• Journal of Biomedical Engineering Research
• /
• v.38 no.1
• /
• pp.25-31
• /
• 2017

Testing TB in chest X-ray images is a typical method to diagnose presence and magnitude of PTB lesion. However, the method has limitation due to inter-reader variability. Therefore, it is essential to overcome this drawback with automatic interpretation. In this study, we propose a novel method for detection of PTB using SegNet, which is a deep learning architecture for semantic pixel wise image labelling. SegNet is composed of a stack of encoders followed by a corresponding decoder stack which feeds into a soft-max classification layer. We modified parameters of SegNet to change the number of classes from 12 to 2 (TB or none-TB) and applied the architecture to automatically interpret chest radiographs. 552 chest X-ray images, provided by The Korean Institute of Tuberculosis, used for training and test and we constructed a receiver operating characteristic (ROC) curve. As a consequence, the area under the curve (AUC) was 90.4% (95% CI:[85.1, 95.7]) with a classification accuracy of 84.3%. A sensitivity was 85.7% and specificity was 82.8% on 431 training images (TB 172, none-TB 259) and 121 test images (TB 63, none-TB 58). This results show that detecting PTB using SegNet is comparable to other PTB detection methods.

• Journal of the Korean Society of Radiology
• /
• v.18 no.1
• /
• pp.21-26
• /
• 2024

The one of the famous deep learning models for object detection task is you only look once version 5 (YOLOv5) framework based on the one stage architecture. In addition, YOLOv5 model indicated high performance for accurate lesion detection using the bottleneck CSP layer and skip connection function. The purpose of this study was to evaluate the performance of YOLOv5 framework according to various hyperparameters in position emission tomogrpahy (PET) phantom images. The dataset was obtained from QIN PET segmentation challenge in 500 slices. We set the bounding box to generate ground truth dataset using labelImg software. The hyperparameters for network train were applied by changing optimization function (SDG, Adam, and AdamW), activation function (SiLU, LeakyRelu, Mish, and Hardwish), and YOLOv5 model size (nano, small, large, and xlarge). The intersection over union (IOU) method was used for performance evaluation. As a results, the condition of outstanding performance is to apply AdamW, Hardwish, and nano size for optimization function, activation function and model version, respectively. In conclusion, we confirmed the usefulness of YOLOv5 network for object detection performance in nuclear medicine images.

• Imaging Science in Dentistry
• /
• v.54 no.1
• /
• pp.49-56
• /
• 2024

Purpose: This study evaluated the impact of artifacts generated by metal crowns on the detection of proximal caries lesions in teeth at various distances using cone-beam computed tomography (CBCT). Additionally, the diagnostic impacts of tube current and metal artifact reduction (MAR) were investigated. Materials and Methods: Thirty teeth were arranged within 10 phantoms, each containing 1 first premolar, 1 second premolar, and 1 second molar. A sound first molar (for the control group) or a tooth with a metal crown was placed. Of the 60 proximal surfaces evaluated, 15 were sound and 45 exhibited enamel caries. CBCT scans were acquired using an OP300 Maxio unit (Instrumentarium, Tuusula, Finland), while varying the tube current (4, 8, or 12.5 mA) and enabling or disabling MAR. Five observers assessed mesial and distal surfaces using a 5-point scale. Multi-way analysis of variance was employed for data comparison, with P<0.05 indicating statistical significance. Results: The area under the curve (AUC) varied from 0.40 to 0.60 (sensitivity: 0.28-0.45, specificity: 0.44-0.80). The diagnostic accuracy was not significantly affected by the presence of a metal crown, milliamperage, or MAR(P>0.05). However, the overall AUC and specificity were significantly lower for surfaces near a crown (P<0.05). Conclusion: CBCT-based caries detection was not influenced by the presence of a metal crown, variations in milliamperage, or MAR activation. However, the diagnostic accuracy was low and was further diminished for surfaces near a crown. Consequently, CBCT is not recommended for the detection of incipient caries lesions.

