ReconU-Net: a direct PET image reconstruction using U-Net architecture with back projection-induced skip connection

Objective.This study aims to introduce a novel back projection-induced U-Net-shaped architecture, called ReconU-Net, based on the original U-Net architecture for deep learning-based direct positron emission tomography (PET) image reconstruction. Additionally, our objective is to visualize the behavior of direct PET image reconstruction by comparing the proposed ReconU-Net architecture with the original U-Net architecture and existing DeepPET encoder-decoder architecture without skip connections.Approach. The proposed ReconU-Net architecture uniquely integrates the physical model of the back projection operation into the skip connection. This distinctive feature facilitates the effective transfer of intrinsic spatial information from the input sinogram to the reconstructed image via an embedded physical model. The proposed ReconU-Net was trained using Monte Carlo simulation data from the Brainweb phantom and tested on both simulated and real Hoffman brain phantom data.Main results. The proposed ReconU-Net method provided better reconstructed image in terms of the peak signal-to-noise ratio and contrast recovery coefficient than the original U-Net and DeepPET methods. Further analysis shows that the proposed ReconU-Net architecture has the ability to transfer features of multiple resolutions, especially non-abstract high-resolution information, through skip connections. Unlike the U-Net and DeepPET methods, the proposed ReconU-Net successfully reconstructed the real Hoffman brain phantom, despite limited training on simulated data.Significance. The proposed ReconU-Net can improve the fidelity of direct PET image reconstruction, even with small training datasets, by leveraging the synergistic relationship between data-driven modeling and the physics model of the imaging process.

Whole Reconstruction-Free System Design for Direct Positron Emission Imaging From Image Generation to Attenuation Correction

Direct positron emission imaging (dPEI), which does not require a mathematical reconstruction step, is a next-generation molecular imaging modality. To maximize the practical applicability of the dPEI system to clinical practice, we introduce a novel reconstruction-free image-formation method called direct μCompton imaging, which directly localizes the interaction position of Compton scattering from the annihilation photons in a three-dimensional space by utilizing the same compact geometry as that for dPEI, involving ultrafast time-of-flight radiation detectors. This unique imaging method not only provides the anatomical information about an object but can also be applied to attenuation correction of dPEI images. Evaluations through Monte Carlo simulation showed that functional and anatomical hybrid images can be acquired using this multimodal imaging system. By fusing the images, it is possible to simultaneously access various object data, which ensures the synergistic effect of the two imaging methodologies. In addition, attenuation correction improves the quantification of dPEI images. The realization of the whole reconstruction-free imaging system from image generation to quantitative correction provides a new perspective in molecular imaging.

Deep learning-based PET image denoising and reconstruction: a review

This review focuses on positron emission tomography (PET) imaging algorithms and traces the evolution of PET image reconstruction methods. First, we provide an overview of conventional PET image reconstruction methods from filtered backprojection through to recent iterative PET image reconstruction algorithms, and then review deep learning methods for PET data up to the latest innovations within three main categories. The first category involves post-processing methods for PET image denoising. The second category comprises direct image reconstruction methods that learn mappings from sinograms to the reconstructed images in an end-to-end manner. The third category comprises iterative reconstruction methods that combine conventional iterative image reconstruction with neural-network enhancement. We discuss future perspectives on PET imaging and deep learning technology.

Deep learning-based correction of cataract-induced influence on macular pigment optical density measurement by autofluorescence spectroscopy

PURPOSE

Measurements of macular pigment optical density (MPOD) using the autofluorescence spectroscopy yield underestimations of actual values in eyes with cataracts. Previously, we proposed a correction method for this error using deep learning (DL); however, the correction performance was validated through internal cross-validation. This cross-sectional study aimed to validate this approach using an external validation dataset.

METHODS

MPODs at 0.25°, 0.5°, 1°, and 2° eccentricities and macular pigment optical volume (MPOV) within 9° eccentricity were measured using SPECTRALIS (Heidelberg Engineering, Heidelberg, Germany) in 197 (training dataset inherited from our previous study) and 157 eyes (validating dataset) before and after cataract surgery. A DL model was trained to predict the corrected value from the pre-operative value using the training dataset, and we measured the discrepancy between the corrected value and the actual postoperative value. Subsequently, the prediction performance was validated using a validation dataset.

RESULTS

Using the validation dataset, the mean absolute values of errors for MPOD and MPOV corrected using DL ranged from 8.2 to 12.4%, which were lower than values with no correction (P < 0.001, linear mixed model with Tukey's test). The error depended on the autofluorescence image quality used to calculate MPOD. The mean errors in high and moderate quality images ranged from 6.0 to 11.4%, which were lower than those of poor quality images.

