An Evaluation of Image Enhancement Techniques for Nailfold Capillary Skeletonisation

Improved nested U-structure for accurate nailfold capillary segmentation

Changes in the structure and function of nailfold capillaries may be indicators of numerous diseases. Noninvasive diagnostic tools are commonly used for the extraction of morphological information from segmented nailfold capillaries to study physiological and pathological changes therein. However, current segmentation methods for nailfold capillaries cannot accurately separate capillaries from the background, resulting in issues such as unclear segmentation boundaries. Therefore, improving the accuracy of nailfold capillary segmentation is necessary to facilitate more efficient clinical diagnosis and research. Herein, we propose a nailfold capillary image segmentation method based on a U2-Net backbone network combined with a Transformer structure. This method integrates the U2-Net and Transformer networks to establish a decoder-encoder network, which inserts Transformer layers into the nested two-layer U-shaped architecture of the U2-Net. This structure effectively extracts multiscale features within stages and aggregates multilevel features across stages to generate high-resolution feature maps. The experimental results demonstrate an overall accuracy of 98.23 %, a Dice coefficient of 88.56 %, and an IoU of 80.41 % compared to the ground truth. Furthermore, our proposed method improves the overall accuracy by approximately 2 %, 3 %, and 5 % compared to the original U2-Net, Res-Unet, and U-Net, respectively. These results indicate that the Transformer-U2Net network performs well in nailfold capillary image segmentation and provides more detailed and accurate information on the segmented nailfold capillary structure, which may aid clinicians in the more precise diagnosis and treatment of nailfold capillary-related diseases.

Analysis of microvascular pattern in diabetes mellitus condition using the nailfold capillaroscopy images

Diabetes is often considered a vascular disease due to its impact on blood vessels, it is a complex condition with various metabolic and autoimmune factors involved. One of the long term comorbidities of diabetes includes microvascular complications. The microvascular complications can be analyzed using the Nailfold capillaroscopy, a non-invasive technique that allows for the visualization and analysis of capillaries in the proximal nailfold area. Using advanced video capillaroscopy with high magnification, capillary images can be captured from and processed to analyze their morphology. The capillary images of normal group and diabetic group are acquired from 118 participants using nailfold capillaroscopy and the obtained images are preprocessed using image processing filters. The identification and segmentation of the capillaries are the challenges to be addressed in the processing of the images. Hence segmentation of capillaries is done using morphological operations, thresholding and convolutional neural networks. The performance of the filters and segmentation methods are evaluated using Mean Square Error (MSE), Peak signal to Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), Jaccard Index and Sorensen coefficient. By analyzing the morphological features namely the capillary diameter, density, distribution, presence of hemorrhage and the shape of the capillaries from both the groups, the capillary changes associated with diabetic condition were studied. It was found that the non diabetic participants considered in this study has capillary diameter in the range of 8-14 µm and the capillary density in the range of 10-30 capillaries per mm2 whereas the diabetic participants has capillary diameter greater than 30 µm and the capillary density is less than 10 capillaries per mm2. In addition to capillary density and diameter, the presence of hemorrhage, the orientation and distribution of the capillaries are also considered to differentiate the diabetic group from the non diabetic group. The classification of the participants are validated with the clinical history of the participants.

Automated Stabilization, Enhancement and Capillaries Segmentation in Videocapillaroscopy

Oral capillaroscopy is a critical and non-invasive technique used to evaluate microcirculation. Its ability to observe small vessels in vivo has generated significant interest in the field. Capillaroscopy serves as an essential tool for diagnosing and prognosing various pathologies, with anatomic-pathological lesions playing a crucial role in their progression. Despite its importance, the utilization of videocapillaroscopy in the oral cavity encounters limitations due to the acquisition setup, encompassing spatial and temporal resolutions of the video camera, objective magnification, and physical probe dimensions. Moreover, the operator's influence during the acquisition process, particularly how the probe is maneuvered, further affects its effectiveness. This study aims to address these challenges and improve data reliability by developing a computerized support system for microcirculation analysis. The designed system performs stabilization, enhancement and automatic segmentation of capillaries in oral mucosal video sequences. The stabilization phase was performed by means of a method based on the coupling of seed points in a classification process. The enhancement process implemented was based on the temporal analysis of the capillaroscopic frames. Finally, an automatic segmentation phase of the capillaries was implemented with the additional objective of quantitatively assessing the signal improvement achieved through the developed techniques. Specifically, transfer learning of the renowned U-net deep network was implemented for this purpose. The proposed method underwent testing on a database with ground truth obtained from expert manual segmentation. The obtained results demonstrate an achieved Jaccard index of 90.1% and an accuracy of 96.2%, highlighting the effectiveness of the developed techniques in oral capillaroscopy. In conclusion, these promising outcomes encourage the utilization of this method to assist in the diagnosis and monitoring of conditions that impact microcirculation, such as rheumatologic or cardiovascular disorders.

Image enhancement effect on inter and intra-observer reliability of nailfold capillary assessment

BACKGROUND

Nailfold capillaroscopy (NC) is a diagnostic imaging technique that is used to assess the blood capillary network in the nailfold area. NC is routinely used for patients with microcirculation problems, such as systemic sclerosis and other connective tissue diseases. Experts commonly use subjective evaluation as a reference point in images of nailfold video capillaroscopy, so it is important to reduce the inherent ambiguities in human judgment and diagnosis. Image quality is an important factor that affects measurement error and assessment time of NC images.

OBJECTIVE

In this study, a new image enhancement technique was introduced and evaluated subjectively.

METHODS

In total, 475 nailfold video capillaroscopy images from 18 healthy subjects and 41 systemic lupus erythematosus patients were used. The images were randomly divided into two sets, one each with 275 and 200. Eight independent observers who were familiar with the capillaroscopy technique participated in this study. The set of 275 images was evaluated by three observers with the forced-choice pairwise comparison method. Elliptic broken line (EBL) was used to count the number of capillaries. The intra- and inter-observer reliability of the original and enhanced images was evaluated on 200 images by five observers.

RESULT

Except for eight images, all observers preferred the enhanced images in the visual quality comparison method. The intra-class correlation coefficient (ICC) of intra- and inter-observer reliability increased from 0.76-0.84 to 0.82-0.89, respectively, when using the enhancement method.

CONCLUSION

By improving the image quality, more capillary details will be visible, and an observer can document more details that may not be visible in the original image and can do so more efficiently.