• 한국가시화정보학회지
• /
• 제21권2호
• /
• pp.14-23
• /
• 2023

The pressure difference across stenotic blood vessels is a commonly used clinical metric for diagnosing many cardiovascular diseases. At present, most clinical pressure measurements rely solely on invasive catheterization. In this study, we propose a novel method for non-invasive pressure estimation using the incompressible Navier-Stokes equations and a 3D multi-path integration approach. We verify spatio-temporal convergence on an in-silico dataset of a cylindrical straight pipe phantom with steady and pulsatile flow fields. We then evaluate the proposed method on an in vitro dataset of reconstructed control, pre-operative, and post-operative carotid artery cases acquired from 4D flow MRI. The performance of our method is compared to existing approaches based on the pressure Poisson equation and work-energy relative pressure. The results demonstrate the proposed method's high accuracy, robustness to spatio-temporal subsampling, and reduced sensitivity to noise, highlighting its great potential for non-invasive pressure estimation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권12호
• /
• pp.4816-4834
• /
• 2020

This paper proposes transfer learning and fine-tuning techniques for a deep learning model to detect three distinct brain tumors from Magnetic Resonance Imaging (MRI) scans. In this work, the recent YOLOv4 model trained using a collection of 3064 T1-weighted Contrast-Enhanced (CE)-MRI scans that were pre-processed and labeled for the task. This work trained with the partial 29-layer YOLOv4-Tiny and fine-tuned to work optimally and run efficiently in most platforms with reliable performance. With the help of transfer learning, the model had initial leverage to train faster with pre-trained weights from the COCO dataset, generating a robust set of features required for brain tumor detection. The results yielded the highest mean average precision of 93.14%, a 90.34% precision, 88.58% recall, and 89.45% F1-Score outperforming other previous versions of the YOLO detection models and other studies that used bounding box detections for the same task like Faster R-CNN. As concluded, the YOLOv4-Tiny can work efficiently to detect brain tumors automatically at a rapid phase with the help of proper fine-tuning and transfer learning. This work contributes mainly to assist medical experts in the diagnostic process of brain tumors.

• 방송공학회논문지
• /
• 제28권3호
• /
• pp.267-274
• /
• 2023

의료영상 공개 데이터는 수집에 한계가 있어 데이터셋의 양이 부족하다는 문제점이 있다. 때문에 기존 연구들은 공개 데이터셋에 과적합 되었을 우려가 있다. 본 논문은 실험을 통해 8개의 (Unet, X-Net, HarDNet, SegNet, PSPNet, SwinUnet, 3D-ResU-Net, UNETR) 의료영상 분할 모델의 성능을 비교함으로써 기존 모델의 성능을 재검증하고자 한다. 뇌졸중 진단 공개 데이터 셋인 Anatomical Tracings of Lesions After Stroke(ATLAS) V1.2과 ATLAS V2.0에서 모델들의 성능 비교 실험을 진행한다. 실험결과 대부분 모델은 V1.2과 V2.0에서 성능이 비슷한 결과를 보였다. 하지만 X-net과 3D-ResU-Net는 V1.2 데이터셋에서 더 높은 성능을 기록했다. 이러한 결과는 해당 모델들이 V1.2에 과적합 되었을 것으로 해석할 수 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권1호
• /
• pp.16-37
• /
• 2022

Accurate liver segment segmentation based on radiological images is indispensable for the preoperative analysis of liver tumor resection surgery. However, most of the existing segmentation methods are not feasible to be used directly for this task due to the challenge of exact edge prediction with some tiny and slender vessels as its clinical segmentation criterion. To address this problem, we propose a novel deep learning based segmentation model, called Boundary-Aware Dual Attention Liver Segment Segmentation Model (BADA). This model can improve the segmentation accuracy of liver segments with enhancing the edges including the vessels serving as segment boundaries. In our model, the dual gated attention is proposed, which composes of a spatial attention module and a semantic attention module. The spatial attention module enhances the weights of key edge regions by concerning about the salient intensity changes, while the semantic attention amplifies the contribution of filters that can extract more discriminative feature information by weighting the significant convolution channels. Simultaneously, we build a dataset of liver segments including 59 clinic cases with dynamically contrast enhanced MRI(Magnetic Resonance Imaging) of portal vein stage, which annotated by several professional radiologists. Comparing with several state-of-the-art methods and baseline segmentation methods, we achieve the best results on this clinic liver segment segmentation dataset, where Mean Dice, Mean Sensitivity and Mean Positive Predicted Value reach 89.01%, 87.71% and 90.67%, respectively.