CONCLUSION

The usefulness of the DL correction method was validated. Deep learning reduced the error for a relatively good autofluorescence image quality. Poor-quality images were not corrected.

Fully 3D implementation of the end-to-end deep image prior-based PET image reconstruction using block iterative algorithm

Objective. Deep image prior (DIP) has recently attracted attention owing to its unsupervised positron emission tomography (PET) image reconstruction method, which does not require any prior training dataset. In this paper, we present the first attempt to implement an end-to-end DIP-based fully 3D PET image reconstruction method that incorporates a forward-projection model into a loss function.Approach. A practical implementation of a fully 3D PET image reconstruction could not be performed at present because of a graphics processing unit memory limitation. Consequently, we modify the DIP optimization to a block iteration and sequential learning of an ordered sequence of block sinograms. Furthermore, the relative difference penalty (RDP) term is added to the loss function to enhance the quantitative accuracy of the PET image.Main results. We evaluated our proposed method using Monte Carlo simulation with [18F]FDG PET data of a human brain and a preclinical study on monkey-brain [18F]FDG PET data. The proposed method was compared with the maximum-likelihood expectation maximization (EM), maximuma posterioriEM with RDP, and hybrid DIP-based PET reconstruction methods. The simulation results showed that, compared with other algorithms, the proposed method improved the PET image quality by reducing statistical noise and better preserved the contrast of brain structures and inserted tumors. In the preclinical experiment, finer structures and better contrast recovery were obtained with the proposed method.Significance.The results indicated that the proposed method could produce high-quality images without a prior training dataset. Thus, the proposed method could be a key enabling technology for the straightforward and practical implementation of end-to-end DIP-based fully 3D PET image reconstruction.

List-Mode PET Image Reconstruction Using Deep Image Prior

List-mode positron emission tomography (PET) image reconstruction is an important tool for PET scanners with many lines-of-response and additional information such as time-of-flight and depth-of-interaction. Deep learning is one possible solution to enhance the quality of PET image reconstruction. However, the application of deep learning techniques to list-mode PET image reconstruction has not been progressed because list data is a sequence of bit codes and unsuitable for processing by convolutional neural networks (CNN). In this study, we propose a novel list-mode PET image reconstruction method using an unsupervised CNN called deep image prior (DIP) which is the first trial to integrate list-mode PET image reconstruction and CNN. The proposed list-mode DIP reconstruction (LM-DIPRecon) method alternatively iterates the regularized list-mode dynamic row action maximum likelihood algorithm (LM-DRAMA) and magnetic resonance imaging conditioned DIP (MR-DIP) using an alternating direction method of multipliers. We evaluated LM-DIPRecon using both simulation and clinical data, and it achieved sharper images and better tradeoff curves between contrast and noise than the LM-DRAMA, MR-DIP and sinogram-based DIPRecon methods. These results indicated that the LM-DIPRecon is useful for quantitative PET imaging with limited events while keeping accurate raw data information. In addition, as list data has finer temporal information than dynamic sinograms, list-mode deep image prior reconstruction is expected to be useful for 4D PET imaging and motion correction.

Unbiased TOF estimation using leading-edge discriminator and convolutional neural network trained by single-source-position waveforms

Objective. Convolutional neural networks (CNNs) are a strong tool for improving the coincidence time resolution (CTR) of time-of-flight (TOF) positron emission tomography detectors. However, several signal waveforms from multiple source positions are required for CNN training. Furthermore, there is concern that TOF estimation is biased near the edge of the training space, despite the reduced estimation variance (i.e. timing uncertainty). Approach. We propose a simple method for unbiased TOF estimation by combining a conventional leading-edge discriminator (LED) and a CNN that can be trained with waveforms collected from one source position. The proposed method estimates and corrects the time difference error calculated by the LED rather than the absolute time difference. This model can eliminate the TOF estimation bias, as the combination with the LED converts the distribution of the label data from discrete values at each position into a continuous symmetric distribution. Main results. Evaluation results using signal waveforms collected from scintillation detectors show that the proposed method can correctly estimate all source positions without bias from a single source position. Moreover, the proposed method improves the CTR of the conventional LED. Significance. We believe that the improved CTR will not only increase the signal-to-noise ratio but will also contribute significantly to a part of the direct positron emission imaging.

Ultrafast timing enables reconstruction-free positron emission imaging

X-ray and gamma-ray photons are widely used for imaging but require a mathematical reconstruction step, known as tomography, to produce cross-sectional images from the measured data. Theoretically, the back-to-back annihilation photons produced by positron-electron annihilation can be directly localized in three-dimensional space using time-of-flight information without tomographic reconstruction. However, this has not yet been demonstrated due to the insufficient timing performance of available radiation detectors. Here, we develop techniques based on detecting prompt Cerenkov photons, which when combined with a convolutional neural network for timing estimation resulted in an average timing precision of 32 picoseconds, corresponding to a spatial precision of 4.8 mm. We show this is sufficient to produce cross-sectional images of a positron-emitting radionuclide directly from the detected coincident annihilation photons, without using any tomographic reconstruction algorithm. The reconstruction-free imaging demonstrated here directly localizes positron emission, and frees the design of an imaging system from the geometric and sampling constraints that normally present for tomographic reconstruction.