• Journal of Biomedical Engineering Research
• /
• v.19 no.6
• /
• pp.551-562
• /
• 1998

Amultiresolution filtering algorithm based on the physical SPECT lesion detachability provides and optimal solution for SPECT reconstruction problem. Related to the previous study, we estimated the SPECT lesion detection capability by m minimum detectable lesion sizes (MDLSs), and generated m reconstruction filters which are designed to maximize the smoothing effect at a fixed MDLS-dependent resolution level $\frac{MDLS}{4\sqrt{2In2}}$. The proposed multiresolution filtering algorithm used a coarse-to-fine approach for the m-level resolution filter images obtained from these m filters for a given projection image. First, the local homogeneity is determined for every pixel of the filter images, by comparing the local variance value computed in a window centered at the pixel and the mode determined from the distribution of the local variances. Based on the local homogeneity, the pixels declared as homogeneous are chosen from the filter image of the lowest resolution, and for the other pixels the same process is repeated for the higher resolution filter images. For the non-homogeneous pixels after this pixels after this repetition process ends, the pixel values of the highest resolution filter image are substituted. From the results of the simulated experiments, the proposed multiresolution filtering algorithm showed a strong smoothing effect in the homogeneous regions and a significant resolution improvement near the edge regions of the projection images, and so produced good adaptability effects in the reconstructed images.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.31 no.3
• /
• pp.506-515
• /
• 2004

The objective of this study was to compare the rate of in vitro demineralization of bovine permanent (BP), human deciduous (HD) and human permanent (HP) enamel. Twenty aye flattened and polished enamel samples for each group (BP, HD, HP) were immersed in a demineralizing solution (0.1 mol/L lactic acid, 0.2% Carbopol 907, and 50% saturated hydroxyapatite) for 1, 2, 4 or 8 days. All 25 samples from each group were subjected to Quantitative light induced fluorescence analysis (QLF) and 5 samples from each group were randomly selected for Transverse Microradiography analysis (TMR). Integrated mineral loss (IML) and lesion depth (LD) were determined by TMR. The fluorescence radiance (FR) of sound enamel $(FR_S)$, demineralized enamel $(FR_D)$ were determined by QLF and FR ratio $(FR_D/FR_S)$ was calculated. Bovine enamel samples showed significant correlation between FR ratio and lesion depth(p<0.05) and deciduous enamel samples does not showed significant correlation between FR ratio and lesion depth(p>0.05). Permanent enamel samples showed significant correlation between FR ratio and lesion depth(p<0.05) The constant of demineralization time between FR ratio from regression analysis were as follows: bovine enamel was -4.643(p<0.05) deciduous enamel was -5.421(p<0.05) and permanent enamel was -4.435(p<0.05).

• Investigative Magnetic Resonance Imaging
• /
• v.12 no.2
• /
• pp.161-169
• /
• 2008

Purpose : The aim of this study was to evaluate the usefulness of MRI performed after excisional biopsy to diagnose residual cancer and additional lesions. Materials and Methods : A total of 16 patients who had breast cancer diagnosed by excisional biopsy underwent preoperative breast MRI between March 2005 and August 2007 were included. MRI findings were considered positive for residual cancer if focal, thick, or irregular rim enhancement or adjacent enhancing lesion was identified around the postoperative biopsy cavity. And additional lesions separated from biopsy cavity including multifocal, multicentric, or contralateral lesion were evaluated. We evaluated the diagnostic accuracy of MRI, comparing MRI with histopathologic finding, and the impact of MRI on surgical planning. Results : The sensitivity and specificity of MRI for detecting residual disease considering rim enhancement were 85.7%(6/7) and 55.6%(5/9). Additional lesions including multifocal, multicentric, or contralateral lesion were found in 6 patients. In 7 patients, results of MRI findings changed surgical treatment planning. Conclusion : The pattern of rim enhancement on MRI after excisional biopsy is not differential point to evaluate remnant lesion. But MRI has an important role to help the detection of multifocal or multicentric, or contralateral breast malignancies, resulting in beneficial change in surgical treatment planning.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.46 no.3
• /
• pp.265-273
• /
• 2019