• International Journal of Computer Science & Network Security
• /
• 제24권5호
• /
• pp.53-63
• /
• 2024

Alzheimer's disease is a brain disorder that worsens over time and affects millions of people around the world. It leads to a gradual deterioration in memory, thinking ability, and behavioral and social skills until the person loses his ability to adapt to society. Technological progress in medical imaging and the use of artificial intelligence, has provided the possibility of detecting Alzheimer's disease through medical images such as magnetic resonance imaging (MRI). However, Deep learning algorithms, especially convolutional neural networks (CNNs), have shown great success in analyzing medical images for disease diagnosis and classification. Where CNNs can recognize patterns and objects from images, which makes them ideally suited for this study. In this paper, we proposed to compare the performances of Alzheimer's disease detection by using two deep learning methods: You Only Look Once (YOLO), a CNN-enabled object recognition algorithm, and Visual Geometry Group (VGG16) which is a type of deep convolutional neural network primarily used for image classification. We will compare our results using these modern models Instead of using CNN only like the previous research. In addition, the results showed different levels of accuracy for the various versions of YOLO and the VGG16 model. YOLO v5 reached 56.4% accuracy at 50 epochs and 61.5% accuracy at 100 epochs. YOLO v8, which is for classification, reached 84% accuracy overall at 100 epochs. YOLO v9, which is for object detection overall accuracy of 84.6%. The VGG16 model reached 99% accuracy for training after 25 epochs but only 78% accuracy for testing. Hence, the best model overall is YOLO v9, with the highest overall accuracy of 86.1%.

• 대한영상의학회지
• /
• 제82권2호
• /
• pp.406-416
• /
• 2021

목적 원발성 비점액성 직장암 환자에서 자기공명영상 기반 텍스처 분석 변수와 Kirsten rat sarcoma viral oncogene homolog (이하 KRAS) 유전자 변이와의 연관성을 조사한다. 방법 조직학적으로 비점액성 직장 선암종으로 진단받고 KRAS 유전자 정보가 있으며 치료 전 직장 자기공명영상을 시행한 79명의 환자를 훈련 데이터셋(n = 46)과 검증 데이터셋(n = 33)으로 나누었다. 텍스처 분석은 축상면 T2 강조영상에서 시행되었다. 텍스처 변수와 KRAS 유전자 변이와의 연관성은 Mann-Whitney U 검정을 통해 통계적으로 분석하였다. 수신기작동 특성 곡선(receiver operating characteristic) 분석을 이용하여 KRAS 유전자 변이를 예측하기 위한 최적의 절단값을 산출하였다. 이 절단값은 검증 데이터셋을 사용해 검증되었다. 결과 훈련 데이터셋에서 왜도(skewness)는 유전자 변이가 있는 집단(n = 22명)에서 유전자 변이가 없는 집단(n = 24명)보다 유의하게 높았다(0.221 ± 0.283; -0.006 ± 0.178, p = 0.003). 왜도의 곡선 하 면적 값(area under the curve)은 0.757 (95% 신뢰구간, 0.606-0.872)로 정확도는 71%, 민감도는 64%, 특이도는 78%였다. 다른 텍스처 변수들은 두 집단 간 유의한 차이를 보이지 않았다(p > 0.05). 검증 데이터셋에 절단값 0.078을 적용하였을 때 정확도는 76%, 민감도는 86%, 특이도는 68%였다. 결론 원발성 비점액성 직장암 환자에서 왜도는 KRAS 유전자 변이와 연관성을 보였다.

• International Journal of Computer Science & Network Security
• /
• 제22권5호
• /
• pp.143-148
• /
• 2022

Brain tumor classification is an important process that allows doctors to plan treatment for patients based on the stages of the tumor. To improve classification performance, various CNN-based architectures are used for brain tumor classification. Existing methods for brain tumor segmentation suffer from overfitting and poor efficiency when dealing with large datasets. The enhanced CNN architecture proposed in this study is based on U-Net for brain tumor segmentation, RefineNet for pattern analysis, and SegNet architecture for brain tumor classification. The brain tumor benchmark dataset was used to evaluate the enhanced CNN model's efficiency. Based on the local and context information of the MRI image, the U-Net provides good segmentation. SegNet selects the most important features for classification while also reducing the trainable parameters. In the classification of brain tumors, the enhanced CNN method outperforms the existing methods. The enhanced CNN model has an accuracy of 96.85 percent, while the existing CNN with transfer learning has an accuracy of 94.82 percent.