Anatomical-guided attention enhances unsupervised PET image denoising performance

Although supervised convolutional neural networks (CNNs) often outperform conventional alternatives for denoising positron emission tomography (PET) images, they require many low- and high-quality reference PET image pairs. Herein, we propose an unsupervised 3D PET image denoising method based on an anatomical information-guided attention mechanism. The proposed magnetic resonance-guided deep decoder (MR-GDD) utilizes the spatial details and semantic features of MR-guidance image more effectively by introducing encoder-decoder and deep decoder subnetworks. Moreover, the specific shapes and patterns of the guidance image do not affect the denoised PET image, because the guidance image is input to the network through an attention gate. In a Monte Carlo simulation of [¹⁸F]fluoro-2-deoxy-D-glucose (FDG), the proposed method achieved the highest peak signal-to-noise ratio and structural similarity (27.92 ± 0.44 dB/0.886 ± 0.007), as compared with Gaussian filtering (26.68 ± 0.10 dB/0.807 ± 0.004), image guided filtering (27.40 ± 0.11 dB/0.849 ± 0.003), deep image prior (DIP) (24.22 ± 0.43 dB/0.737 ± 0.017), and MR-DIP (27.65 ± 0.42 dB/0.879 ± 0.007). Furthermore, we experimentally visualized the behavior of the optimization process, which is often unknown in unsupervised CNN-based restoration problems. For preclinical (using [¹⁸F]FDG and [¹¹C]raclopride) and clinical (using [¹⁸F]florbetapir) studies, the proposed method demonstrates state-of-the-art denoising performance while retaining spatial resolution and quantitative accuracy, despite using a common network architecture for various noisy PET images with 1/10th of the full counts. These results suggest that the proposed MR-GDD can reduce PET scan times and PET tracer doses considerably without impacting patients.

Correction for the Influence of Cataract on Macular Pigment Measurement by Autofluorescence Technique Using Deep Learning

Purpose

Measurements of macular pigment optical density (MPOD) by the autofluorescence technique yield underestimations of actual values in eyes with cataract. We applied deep learning (DL) to correct this error.

Subjects and Methods

MPOD was measured by SPECTRALIS (Heidelberg Engineering, Heidelberg, Germany) in 197 eyes before and after cataract surgery. The nominal MPOD values (= preoperative value) were corrected by three methods: the regression equation (RE) method, subjective classification (SC) method (described in our previous study), and DL method. The errors between the corrected and true values (= postoperative value) were calculated for local MPODs at 0.25°, 0.5°, 1°, and 2° eccentricities and macular pigment optical volume (MPOV) within 9° eccentricity.

Results

The mean error for MPODs at four eccentricities was 32% without any correction, 15% with correction by RE, 16% with correction by SC, and 14% with correction by DL. The mean error for MPOV was 21% without correction and 14%, 10%, and 10%, respectively, with correction by the same methods. The errors with any correction were significantly lower than those without correction (P < 0.001, linear mixed model with Tukey's test). The errors with DL correction were significantly lower than those with RE correction in MPOD at 1° eccentricity and MPOV (P < 0.001) and were equivalent to those with SC correction.

Conclusions

The objective method using DL was useful to correct MPOD values measured in aged people.

Translational Relevance

MPOD can be obtained with small errors in eyes with cataract using DL.

Deep learning-based attenuation correction for brain PET with various radiotracers

OBJECTIVES

Attenuation correction (AC) is crucial for ensuring the quantitative accuracy of positron emission tomography (PET) imaging. However, obtaining accurate μ-maps from brain-dedicated PET scanners without AC acquisition mechanism is challenging. Therefore, to overcome these problems, we developed a deep learning-based PET AC (deep AC) framework to synthesize transmission computed tomography (TCT) images from non-AC (NAC) PET images using a convolutional neural network (CNN) with a huge dataset of various radiotracers for brain PET imaging.

METHODS

The proposed framework is comprised of three steps: (1) NAC PET image generation, (2) synthetic TCT generation using CNN, and (3) PET image reconstruction. We trained the CNN by combining the mixed image dataset of six radiotracers to avoid overfitting, including [¹⁸F]FDG, [¹⁸F]BCPP-EF, [¹¹C]Racropride, [¹¹C]PIB, [¹¹C]DPA-713, and [¹¹C]PBB3. We used 1261 brain NAC PET and TCT images (1091 for training and 70 for testing). We did not include [¹¹C]Methionine subjects in the training dataset, but included them in the testing dataset.

