Image Processing for Enhanced Learning (IMPEL) Lab. Our group carries out the research for artificial intelligence based on the digital image and signal datasets.

Prior-aware and Context-guided Group Sampling for Active Probabilistic Subsampling

Physics-informed optimization of saturation-transfer MRI protocols using non-differentiable Bloch models

Cyclic Conditional Diffusion Models for CT-to-MR Synthetic Image Segmentation with Misaligned Image Pairs

An Unsupervised Metal Artifact Reduction Model via Physics-Based Metal-Aware Networks in CT Images

Integrating Depth Priors Into PixelNeRF for Enhanced Global Guidance

Artificial Intelligence Model for Detection of Colorectal Cancer on Routine Abdominopelvic CT Examinations: A Training and External-Testing Study

Self-supervised learning for denoising of multidimensional MRI data

A Noise Robust Image Reconstruction using Slice Aware Cycle Interpolator Network for Parallel Imaging in MRI

Attention Guided CAM: Visual Explanations of Vision Transformer Guided by Self-Attention

Spectrum Translation for Refinement of Image Generation (STIG) Based on Contrastive Learning and Spectral Filter Profile

RenseNet: A Deep Learning Network Incorporating Residual and Dense Blocks with Edge Conservative Module to Improve Small Lesion Classification

Unsupervised Motion Artifact Correction of Turbo Spin-Echo MRI using Deep Image Prior

An Endodontic forecasting model based on the analysis of preoperative dental radiographs: A pilot study on an endodontic predictive deep neural network

Spatial Feature Conservation Networks (SFCNs) for Dilated Convolutions to Improve Breast Cancer Segmentation from DCE-MRI

Computed Tomography Vertebral Segmentation from Multi-Vendor Scanner Data

Deep Learning-Based Automatic Segmentation of Mandible and Maxilla in Multi-Center CT Images

Deep Neural Network with Consistency Regularization of Multi-Output Channels to Improve Tumor Detection and Delineation

Closing the Gap between Deep Neural Network Modeling and Decision‐Making Metric via Generalized Loss Function

A Dual Domain Network for MRI Reconstruction Using Gabor Loss

Modified U‐Net (mU‐Net) with Incorporation of Object‐Dependent High Level Features for Improved Liver and Liver‐Tumor Segmentation in CT Images

Machine Learning Techniques for Biomedical Image Segmentation: An Overview of Technical Aspects and Introduction to State‐of‐Art Applications

A locally segmented reconstruction method for parallel imaging

Automatic Polyp Recognition in Colonoscopy Images Using Deep Learning and Two-Stage Pyramidal Feature Prediction

A Multi‐Contrast Imaging Method Using Steady‐State Free Precession with Alternating RF Flip Angles

Retrospective motion gating in cardiac MRI using a simultaneously acquired navigator

Non‐Contrast‐Enhanced Peripheral MR Angiography using Velocity‐Selective Excitation

Self‐Gated Cardiac Cine Imaging Using Phase Information

Multi‐slice imAGe generation using intra‐slice parallel imaging and inter‐slice shifting (MAGGULLI)

Sliding TOF: Sliding time of flight MR angiography using a dynamic image reconstruction method

Isotropic diffusion weighting for measurement of high‐resolution apparent diffusion coefficient map using a single radial scan in MRI

A Novel Detection Method of Periodically Moving Region in Radial MRI

Deep Probabilistic Multishell Undersampling for Diffusion Spectrum Imaging

Self-supervised Denoising Model for 7T Angiography with Z-Directional Artery Focused DDPM

Self-supervised deep learning model for denoising and distortion correction in accelerated echo planar imaging

Optimization of acquisition schedules and correction of B0 and B1 field inhomogeneities in saturation transfer MR fingerprinting (ST-MRF)

DViTPolyp: Small Polyp Detection in Colonoscopy Videos based on Deep Vision Transformer

Multi-Slice-Aware Denoising Model for 7T MR Angiography

A Dual-Flip Angle bSSFP Imaging Method for Proton Resonance Frequency-Based Thermometry

A Bounding-Box Regression Model for Colorectal Tumor Detection in CT Images Via Two Contrary Networks

A Localized Reconstruction Method for Parallel Imaging in MRI

Modified U‐Net for High‐Resolution High‐Level Feature Extraction and Its Application to Liver‐Tumor Segmentation

A Fast Multi‐Contrast Knee Imaging Method Using a Hybrid bSSFP Pulse Sequence

A Retrospective Cardiac Gating Method using Simultaneously Acquired Navigator

A New High‐Resolution Perfusion‐Weighted Imaging Technique without exogenous tracer and Its Quantitative Model