This study was performed to evaluate clinical use of laser fluorescence (LF) to identify early childhood caries lesions suitable for applying resin infiltration. 20 exfoliated primary molars with proximal caries were selected and cut buccolingually cross the central pit for regarding the mesial and distal surfaces respectively. 27 specimens corresponding to ICDAS code 1 and 2 were selected and the LF values were measured. When infiltrant resin was applied, double staining for microscopy detection has done simultaneously. Tooth samples were sliced with 0.7 mm thick. The maximum lesion depth, maximum penetration depth, and average penetration rate were measured from the confocal scanning laser microscope image. Pearson correlation analysis was performed. The intraclass correlation coefficient of LF values shows excellent agreement. LF values had positive correlation with penetration rate, but not lesion depth and penetration depth. Significant correlation between LF readings and penetration rate was verified in deep enamel caries and dentin caries except shallow enamel caries. Infiltrant resin could penetrate with a higher rate and LF values could be increased in more active caries lesions. In assessing radiologically similar caries lesion, laser fluorescence might be useful for identifying caries activity.

• Research in Plant Disease
• /
• v.29 no.4
• /
• pp.459-463
• /
• 2023

This study was conducted to investigate the plant-parasitic nematodes of ornamental foliage plants, citrus orchards, and tea plantations from July to December 2022. As a result of the investigation of plant-parasitic nematodes in 415 foliage plants, root-lesion nematodes were detected most frequently, followed by root-knot nematodes, pin nematodes, and other nematodes. In tea plantations, spiral nematodes, cyst nematodes, and root-knot nematodes were detected. Citrus nematodes, ring nematodes, and root-knot nematodes were discovered in citrus orchards. In foliage plants, tea plantations, and citrus orchards, the detection rate of plant-parasitic nematodes was not high, but root-lesion nematodes were detected. Therefore, it is necessary to apply appropriate control methods to manage root-lesion nematodes during the cultivation of foliage, tea, and citrus plants.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.1
• /
• pp.1-7
• /
• 2008

PET allows non-invasive, quantitative and repetitive imaging of biological function in living animals. Small animal PET imaging with $[^{18}F]$FDG has been successfully applied to investigation of metabolism, receptor-ligand interactions, gene expression, adoptive cell therapy and somatic gene therapy. Experimental condition of animal handling impacts on the biodistribution of $[^{18}F]$FDG in small animal study. The small animal PET and CT images were registered using the hardware fiducial markers and small animal contour point. Tumor imaging in small animal with small animal $[^{18}F]$FDG PET should be considered fasting, warming, and isoflurane anesthesia level. Registered imaging with small animal PET and CT image could be useful for the detection of tumor. Small animal experimental condition of animal handling and registration method will be of most importance for small lesion detection of metastases tumor model.

• Journal of Korean Neurosurgical Society
• /
• v.30 no.12
• /
• pp.1417-1421
• /
• 2001

Objective : The purposes of this study are to compare imaging features with operative findings and to determine significance of imaging studies for early detection of craniosynostosis(CS). Methods : Plain radiograph of skull and three-dimensional(3D) CT reconstruction were analyzed in 10 consecutive patients with CS to assess the presence and the extent of synostosis. The radiological findings were investigated and compared with operative findings. Results : The locations of lesion were coronal suture in 6, sagittal suture in 3 and multiple sutures in one patient, and the age ranged 1 to 53 months(mean age : 17.4 months). Reconstructive procedures with or without advancement of supraorbital rim were performed in coronal CS patients and ${\pi}$-procedures or synostectomy were done in sagittal CS patients. Radi-ological abnormalities such as sutural indistinctness or sclerosis, bony ridge, bossing and other bony deformities were nearly consistent with surgical findings. Conclusion : The interpretation of imaging study are very important for early detection of craniosynostosis, especially, the plain radiographs of skull. Also 3D CT imaging is helpful in diagnosis and surgical planing of craniosynostosis. There are no significant differences between imaging features and operative findings in CS patients.