RESULTS

The image quality of the synthetic TCT images obtained using the CNN trained on the mixed dataset of six radiotracers was superior to those obtained using the CNN trained on the split dataset generated from each radiotracer. In the [¹⁸F]FDG study, the mean relative PET biases of the emission-segmented AC (ESAC) and deep AC were 8.46 ± 5.24 and - 5.69 ± 4.97, respectively. The deep AC PET and TCT AC PET images exhibited excellent correlation for all seven radiotracers (R² = 0.912-0.982).

CONCLUSION

These results indicate that our proposed deep AC framework can be leveraged to provide quantitatively superior PET images when using the CNN trained on the mixed dataset of PET tracers than when using the CNN trained on the split dataset which means specific for each tracer.

4D deep image prior: dynamic PET image denoising using an unsupervised four-dimensional branch convolutional neural network

Although convolutional neural networks (CNNs) demonstrate the superior performance in denoising positron emission tomography (PET) images, a supervised training of the CNN requires a pair of large, high-quality PET image datasets. As an unsupervised learning method, a deep image prior (DIP) has recently been proposed; it can perform denoising with only the target image. In this study, we propose an innovative procedure for the DIP approach with a four-dimensional (4D) branch CNN architecture in end-to-end training to denoise dynamic PET images. Our proposed 4D CNN architecture can be applied to end-to-end dynamic PET image denoising by introducing a feature extractor and a reconstruction branch for each time frame of the dynamic PET image. In the proposed DIP method, it is not necessary to prepare high-quality and large patient-related PET images. Instead, a subject's own static PET image is used as additional information, dynamic PET images are treated as training labels, and denoised dynamic PET images are obtained from the CNN outputs. Both simulation with [¹⁸F]fluoro-2-deoxy-D-glucose (FDG) and preclinical data with [¹⁸F]FDG and [¹¹C]raclopride were used to evaluate the proposed framework. The results showed that our 4D DIP framework quantitatively and qualitatively outperformed 3D DIP and other unsupervised denoising methods. The proposed 4D DIP framework thus provides a promising procedure for dynamic PET image denoising.

Imaging mitochondrial complex I activation during a vibrotactile stimulation: A PET study using [18F]BCPP-EF in the conscious monkey brain

In order to evaluate the capability of 2-tert-butyl-4-chloro-5-{6-[2-(2-[¹⁸F]fluoroethoxy)-ethoxy]-pyridin-3-ylmethoxy}-2H-pyridazin-3-one ([¹⁸F]BCPP-EF), a novel positron emission tomography (PET) probe for mitochondrial complex I (MC-I) activity, to assess neuronal activation, an activation PET study was conducted in the conscious monkey brain with a continuous unilateral vibrotactile stimulation. PET scans with [¹⁵O]H₂O, [¹⁸F]BCPP-EF, or 2-deoxy-2-[¹⁸F]fluoroglucose ([¹⁸F]FDG) were conducted under: (1) resting conditions; (2) a continuous vibration stimulation; (3) a continuous vibration stimulation after 15-min pre-vibration; and (4) a continuous vibration stimulation after 30-min pre-vibration. The contralateral/ipsilateral ratio (CIR) in the somatosensory cortex showed significant increases in the uptake of [¹⁵O]H₂O, [¹⁸F]BCPP-EF, and [¹⁸F]FDG with the vibration stimulation. The longer pre-vibration duration induced significantly lower CIR in regional cerebral blood flow (rCBF) measured using [¹⁵O]H₂O, whereas it did not affect the CIR in [¹⁸F]BCPP-EF or the regional cerebral metabolic rate of glucose (rCMRglc) measured using [¹⁸F]FDG 30-60 min after the injection. These results suggest that the [¹⁸F]BCPP-EF response in the later phase of scans was not influenced by the increase in rCBF, indicating the capability of [¹⁸F]BCPP-EF to detect acute changes in MC-I activity induced by neuronal activation. However, the metabolic shift from glycolysis to oxidation was not observed under the stimulation used here.