Measurement of the Blood Pressure Using the Phase Difference in MRI

Peripheral MR Angiography using velocity‐sensitive gradient‐echo technique

Respiratory Motion Estimation within Echo Train of Fast Spin Echo

Reduction of RF Pulse Duration using Dephased Transition

Learning‐based Reconstruction using Artificial Neural Network for Higher Acceleration

Correction of Chemical‐Shift Ghost Artifact in Blipped Controlled Aliasing Parallel Imaging

Self‐Gated CMR (Cardiac MRI) Using Phase Information

A New Self‐Gating Method for Cardiac MRI Using Phase Information

A New bSSFP‐MRI Pulse Sequence for Interleaved Acquisition of k‐space Lines from Different Slice

Multi‐slice imAGe Generation Using intra‐slice paraLLel imaging and Inter‐slice shifting (MAGGULLI)

Multi‐Contrast Reconstruction using Neural Network for Higher Acceleration

Peripheral MR Angiography using Fourier Velocity Encoding and Dynamic Reconstruction

Multiband Imaging Method for Metal Artifact Correction with 3D Multi‐Spectral Imaging

High‐Resolution Perfusion‐Weighted Imaging Without Tagging Pulses

Ordering of Multiple Diffusion Gradient Directions for High Resolution ADC Maps Using Golden Angle

Fast Acquisition of a High Resolution Apparent Diffusion Coefficient Map Using Radial Sampling

Magnetic resonance imaging apparatus and method of generating magnetic resonance image

Method and apparatus for generating magnetic resonance image

Magnetic resonance imaging apparatus and method

Magnetic resonance imaging (mri) apparatus and method of generating mr image

Method for acquiring water-fat separation image, and magnetic resonance imaging apparatus therefor

Magnetic resonance imaging apparatus and method of reconstructing mr image by using neural network

Magnetic resonance imaging apparatus and method of controlling the same

Magnetic resonance imaging apparatus and image processing method thereof

Resonance imaging apparatus and diffusion weighted image acquiring method thereof

Modified U-Net (mU-Net) with Incorporation of Object-Dependent High Level Features for Improved Liver and Liver-Tumor Segmentation in CT Images

Machine Learning Techniques for Biomedical Image Segmentation: An Overview of Technical Aspects and Introduction to State-of-Art Applications

Automatic polyp recognition in colonoscopy images using deep learning and two-stage pyramidal feature prediction

An endodontic forecasting model based on the analysis of preoperative dental radiographs: a pilot study on an endodontic predictive deep neural network

Closing the Gap between Deep Neural Network Modeling and Biomedical Decision-Making Metrics in Segmentation via Adaptive Loss Functions

Self‐supervised learning for denoising of multidimensional MRI data

Learning-based reconstruction using artificial neural network for higher acceleration

Self‐gated cardiac cine imaging using phase information

Unsupervised motion artifact correction of turbo spin‐echo MRI using deep image prior

A dual domain network for MRI reconstruction using gabor loss

Computed tomography vertebral segmentation from multi-vendor scanner data

Magnetic resonance imaging (MRI) apparatus and method of generating MR image

A multicontrast imaging method using steady‐state free precession with alternating RF flip angles

HyunWook Park

Lequan Yu

Liyue Shen

Lei Xing

Hyunseok Seo

Segmentation of livers and liver tumors is one of the most important steps in radiation therapy of hepatocellular carcinoma. The segmentation task is often done manually, making it tedious, labor intensive, and subject to intra-/inter- operator variations. While various algorithms for delineating organ-at-risks (OARs) and tumor targets have been proposed, automatic segmentation of livers and liver tumors remains intractable due to their low tissue contrast with respect to the surrounding organs and their deformable shape in CT images. The U-Net has gained increasing popularity recently for image analysis tasks and has shown promising results. Conventional U-Net architectures, however, suffer from three major drawbacks. First, skip connections allow for the duplicated transfer of low resolution information in feature maps to improve efficiency in learning, but this often leads to blurring of extracted image features …

In recent years, significant progress has been made in developing more accurate and efficient machine learning algorithms for segmentation of medical and natural images. In this review article, we highlight the imperative role of machine learning algorithms in enabling efficient and accurate segmentation in the field of medical imaging. We specifically focus on several key studies pertaining to the application of machine learning methods to biomedical image segmentation. We review classical machine learning algorithms such as Markov random fields, k ‐means clustering, random forest, etc. Although such classical learning models are often less accurate compared to the deep‐learning techniques, they are often more sample efficient and have a less complex structure. We also review different deep‐learning architectures, such as the artificial neural networks (ANNs), the convolutional neural networks (CNNs), and …

