Learning Framework**: The paper proposes a federated learning framework that allows different hospitals (clients) to exchange learning results without directly sharing raw data. This approach addresses the issues of data privacy and data set fragmentation.
- **Deep Convolutional Neural Networks (DCNN)**: DCNN is used for feature extraction and classification of skin lesion images, combining convolutional layers, pooling layers, and fully connected layers to achieve efficient classification.
- **Data Augmentation**: To improve the model's generalization ability, various data augmentation techniques are employed, including image scaling, rotation, and translation.

#### Experimental Validation
- **Datasets**: Experiments were tested on three datasets of different complexity and scale, namely the ISIC 2018 dataset, the PH2 dataset, and a combined dataset.
- **Evaluation Metrics**: The model's performance is evaluated using metrics such as Accuracy, Sensitivity, Precision, and Specificity.

In summary, this paper aims to overcome the privacy issues of data sharing and improve the accuracy and practicality of the skin cancer classification system using federated learning and deep learning technologies.