• 스마트미디어저널
• /
• 제12권3호
• /
• pp.30-37
• /
• 2023

Alzheimer's disease (AD) symptoms are being treated by early diagnosis, where we can only slow the symptoms and research is still undergoing. In consideration, using T1-weighted images several classification models are proposed in Machine learning to identify AD. In this paper, we consider the improvised feature selection, to reduce the complexity by using wrapping techniques and Restricted Boltzmann Machine (RBM). This present work used the subcortical and cortical features of 278 subjects from the ADNI dataset to identify AD and sMRI. Multi-class classification is used for the experiment i.e., AD, EMCI, LMCI, HC. The proposed feature selection consists of Forward feature selection, Backward feature selection, and Combined PCA & RBM. Forward and backward feature selection methods use an iterative method starting being no features in the forward feature selection and backward feature selection with all features included in the technique. PCA is used to reduce the dimensions and RBM is used to select the best feature without interpreting the features. We have compared the three models with PCA to analysis. The following experiment shows that combined PCA &RBM, and backward feature selection give the best accuracy with respective classification model RF i.e., 88.65, 88.56% respectively.

• 생물정신의학
• /
• 제21권3호
• /
• pp.81-86
• /
• 2014

Objectives Total intracranial volume (TIV) is a major nuisance of neuroimaging research for interindividual differences of brain structure and function. Authors intended to prove the reliability of the atlas scaling factor (ASF) method for TIV estimation in FreeSurfer by comparing it with the results of manual tracing as reference method. Methods The TIVs of 26 normal children and 26 children with attention-deficit hyperactivity disorder (ADHD) were obtained by using FreeSurfer reconstruction and manual tracing with T1-weighted images. Manual tracing performed in every 10th slice of MRI dataset from midline of sagittal plane by one researcher who was blinded from clinical data. Another reseacher performed manual tracing independently for randomly selected 20 dataset to verify interrater reliability. Results The interrater reliability was excellent (intraclass coefficient = 0.91, p < 7.1e-07). There were no significant differences of age and gender distribution between normal and ADHD groups. No significant differences were found between TIVs from ASF method and manual tracing. Strong correlation between TIVs from 2 different methods were shown (r = 0.90, p < 2.2e-16). Conclusions The ASF method for TIV estimation by using FreeSurfer showed good agreement with the reference method. We can use the TIV from ASF method for correction in analysis of structural and functional neuroimaging studies with not only elderly subjects but also children, even with ADHD.

• 한국의학물리학회:학술대회논문집
• /
• 한국의학물리학회 2002년도 Proceedings
• /
• pp.450-453
• /
• 2002

In radiotherapy treatment planning, it is critical to deliver the radiation dose to tumor and protect surrounding normal tissue. Recent developments in functional imaging and radiotherapy treatment technology have been raising chances to control tumor saving normal tissues. A brain phantom which could be used for image registration technique of CT-MR and CT-SPECT images using surface matching was developed. The brain phantom was specially designed to obtain imaging dataset of CT, MR, and SPECT. The phantom had an external frame with 4 N-shaped pipes filled with acryl rods, Pb rods for CT, MR, and SPECT imaging, respectively. 8 acrylic pipes were inserted into the empty space of the brain phantom to be imaged for geometric evaluation of the matching. For an optimization algorithm of image registration, we used Downhill simplex algorithm suggested as a fast surface matching algorithm. Accuracy of image fusion was assessed by the comparison between the center points of the section of N-shaped bars in the external frame and the inserted pipes of the phantom and minimized cost functions of the optimization algorithm. Technique with partially transparent, mixed images using color on gray was used for visual assessment of the image registration process. The errors of image registration of CT-MR and CT-SPECT were within 2mm and 4mm, respectively. Since these errors were considered within a reasonable margin from the phantom study, the phantom is expected to be used for conventional image registration between multimodal image datasets..