Polyp recognition in colonoscopy images is crucial for early colorectal cancer detection and treatment. However, the current manual review requires undivided concentration of the gastroenterologist and is prone to diagnostic errors. In this article, we present an effective, two-stage approach called PLPNet, where the abbreviation “PLP” stands for the word “polyp,” for automated pixel-accurate polyp recognition in colonoscopy images using very deep convolutional neural networks (CNNs). Compared to hand-engineered approaches and previous neural network architectures, our PLPNet model improves recognition accuracy by adding a polyp proposal stage that predicts the location box with polyp presence. Several schemes are proposed to ensure the model's performance. First of all, we construct a polyp proposal stage as an extension of the faster R-CNN, which performs as a region-level polyp detector to …

Vision Transformer(ViT) is one of the most widely used models in the computer vision field with its great performance on various tasks. In order to fully utilize the ViT-based architecture in various applications, proper visualization methods with a decent localization performance are necessary, but these methods employed in CNN-based models are still not available in ViT due to its unique structure. In this work, we propose an attention-guided visualization method applied to ViT that provides a high-level semantic explanation for its decision. Our method selectively aggregates the gradients directly propagated from the classification output to each self-attention, collecting the contribution of image features extracted from each location of the input image. These gradients are additionally guided by the normalized self-attention scores, which are the pairwise patch correlation scores. They are used to supplement the gradients on the patch-level context information efficiently detected by the self-attention mechanism. This approach of our method provides elaborate high-level semantic explanations with great localization performance only with the class labels. As a result, our method outperforms the previous leading explainability methods of ViT in the weakly-supervised localization task and presents great capability in capturing the full instances of the target class object. Meanwhile, our method provides a visualization that faithfully explains the model, which is demonstrated in the perturbation comparison test.

Deep learning is becoming an indispensable tool for imaging applications, such as image segmentation, classification, and detection. In this work, we reformulate a standard deep learning problem into a new neural network architecture with multi-output channels, which reflects different facets of the objective, and apply the deep neural network to improve the performance of image segmentation. By adding one or more interrelated auxiliary-output channels, we impose an effective consistency regularization for the main task of pixelated classification (i.e., image segmentation). Specifically, multi-output-channel consistency regularization is realized by residual learning via additive paths that connect main-output channel and auxiliary-output channels in the network. The method is evaluated on the detection and delineation of lung and liver tumors with public data. The results clearly show that multi-output-channel …

Introduction This study aimed to evaluate the use of deep convolutional neural network (DCNN) algorithms to detect clinical features and predict the three-year outcome of endodontic treatment on preoperative periapical radiographs. Methods A database of single-root premolars that received endodontic treatment or retreatment by endodontists with presence of three-year outcome was prepared ( n = 598). We constructed a 17-layered DCNN with a self-attention layer (Periapical Radiograph Explanatory System with Self-Attention Network [PRESSAN-17]), and the model was trained, validated, and tested to 1) detect 7 clinical features, that is, full coverage restoration, presence of proximal teeth, coronal defect, root rest, canal visibility, previous root filling, and periapical radiolucency and 2) predict the three-year endodontic prognosis by analyzing preoperative periapical radiographs as an input. During the …

Deep learning is becoming an indispensable tool for various tasks in science and engineering. A critical step in constructing a reliable deep learning model is the selection of a loss function, which measures the discrepancy between the network prediction and the ground truth. While a variety of loss functions have been proposed in the literature, a truly optimal loss function that maximally utilizes the capacity of neural networks for deep learning-based decision-making has yet to be established. Here, we devise a generalized loss function with functional parameters determined adaptively during model training to provide a versatile framework for optimal neural network-based decision-making in small target segmentation. The method is showcased by more accurate detection and segmentation of lung and liver cancer tumors as compared with the current state-of-the-art. The proposed formalism opens new …

Sophisticated segmentation of the craniomaxillofacial bones (the mandible and maxilla) in computed tomography (CT) is essential for diagnosis and treatment planning for craniomaxillofacial surgeries. Conventional manual segmentation is time-consuming and challenging due to intrinsic properties of craniomaxillofacial bones and head CT such as the variance in the anatomical structures, low contrast of soft tissue, and artifacts caused by metal implants. However, data-driven segmentation methods, including deep learning, require a large consistent dataset, which creates a bottleneck in their clinical applications due to limited datasets. In this study, we propose a deep learning approach for the automatic segmentation of the mandible and maxilla in CT images and enhanced the compatibility for multi-center datasets. Four multi-center datasets acquired by various conditions were applied to create a scenario where the model was trained with one dataset and evaluated with the other datasets. For the neural network, we designed a hierarchical, parallel and multi-scale residual block to the U-Net (HPMR-U-Net). To evaluate the performance, segmentation with in-house dataset and with external datasets from multi-center were conducted in comparison to three other neural networks: U-Net, Res-U-Net and mU-Net. The results suggest that the segmentation performance of HPMR-U-Net is comparable to that of other models, with superior data compatibility.