Chemical and molecular bases of dome formation in human colorectal cancer cells mediated by sulphur compounds from Cucumis melo var. conomon

Colorectal cancer was the third most commonly diagnosed malignant tumor and the fourth leading cause of cancer deaths worldwide in 2012. A human colorectal cancer cell line, RCM‐1, was established from a colon cancer tissue diagnosed as a well‐differentiated rectum adenocarcinoma. RCM‐1 cells spontaneously form ‘domes’ (formerly designated ‘ducts’) resembling villiform structures. Two sulphur‐containing compounds from Cucumis melo var. conomon (Katsura‐uri, or Japanese pickling melon), referred to as 3‐methylthiopropionic acid ethyl ester (MTPE) and methylthioacetic acid ethyl ester (MTAE), can induce the differentiation of the unorganized cell mass of an RCM‐1 human colorectal cancer cell culture into a dome. However, the underlying molecular mechanisms of such dome formation have not been previously reported. Here, we performed a structure–activity relationship analysis, which indicated that methylthioacetic acid (MTA) was the lowest molecular weight compound with the most potent dome‐inducing activity among 37 MTPE and MTAE analogues, and the methylthio group was essential for this activity. According to our microarray analysis, MTA resulted in down‐regulation of 537 genes and up‐regulation of 117 genes. Furthermore, MTA caused down‐regulation of many genes involved in cell‐cycle control, with the cyclin E2 (CCNE2) and cell division cycle 25A (CDC25A) genes being the most significantly reduced. Pharmacological analysis showed that the administration of two cell‐cycle inhibitors for inactivating CDC25A phosphatase (NSC95397) and the cyclin E2/cyclin‐dependent kinase 2 complex (purvalanol A) increased the dome number independently of MTA. Altogether, our results indicate that MTA is the minimum unit required to induce dome formation, with the down‐regulation of CDC25A and possibly CCNE2 being important steps in this process.

Microbial Community Structure and Chemical Constituents in Shinkiku, a Fermented Crude Drug Used in Kampo Medicine

Shinkiku (Massa Medicata Fermentata) is a traditional crude drug used to treat anorexia and dyspepsia of elder patients in east Asia. Shinkiku is generally prepared by the microbial fermentation of wheat and herbs. Shinkiku is also used in Japanese Kampo medicine as a component of (Hangebyakujutsutemmato). However, the quality of shinkiku varies by manufacture because there are no reference standards to control the quality of medicinal shinkiku. Thus, we aim to characterize the quality of various commercially available shinkiku by chemical and microbial analysis. We collected 13 shinkiku products manufactured in China and Korea and investigated the microbial structure and chemical constituents. Amplicon sequence analysis revealed that Aspergillus sp. was common microorganism in shinkiku products. Digestive enzymes (α-amylase, protease, and lipase), organic acids (ferulic acid, citric acid, lactic acid, and acetic acid), and 39 volatile compounds were commonly found in shinkiku products. Although there were some commonalities in shinkiku products, microbial and chemical characteristic considerably differed as per the manufacturer. Aspergillus sp. was predominant in Korean products, and Korean products showed higher enzyme activities than Chinese products. Meanwhile, Bacillus sp. was commonly detected in Chinese shinkiku, and ferulic acid was higher in Chinese products. Principal component analysis based on the GC-MS peak area of the volatiles also clearly distinguished shinkiku products manufactured in China from those in Korea. Chinese products contained higher amounts of benzaldehyde and anethole than Korean ones. Korean products were further separated into two groups: one with relatively higher linalool and terpinen-4-ol and another with higher hexanoic acid and 1-octen-3-ol. Thus, our study revealed the commonality and diversity of commercial shinkiku products, in which the commonalities can possibly be the reference standard for quality control of shinkiku, and the diversity suggested the importance of microbial management to stabilize the quality of shinkiku.

Single-base deletion in GmCHR5 increases the genistein-to-daidzein ratio in soybean seed

Novel mutant alleles related to isoflavone content are useful for breeding programs to improve the disease resistance and nutritional content of soybean. However, identification of mutant alleles from high-density mutant libraries is expensive and time-consuming because soybean has a large, complicated genome. Here, we identified the gene responsible for increased genistein-to-daidzein ratio in seed of the mutant line F333ES017D9. For this purpose, we used a time- and cost-effective approach based on selective genotyping of a small number of F₂ plants showing the mutant phenotype with nearest-neighboring-nucleotide substitution-high-resolution melting analysis markers, followed by alignment of short reads obtained by next-generation sequencing analysis with the identified locus. In the mutant line, GmCHR5 harbored a single-base deletion that caused a change in the substrate flow in the isoflavone biosynthetic pathway towards genistein. Mutated GmCHR5 was expressed at a lower level during seed development than wild-type GmCHR5. Ectopic overexpression of GmCHR5 increased the production of daidzein derivatives in both the wild-type and mutant plants. The present strategy will be useful for accelerating identification of mutant alleles responsible for traits of interest in agronomically important crops.

