Abstract:Hyperspectral imaging (HSI) has become a very compelling technique in different scientific areas; indeed, many researchers use it in the fields of remote sensing, agriculture, forensics, and medicine. In the latter, HSI plays a crucial role as a diagnostic support and for surgery guidance. However, the computational effort in elaborating hyperspectral data is not trivial. Furthermore, the demand for detecting diseases in a short time is undeniable. In this paper, we take up this challenge by parallelizing three machine-learning methods among those that are the most intensively used: Support Vector Machine (SVM), Random Forest (RF), and eXtreme Gradient Boosting (XGB) algorithms using the Compute Unified Device Architecture (CUDA) to accelerate the classification of hyperspectral skin cancer images. They all showed a good performance in HS image classification, in particular when the size of the dataset is limited, as demonstrated in the literature. We illustrate the parallelization techniques adopted for each approach, highlighting the suitability of Graphical Processing Units (GPUs) to this aim. Experimental results show that parallel SVM and XGB algorithms significantly improve the classification times in comparison with their serial counterparts.

What problem does this paper attempt to address?

### Problems the paper attempts to solve This paper aims to accelerate the hyperspectral skin - cancer image classification through parallel machine - learning methods. Specifically, the authors focus on the following issues: 1. **Computational complexity**: Hyperspectral Imaging (HSI) plays an important role in medical diagnosis, but the computational complexity of processing hyperspectral data is high and requires a large amount of computing resources and time. 2. **Requirement for rapid diagnosis**: Early detection of skin cancer is crucial for improving treatment outcomes and patient survival rates. Therefore, a fast and reliable diagnosis system is required to meet this need. 3. **Limitations of existing methods**: Although existing machine - learning and deep - learning methods perform well in hyperspectral image classification, there are still performance bottlenecks when dealing with large - scale data sets. To solve the above problems, the authors selected three commonly used machine - learning methods: Support Vector Machine (SVM), Random Forest (RF) and eXtreme Gradient Boosting (XGB), and parallelized these algorithms through NVIDIA's Compute Unified Device Architecture (CUDA) framework to accelerate the classification of hyperspectral skin - cancer images. ### Main contributions 1. **Parallelization method**: It describes in detail how to parallelize SVM, RF and XGB methods on GPU, including specific parallelization techniques and optimization strategies. 2. **Comparison of different devices**: The parallelized algorithms are implemented on different NVIDIA GPU devices, including GeForce RTX 2080, GeForce RTX 4090 and a Tesla A16 cluster consisting of five nodes, and their performance differences in hyperspectral skin - cancer image classification are compared. 3. **Performance improvement**: The experimental results show that compared with the serial versions, the parallel versions of SVM and XGB achieve a speed - up of 130 times and 1.4 times respectively, verifying the effectiveness of GPU in accelerating the medical diagnosis process. ### Conclusion Through parallelization techniques, the authors have successfully improved the efficiency of hyperspectral skin - cancer image classification, providing technical support for fast and accurate skin - cancer diagnosis. Future research directions may include further optimizing parallelization algorithms and exploring more parallel computing techniques suitable for hyperspectral image processing.

Acceleration of Hyperspectral Skin Cancer Image Classification through Parallel Machine-Learning Methods

Parallel Implementations Assessment of a Spatial-Spectral Classifier for Hyperspectral Clinical Applications

Edge and cloud computing approaches in the early diagnosis of skin cancer with attention-based vision transformer through hyperspectral imaging

High Performance Hyperspectral Image Classification using Graphics Processing Units

Parallel Adaptive Sparsity-Constrained NMF Algorithm for Hyperspectral Unmixing.

GPU-based Acceleration of the Hyperspectral Band Selection by SNR Estimation Using Wavelet Transform

Supervised Machine Learning Methods and Hyperspectral Imaging Techniques Jointly Applied for Brain Cancer Classification

GPU Implementation of an Automatic Target Detection and Classification Algorithm for Hyperspectral Image Analysis

Blood Cancer Diagnosis Using Hyperspectral Imaging Assisted with Forward Searching Method and Machine Learning

Non-Invasive Skin Cancer Diagnosis Using Hyperspectral Imaging for In-Situ Clinical Support

GPU-Based Parallel Computing Methods for Medical Photoacoustic Image Reconstruction

Accelerating Support Vector Machine Learning With Gpu-Based Mapreduce

Combining hyperspectral imaging techniques with deep learning to aid in early pathological diagnosis of melanoma

Evaluation of Preprocessing Methods on Independent Medical Hyperspectral Databases to Improve Analysis

Classification of Hyperspectral Imagery Based on Spectral Gradient, SVM and Spatial Random Forest

GPU-based key-frame selection of pulmonary ultrasound images to detect COVID-19

Advancing laser ablation assessment in hyperspectral imaging through machine learning

Parallel Multiple Instance Learning for Extremely Large Histopathology Image Analysis.

Hyperspectral Imaging for Glioblastoma Surgery: Improving Tumor Identification Using a Deep Spectral-Spatial Approach

Optimized Learning for X-Ray Image Classification for Multi-Class Disease Diagnoses with Accelerated Computing Strategies

Hyperspectral imaging for tissue classification in glioblastoma tumor patients: a deep spectral-spatial approach