Purpose To develop a fast denoising framework for high‐dimensional MRI data based on a self‐supervised learning scheme, which does not require ground truth clean image. Theory and Methods Quantitative MRI faces limitations in SNR, because the variation of signal amplitude in a large set of images is the key mechanism for quantification. In addition, the complex non‐linear signal models make the fitting process vulnerable to noise. To address these issues, we propose a fast deep‐learning framework for denoising, which efficiently exploits the redundancy in multidimensional MRI data. A self‐supervised model was designed to use only noisy images for training, bypassing the challenge of clean data paucity in clinical practice. For validation, we used two different datasets of simulated magnetization transfer contrast MR fingerprinting (MTC‐MRF) dataset and in vivo DWI image dataset to show the …

A long imaging time has been regarded as a major drawback of MRI, and many techniques have been proposed to overcome this problem. Parallel imaging (PI) and compressed sensing (CS) techniques utilize different sensitivity of multi-channel RF coils and sparsity of signal in a certain domain to remove aliasing artifacts that are generated by subsampling, respectively. In this study, an artificial neural networks (ANN) are applied to MR reconstruction to reduce imaging time, and it is shown that the ANN model has a potential to be comparable to PI and CS.

Purpose To obtain multiphase cardiac cine images with high resolution, a novel self‐gating method for both cardiac and respiratory motions is proposed. Methods The proposed method uses the phase of projection data obtained from a separate axial slice to measure cardiac and respiratory motion, after the acquisition of every k‐space line in the image plane. Cardiac motion is estimated from the phase of the projection data passing through the aorta, which is amplified by superior–inferior directional bipolar gradients, whereas respiratory motion is estimated from the phase of the left–right directional projection data of the abdomen. To verify the proposed self‐gating method, a simulation and in vivo steady state free precession cardiac imaging were performed. Results The proposed method provides high resolution multiphase cardiac cine images. Using the proposed self‐gating method, the phase variation of …

Currently, image generation and synthesis have remarkably progressed with generative models. Despite photo-realistic results, intrinsic discrepancies are still observed in the frequency domain. The spectral discrepancy appeared not only in generative adversarial networks but in diffusion models. In this study, we propose a framework to effectively mitigate the disparity in frequency domain of the generated images to improve generative performance of both GAN and diffusion models. This is realized by spectrum translation for the refinement of image generation (STIG) based on contrastive learning. We adopt theoretical logic of frequency components in various generative networks. The key idea, here, is to refine the spectrum of the generated image via the concept of image-to-image translation and contrastive learning in terms of digital signal processing. We evaluate our framework across eight fake image datasets and various cutting-edge models to demonstrate the effectiveness of STIG. Our framework outperforms other cutting-edges showing significant decreases in FID and log frequency distance of spectrum. We further emphasize that STIG improves image quality by decreasing the spectral anomaly. Additionally, validation results present that the frequency-based deepfake detector confuses more in the case where fake spectrums are manipulated by STIG.

Purpose In MRI, motion artifacts can significantly degrade image quality. Motion artifact correction methods using deep neural networks usually required extensive training on large datasets, making them time‐consuming and resource‐intensive. In this paper, an unsupervised deep learning‐based motion artifact correction method for turbo‐spin echo MRI is proposed using the deep image prior framework. Theory and Methods The proposed approach takes advantage of the high impedance to motion artifacts offered by the neural network parameterization to remove motion artifacts in MR images. The framework consists of parameterization of MR image, automatic spatial transformation, and motion simulation model. The proposed method synthesizes motion‐corrupted images from the motion‐corrected images generated by the convolutional neural network, where an optimization process minimizes the objective …

Fast magnetic resonance imaging (MRI) scan is usually achieved by undersampling in k-space, and reconstruction methods for image domain is indispensable. Conventional reconstruction methods rely on independent sensitivity maps of the receiver coils to synthesize the components of the spatial harmonics in k-space. In recent years, deep learning-based MRI algorithms have been providing more accurate reconstructed image than the traditional results. Nonetheless, there is room for improvement. In this study, we proposed a new deep learning-based reconstruction algorithm to use image and k-space domain data simultaneously. The Gabor filter was also defined to effectively incorporate two different domain data in learning stage. Experimental results using real MRI data showed that the proposed method outperforms other deep learning-based algorithms for three metrics of nMSE, SSIM, and VIF.