Identification of novel MYB transcription factors involved in the isoflavone biosynthetic pathway by using the combination screening system with agroinfiltration and hairy root transformation

Soybean isoflavones are functionally important secondary metabolites that are mainly accumulated in seeds. Their biosynthetic processes are regulated coordinately at the transcriptional level; however, screening systems for key transcription factors (TFs) are limited. Here we developed a combination screening system comprising a simple agroinfiltration assay and a robust hairy root transformation assay. First, we screened for candidate MYB TFs that could activate the promoters of the chalcone synthase (CHS) gene GmCHS8 and the isoflavone synthase (IFS) genes GmIFS1 and GmIFS2 in the isoflavone biosynthetic pathway. In the agroinfiltration assay, we co-transformed a LjUbi (Lotus japonicus polyubiquitin gene) promoter-fused MYB gene with target promoter-fused GUS (β-glucuronidase) gene constructs, and identified three genes (GmMYB102, GmMYB280, and GmMYB502) as candidate regulators of isoflavone biosynthesis. We then evaluated the functional regulatory role of identified three MYB genes in isoflavone biosynthesis using hairy roots transformation assay in soybean for the accumulation of isoflavones. Three candidate MYB genes showed an increased accumulation of total isoflavones in hairy root transgenic lines. Accumulation of total isoflavones in the three MYB-overexpressing lines was approximately 2-to 4-folds more than that in the vector control, confirming their possible role to regulate isoflavone biosynthesis. However, the significant accumulation of authentic GmCHS8, GmIFS1, and GmIFS2 transcripts could not be observed except for the GmMYB502-overexpressing line. Therefore, the analysis of isoflavone accumulation in transgenic hairy root was effective for evaluation of transactivation activity of MYB TFs for isoflavone biosynthetic genes. Our results demonstrate a simple and robust system that can potentially identify the function of orphan TFs in diverse plant metabolic pathways.

Automated segmentation of 2D low-dose CT images of the psoas-major muscle using deep convolutional neural networks

The psoas-major muscle has been reported as a predictive factor of sarcopenia. The cross-sectional area (CSA) of the psoas-major muscle in axial images has been indicated to correlate well with the whole-body skeletal muscle mass. In this study, we evaluated the segmentation accuracy of low-dose X-ray computed tomography (CT) images of the psoas-major muscle using the U-Net convolutional neural network, which is a deep-learning technique. Deep learning has been recently known to outperform conventional image-segmentation techniques. We used fivefold cross validation to validate the segmentation performance (n = 100) of the psoas-major muscle. For the intersection over union and CSA ratio, segmentation accuracies of 86.0 and 103.1%, respectively, were achieved. These results suggest that the U-Net network is competitive compared with the previous methods. Therefore, the proposed technique is useful for segmenting the psoas-major muscle even in low-dose CT images.

Evaluation of D-isomers of 4-borono-2-18F-fluoro-phenylalanine and O-11C-methyl-tyrosine as brain tumor imaging agents: a comparative PET study with their L-isomers in rat brain glioma

Background

The potential of the D-isomerization of 4-borono-2-¹⁸F-fluoro-phenylalanine (¹⁸F-FBPA) to improve its target tumor to non-target normal brain tissue ratio (TBR) was evaluated in rat brain glioma and compared with those of L- and D-¹¹C-methyl-tyrosine (¹¹C-CMT).

The L- or D-isomer of ¹⁸F-FBPA was injected into rats through the tail vein, and their whole body kinetics and distributions were assessed using the tissue dissection method up to 90 min after the injection. The kinetics of L- and D-¹⁸F-FBPA or L- and D-¹¹C-CMT in the C-6 glioma-inoculated rat brain were measured for 90 or 60 min, respectively, using high-resolution animal PET, and their TBRs were assessed.

Results

Tissue dissection analyses showed that D-¹⁸F-FBPA uptake was significantly lower than that of L-¹⁸F-FBPA in the brain and abdominal organs, except for the kidney and bladder, reflecting the faster elimination rate of D-¹⁸F-FBPA than L-¹⁸F-FBPA from the blood to the urinary tract. PET imaging using ¹⁸F-FBPA revealed that although the brain uptake of D-¹⁸F-FBPA was significantly lower than that of L-¹⁸F-FBPA, the TBR of the D-isomer improved to 6.93 from 1.45 for the L-isomer. Similar results were obtained with PET imaging using ¹¹C-CMT with a smaller improvement in TBR to 1.75 for D-¹¹C-CMT from 1.33 for L-¹¹C-CMT.

Conclusions

The present results indicate that D-¹⁸F-FBPA is a better brain tumor imaging agent with higher TBR than its original L-isomer and previously reported tyrosine-based PET imaging agents. This improved TBR of D-¹⁸F-FBPA without any pre-treatments, such as tentative blood-brain barrier disruption using hyperosmotic agents or sonication, suggests that the D-isomerization of BPA results in the more selective accumulation of ¹⁰B in tumor cells that is more effective and less toxic than conventional L-BPA.

