Very deep super-resolution for efficient cone-beam computed tomographic image restoration

Hwang, Jae Joon; Jung, Yun‐Hoa; Heo, Min-Suk

doi:10.5624/isd.2020.50.4.331

Cited by 8 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hwang et al addressed the storage space and cost issues associated with storing large-capacity CBCT data in dentistry. 19 The network called very deep super-resolution was applied to restore compressed virtual CBCT images using publicly available data. The network enhanced the resolution from low-resolution CBCT images, resulting in superior quality compared to bicubic interpolation.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Improving resolution of panoramic radiographs: super-resolution concept

Çelik,

Mikaeili,

Çelik

2024

Dentomaxillofacial Radiology

View full text Add to dashboard Cite

Objectives Dental imaging plays a key role in the diagnosis and treatment of dental conditions, yet limitations regarding the quality and resolution of dental radiographs sometimes hinder precise analysis. Super-resolution with deep learning refers to a set of techniques used to enhance the resolution of images beyond their original size or quality using deep neural networks instead of traditional image interpolation methods which often result in blurred or pixelated images when attempting to increase resolution. Leveraging advancements in technology, this study aims to enhance the resolution of dental panoramic radiographs, thereby enabling more accurate diagnoses and treatment planning. Methods About 1714 panoramic radiographs from 3 different open datasets are used for training (n = 1364) and testing (n = 350). The state of the art 4 different models is explored, namely Super-Resolution Convolutional Neural Network (SRCNN), Efficient Sub-Pixel Convolutional Neural Network, Super-Resolution Generative Adversarial Network, and Autoencoder. Performances in reconstructing high-resolution dental images from low-resolution inputs with different scales (s = 2, 4, 8) are evaluated by 2 well-accepted metrics Structural Similarity Index (SSIM) and Peak Signal-to-Noise Ratio (PSNR). Results SSIM spans between 0.82 and 0.98 while PSNR are between 28.7 and 40.2 among all scales and models. SRCNN provides the best performance. Additionally, it is observed that performance decreased when images are scaled with higher values. Conclusion The findings highlight the potential of super-resolution concepts to significantly improve the quality and detail of dental panoramic radiographs, thereby contributing to enhanced interpretability.

show abstract

Section: Discussionmentioning

confidence: 99%

“… 18 Hwang et al addressed the storage space and cost issues associated with storing large-capacity CBCT data in dentistry. 19 …”

Section: Introductionmentioning

confidence: 99%

Improving resolution of panoramic radiographs: super-resolution concept

Çelik,

Mikaeili,

Çelik

2024

Dentomaxillofacial Radiology

View full text Add to dashboard Cite

show abstract

“…Additionally, network design does not allow for a high learning rate, leading to high computing resource costs [ 32 , 33 ]. Therefore, this study focused on the characteristics of VDSR, which exhibits high performance in improving image quality with its deep layers [ 32 , 33 , 34 ]. Compared with existing networks, VDSR is designed with 20 deep layers, enabling superior image quality improvement effects based on its excellent learning performance, allowing for the application of a high learning rate for smoother learning [ 32 , 34 ].…”

Section: Discussionmentioning

confidence: 99%

“…Therefore, this study focused on the characteristics of VDSR, which exhibits high performance in improving image quality with its deep layers [ 32 , 33 , 34 ]. Compared with existing networks, VDSR is designed with 20 deep layers, enabling superior image quality improvement effects based on its excellent learning performance, allowing for the application of a high learning rate for smoother learning [ 32 , 34 ]. Although VDSR’s deep structure presents another limitation by complicating the learning process, this study aimed to implement single image super resolution (SISR) for a single trigeminal nerve in a single MRI slice among multiple MRI slices.…”

Section: Discussionmentioning

confidence: 99%

Deep Learning Super-Resolution Technique Based on Magnetic Resonance Imaging for Application of Image-Guided Diagnosis and Surgery of Trigeminal Neuralgia

Hwang,

Park,

Kang

et al. 2024

Life

View full text Add to dashboard Cite

This study aimed to implement a deep learning-based super-resolution (SR) technique that can assist in the diagnosis and surgery of trigeminal neuralgia (TN) using magnetic resonance imaging (MRI). Experimental methods applied SR to MRI data examined using five techniques, including T2-weighted imaging (T2WI), T1-weighted imaging (T1WI), contrast-enhancement T1WI (CE-T1WI), T2WI turbo spin–echo series volume isotropic turbo spin–echo acquisition (VISTA), and proton density (PD), in patients diagnosed with TN. The image quality was evaluated using the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM). High-quality reconstructed MRI images were assessed using the Leksell coordinate system in gamma knife radiosurgery (GKRS). The results showed that the PSNR and SSIM values achieved by SR were higher than those obtained by image postprocessing techniques, and the coordinates of the images reconstructed in the gamma plan showed no differences from those of the original images. Consequently, SR demonstrated remarkable effects in improving the image quality without discrepancies in the coordinate system, confirming its potential as a useful tool for the diagnosis and surgery of TN.

show abstract

“… 6 For medical imaging, this method is being actively studied for noise reduction and resolution enhancement in MRI and computed tomography. 7 8 However, few studies have explored ways of generating different tissue contrast images based on other pulse sequence images.…”

Section: Introductionmentioning

confidence: 99%

Synthesis of T2-weighted images from proton density images using a generative adversarial network in a temporomandibular joint magnetic resonance imaging protocol

Lee

Choi

et al. 2022

Imaging Sci Dent

View full text Add to dashboard Cite

Purpose This study proposed a generative adversarial network (GAN) model for T2-weighted image (WI) synthesis from proton density (PD)-WI in a temporomandibular joint (TMJ) magnetic resonance imaging (MRI) protocol. Materials and Methods From January to November 2019, MRI scans for TMJ were reviewed and 308 imaging sets were collected. For training, 277 pairs of PD- and T2-WI sagittal TMJ images were used. Transfer learning of the pix2pix GAN model was utilized to generate T2-WI from PD-WI. Model performance was evaluated with the structural similarity index map (SSIM) and peak signal-to-noise ratio (PSNR) indices for 31 predicted T2-WI (pT2). The disc position was clinically diagnosed as anterior disc displacement with or without reduction, and joint effusion as present or absent. The true T2-WI-based diagnosis was regarded as the gold standard, to which pT2-based diagnoses were compared using Cohen’s ĸ coefficient. Results The mean SSIM and PSNR values were 0.4781(±0.0522) and 21.30(±1.51) dB, respectively. The pT2 protocol showed almost perfect agreement (ĸ=0.81) with the gold standard for disc position. The number of discordant cases was higher for normal disc position (17%) than for anterior displacement with reduction (2%) or without reduction (10%). The effusion diagnosis also showed almost perfect agreement (ĸ=0.88), with higher concordance for the presence (85%) than for the absence (77%) of effusion. Conclusion The application of pT2 images for a TMJ MRI protocol useful for diagnosis, although the image quality of pT2 was not fully satisfactory. Further research is expected to enhance pT2 quality.

show abstract

Very deep super-resolution for efficient cone-beam computed tomographic image restoration

Cited by 8 publications

References 12 publications

Improving resolution of panoramic radiographs: super-resolution concept

Improving resolution of panoramic radiographs: super-resolution concept

Deep Learning Super-Resolution Technique Based on Magnetic Resonance Imaging for Application of Image-Guided Diagnosis and Surgery of Trigeminal Neuralgia

Synthesis of T2-weighted images from proton density images using a generative adversarial network in a temporomandibular joint magnetic resonance imaging protocol

Contact Info

Product

Resources

About