A novel visible and near-infrared hyperspectral imaging platform for automated breast-cancer detection

BACKGROUND

Breast cancer is a leading cause of cancer-related deaths among women worldwide. Early and accurate detection is crucial for improving patient outcomes. Our study utilizes Visible and Near-Infrared Hyperspectral Imaging (VIS-NIR HSI), a promising non-invasive technique, to detect cancerous regions in ex-vivo breast specimens based on their hyperspectral response.

METHODS

In this paper, we present a novel HSI platform integrated with fuzzy c-means clustering for automated breast cancer detection. We acquire hyperspectral data from breast tissue samples, and preprocess it to reduce noise and enhance hyperspectral features. Fuzzy c-means clustering is then applied to segment cancerous regions based on their spectral characteristics.

RESULTS

Our approach demonstrates promising results. We evaluated the quality of the clustering using metrics like Silhouette Index (SI), Davies-Bouldin Index (DBI), and Calinski-Harabasz Index (CHI). The clustering metrics results revealed an optimal number of 6 clusters for breast tissue classification, and the SI values ranged from 0.68 to 0.72, indicating well-separated clusters. Moreover, the CHI values showed that the clusters were well-defined, and the DBI values demonstrated low cluster dispersion. Additionally, the sensitivity, specificity, and accuracy of our system were evaluated on a dataset of breast tissue samples. We achieved an average sensitivity of 96.83%, specificity of 93.39%, and accuracy of 95.12%. These results indicate the effectiveness of our HSI-based approach in distinguishing cancerous and non-cancerous regions.

CONCLUSIONS

The paper introduces a robust hyperspectral imaging platform coupled with fuzzy c-means clustering for automated breast cancer detection. The clustering metrics results support the reliability of our approach in effectively segmenting breast tissue samples. In addition, the system shows high sensitivity and specificity, making it a valuable tool for early-stage breast cancer diagnosis. This innovative approach holds great potential for improving breast cancer screening and, thereby, enhancing our understanding of the disease and its detection patterns.

Fully connected network with multi-scale dilation convolution module in evaluating atrial septal defect based on MRI segmentation

BACKGROUND AND OBJECTIVE

Atrial septal defect (ASD) is a common congenital heart disease. During embryonic development, abnormal atrial septal development leads to pores between the left and right atria. ASD accounts for the largest proportion of congenital heart disease. Therefore, the design and implementation of an ASD intelligent auxiliary segmentation system based on deep learning segmentation of the atria has very important practical significance, which we aim to achieve in this paper.

METHODS

This study proposes a multi-scale dilated convolution module, which is composed of three parallel dilated convolutions with different expansion coefficients. The original FCN network usually adopts bilinear interpolation or deconvolution methods when upsampling, both of which lead to information loss to a certain extent. In order to make up for the loss of information, it is expected that the final segmentation result can be directly connected to the deep features in the cardiac MRI. This study uses a dense upsampling convolution module, and in order to obtain the shallow position information, the original FCN jump connection module is still retained. In this research, a deep convolutional neural network for multi-scale feature extraction is designed through the multi-scale expansion convolution module. At the same time, this paper also implements two traditional machine learning segmentation methods (K-means and Watershed algorithms) and a deep learning algorithm (U-net) for comparison.

RESULTS

The intelligent auxiliary segmentation algorithm for atrial images proposed in this framework based on multi-scale expansion convolution and adversarial learning can achieve superior results. Among them, the segmentation algorithm based on multi-scale expansion convolution can extract the associated features of pixels in multiple ranges, and can obtain deeper feature information when using a limited downsampling layer. According to the experimental results of the multi-scale expanded convolutional network on the data set, the Proportion of Greater Contour (PGC) index of the multi-scale expanded convolutional network is 98.78, the value of Average Perpendicular Distance (ADP) is 1.72mm, and the value of Overlapping Dice Metric (ODM) is 0.935, which are higher than other models.

CONCLUSION

The experimental results show that compared with other segmentation models, the model based on multi-scale expansion convolution has significantly improved the accuracy of segmentation. Our technique will be able to assist in the segmentation of ASD, evaluation of the extent of the defect and enhance surgical planning via atrial septal occlusion.

Vessel Segmentation in Retinal Images Using Multi-scale Line Operator and K-Means Clustering

Detecting blood vessels is a vital task in retinal image analysis. The task is more challenging with the presence of bright and dark lesions in retinal images. Here, a method is proposed to detect vessels in both normal and abnormal retinal fundus images based on their linear features. First, the negative impact of bright lesions is reduced by using K-means segmentation in a perceptive space. Then, a multi-scale line operator is utilized to detect vessels while ignoring some of the dark lesions, which have intensity structures different from the line-shaped vessels in the retina. The proposed algorithm is tested on two publicly available STARE and DRIVE databases. The performance of the method is measured by calculating the area under the receiver operating characteristic curve and the segmentation accuracy. The proposed method achieves 0.9483 and 0.9387 localization accuracy against STARE and DRIVE respectively.