Influence of Fruit Ripening on Color, Organic Acid Contents, Capsaicinoids, Aroma Compounds, and Antioxidant Capacity of Shimatogarashi (Capsicum frutescens)

Shimatogarashi (Capsicum frutescens) is a typical chili pepper domesticated in southern Japan. Important traits of Shimatogarashi peppers, such as color; proportion of organic acids, capsaicinoids, and aromatic compounds; and antioxidant activity in three stages of maturity (green (immature), orange (turning), and red (mature) stages) were characterized. The results indicated that the concentration of organic acids, including ascorbic, citric, and malic acid, increased during ripening. In addition, the amount of capsaicinoids, which are responsible for the pungent taste of chili peppers, increased as the fruit matured to the orange and red stages. The volatile compound profile of Shimatogarashi was dominated by the presence of esters, which mainly contributed to fruity notes. The total amount of volatile compounds analyzed by gas chromatography-headspace solid-phase microextraction (GC-HS-SPME), especially esters, decreased as the fruit changed in color from green to red. This was in contrast to the amount of terpenoids, especially limonene, which increased at the red stage, denoting a change in flavor from fruity to a more citrus-like aroma. Based on the total phenolic content (TPC), the oxygen radical absorbance capacity (ORAC) and the diphenylpicrylhydrazyl (DPPH) free radical method, the antioxidant capacity of Shimatogarashi showed an increase at the mature red stage. However, while the red stage showed higher pungency and antioxidant capacity as well as an attractive color, the results of aromatic compound analysis revealed that the immature green stage had the advantages of having pleasant fruity smell, making it suitable for use in condiments.

Anti-inflammatory activity and molecular mechanism of Oolong tea theasinensin

Oolong tea theasinensins are a group of tea polyphenols different from green tea catechins and black tea theaflavins, and they are considered as bioactive compounds in Oolong tea. In the present study, based on the properties of theasinensin and information about inflammatory processes, we investigated the anti-inflammatory activity and molecular mechanisms of theasinensin A (TSA) in both cell and animal models. In the cell model, TSA reduced the levels of pro-inflammatory mediators including inducible nitric oxide synthase (iNOS), nitric oxide (NO), interleukin-12 (IL-12) (p70), tumor necrosis factor alpha (TNF-α), and monocyte chemotactic protein-1 (MCP-1) induced by lipopolysaccharide (LPS). Cellular signaling analysis revealed that TSA downregulated MAPK/ERK kinase (MEK)-extracellular signal-regulated kinase (ERK) signaling. Pull-down assay and affinity data revealed that TSA might directly bind to MEK-ERK for the inhibitory action. In the animal model, TSA suppressed the production of IL-12 (p70), TNF-α, and MCP-1 and attenuated mouse paw edema induced by LPS.

In vitro anticancer activity of loquat tea by inducing apoptosis in human leukemia cells

Fresh loquat leaves have been used as folk health herb in Asian countries for long time, although the evidence supporting their functions is still minimal. This study aimed to clarify the chemopreventive effect of loquat tea extract (LTE) by investigating the inhibition on proliferation, and underlying mechanisms in human promyelocytic leukemia cells (HL-60). LTE inhibited proliferation of HL-60 in a dose-dependent manner. Molecular data showed that the isolated fraction of LTE induced apoptosis of HL-60 as characterized by DNA fragmentation; activation of caspase-3, -8, and -9; and inactivation of poly(ADP)ribose polymerase. Moreover, LTE fraction increased the ratio of pro-apoptotic Bcl-2-associated X protein (Bax)/anti-apoptotic myeloid cell leukemia 1 (Mcl-1) that caused mitochondrial membrane potential loss and cytochrome c released to cytosol. Thus, our data indicate that LTE might induce apoptosis in HL-60 cells through a mitochondrial dysfunction pathway. These findings enhance our understanding for chemopreventive function of loquat tea.

Genetic analysis of ion-beam induced extremely late heading mutants in rice

Two extremely late heading mutants were induced by ion beam irradiation in rice cultivar 'Taichung 65': KGM26 and KGM27. The F2 populations from the cross between the two mutants and Taichung 65 showed clear 3 early: 1 late segregation, suggesting control of late heading by a recessive gene. The genes identified in KGM26 and KGM27 were respectively designated as FLT1 and FLT2. The two genes were mapped using the crosses between the two mutants and an Indica cultivar 'Kasalath'. FLT1 was located on the distal end of the short arm of chromosome 8. FLT2 was located around the centromere of chromosome 9. FLT1 might share the same locus as EHD3 because their chromosomal location is overlapping. FLT2 is inferred to be a new gene because no gene with a comparable effect to that of this gene was mapped near the centromere of chromosome 9. In crosses with Kasalath, homozygotes of late heading mutant genes showed a large variation of days to heading, suggesting that other genes affected late heading mutant genes.