BACKGROUND. Radiologists are prone to missing some colorectal cancers (CRCs) on routine abdominopelvic CT examinations that are in fact detectable on the images. OBJECTIVE. The purpose of this study was to develop an artificial intelligence (AI) model to detect CRC on routine abdominopelvic CT examinations performed without bowel preparation. METHODS. This retrospective study included 3945 patients (2275 men, 1670 women; mean age, 62 years): a training set of 2662 patients from Severance Hospital with CRC who underwent routine contrast-enhanced abdominopelvic CT before treatment between January 2010 and December 2014 and internal (841 patients from Severance Hospital) and external (442 patients from Gangnam Severance Hospital) test sets of patients who underwent routine contrast-enhanced abdominopelvic CT for any indication and colonoscopy within a 2-month interval …

Automatic medical image segmentation is a crucial procedure for computer-assisted surgery. Especially, three-dimensional reconstruction of medical images of the surgical targets can be accurate in fine anatomical structures with optimal image segmentation, thus leading to successful surgical results. However, the performance of the automatic segmentation algorithm highly depends on the consistent properties of medical images. To address this issue, we propose a model for standardizing computed tomography (CT) images. Hence, our CT image-to-image translation network enables diverse CT images (non-standard images) to be translated to images with identical features (standard images) for the more precise performance of U-Net segmentation. Specifically, we combine an image-to-image translation network with a generative adversarial network, consisting of a residual block-based generative …

(57) ABSTRACT A magnetic resonance imaging apparatus includes a gradient magnetic field controller which applies a spatial encoding gradient to a plurality of slices and applies a gradient magnetic field in a first direction with respect to the plurality of slices, a radio frequency (RF) receiver which receives respective magnetic resonance signals from each of the plurality of slices undersampled in a second direction dif ferent from the first direction, and an image processor which generates a respective magnetic resonance image of each of the plurality of slices based on the magnetic resonance signals received from the plurality of slices.

Provided is a magnetic resonance imaging (MRI) apparatus for generating an MR image based on a steady state free procession (SSFP) sequence. The MRI apparatus includes: a radio frequency (RF) controller configured to apply to an object a first RF pulse corresponding to a first slice and a second RF pulse corresponding to a second slice; a gradient magnetic field controller configured to control respective gradient magnetic fields in a first direction corresponding to the first and second RF pulses, from among spatial encoding gradients, to have opposite polarities with respect to each other; a data acquisition unit configured to acquire first and second MR signals corresponding to the first and second slices; and an image processor configured to generate first and second MR images corresponding to the first and second slices, respectively, based on the acquired first and second MR signals.

Purpose To obtain multicontrast images including fat‐suppressed contrast image, a novel multicontrast imaging method using an SSFP sequence with alternating RF flip angles is proposed. Methods The proposed method uses the balanced SSFP sequence with 2 flip angles. In general, the conventional balanced SSFP sequence has its own unique contrast, which combines both FID signal and echo signal under a steady‐state condition. By using alternating RF flip angles and RF phase cycling, various image contrasts weighted by proton density, T 1 , and T 2 can be obtained. The proposed method offers multicontrast images with fat suppression by using the combination of 2 images obtained just after alternating RF pulses, respectively. Results As demonstrated by simulations, phantom and in vivo experiments, the proposed method provides multicontrast knee images including fat‐suppressed contrast images …

Senior Research Scientist & Associate Professor

Intensified Course Completion of Major, Electrical Engineering

Assistant Professor

IMPEL

교수님 소개

서현석

연구실 소개

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Prior-aware and Context-guided Group Sampling for Active Probabilistic Subsampling

Physics-informed optimization of saturation-transfer MRI protocols using non-differentiable Bloch models

Cyclic Conditional Diffusion Models for CT-to-MR Synthetic Image Segmentation with Misaligned Image Pairs

연구실 보유 특허

대학별 입시끈

우리 연구실 모집

교수님 소개

서현석

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Prior-aware and Context-guided Group Sampling for Active Probabilistic Subsampling

Physics-informed optimization of saturation-transfer MRI protocols using non-differentiable Bloch models

Cyclic Conditional Diffusion Models for CT-to-MR Synthetic Image Segmentation with Misaligned Image Pairs

연구실 보유 특허

최근 3년간 연구실적 높은 연구실

대학별 입시끈

우리 연구실 모집