Antioxidant and α-Amylase Inhibitory Compounds from Aerial Parts ofVarthemia iphionoidesBoiss

Various extracts of aerial parts of Varthemia (Varthemia iphionoides Boiss) were investigated for radical-scavenging activity, antioxidative activity, and porcine pancreas alpha-amylase inhibitory activity. The ethanol and water extracts showed a pronounced 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical-scavenging activity, with inhibition of about 90% at a concentration of 100 microg/ml, and alpha-amylase inhibitory activity of about 70% at a concentration of 200 microg/ml by the 2-chloro-4-nitrophenyl alpha-maltotrioside (CNP-G3) degradation method. The ethanol extract was purified by column chromatography to give seven 3-methoxyflavones (1-7) and eudesmane sesquiterpene, selina-4,11(13)-dien-3-on-12-oic acid (8). The structures of these compounds were established by NMR, MS, and UV spectroscopy. Of 3-methoxyflavones, 5,7,4'-trihydroxy-3,6-dimethoxyflavone (1), 5,7,4'-trihydroxy-3,3'-dimethoxyflavone (2), and 5,4'-dihydroxy-3,7,3'-trimethoxyflavone (3,7,3'-tri-O-methyl-quercetin) (7) exhibited pronounced radical-scavenging activity. The antioxidative activity in the linoleic acid system was considerable in compounds 1, 2, and 5,4'-dihydroxy-3,6,7-trimethoxyflavone (4). Compounds 1, 2, 4, 5 (5,7,4'-trihydroxy-3-methoxyflavone), and 6 (5,4'-dihydroxy-3,7-dimethoxyflavone) showed markedly high inhibitory activity against porcine pancreas alpha-amylase. Eudesmane sesquiterpene did not show any activity.

Antioxidant and anti-inflammatory activities of loquat (Eriobotrya japonica) tea

Background: Fresh loquat leaves contain several kinds of flavonoids and have been reported to have preventive effects against some human diseases such as diabetes, coughs and ulcers,. Recently, fresh loquat leaves in Japan were processed to a beverage, called loquat tea, after the fresh leaves are roasted at 350C for 30 minutes. However, the scientific evidence supporting the functions of these processed leaves is still minimal.Objective: The aim of this study is to investigate the antioxidant and anti-inflammatory activities of roasted loquat tea extract (LTE) in vitro and in culture cells.Methods: Bioactive fractions of LTE were separated by column chromatograph. Antioxidant activities were determined by DPPH and ROS assay. Pro-inflammatory mediators cyclooxygenase-2 (COX-2) and prostaglandin E2 (PGE2) were determined by Western blot and ELISA assay, respectively. Chemical quantification and characterization were analyzed by HPLC, FR-IR, and NMR. Phenolic content was measured by Folin-Ciocalteu assay.Results: The results showed that loquat tea extract (LTE) possessed stronger DPPH scavenging activity than fresh. Cellular data revealed that LTE inhibited the production of reactive oxygen species (ROS), and further suppressed the production of COX-2 and PGE2 in lipopolysaccharide (LPS)-activated RAW 264.7 cells. Chemical quantification and characterization data indicated that LTE contained new bioactive phenolic components that were produced from the roasting processes of fresh loquat leaves.Conclusions: Loquat tea made from roasted loquat leaves contained new bioactive phenolic compounds that contribute to its antioxidant and anti-inflammatory activities.Keywords: Loquat tea, Antioxidant activity, Anti-inflammatory activity, Chemical characterization

Chemical taxonomy of red-flowered wild Camellia species based on floral anthocyanins

This study uses anthocyanins in the red flowers of section Camellia as taxonomic markers to investigate the phenetic relationships among 33 wild species from China, Taiwan, and Japan. The 25 anthocyanins from section Camellia produced 38 pigment patterns that serve as phenetic markers. Principal Component Analysis (PCA) indicated that the attachment of one or two glucoses to the cyanidin-core structure at the 3- or the 3- and 5-positions, respectively, was the most influential pattern against the first factor, Z₁. In addition, two alternative pigment patterns, acylated or non-acylated, and the structural isomerism (cis- or trans-) of the p-coumaroyl group were relatively significant patterns. Ward's minimum-variance cluster analysis (WMVCA) produced a dendrogram that consisted of two sub-clusters. One sub-cluster (A) was constructed by species that have mainly two types of anthocyanins: 3,5-di-O-β-glucosides (Camellia saluenensis) and sambubioside of cyanidin (Camellia reticulata). The other sub-cluster (B) was made up of the 3-O-β-glucosides of cyanidin (Camellia japonica) and delphinidin (Camellia hongkongensis), with a higher proportion of the 3-O-β-galactosides (Camellia mairei and Camellia boreali-yunnanica). The former group showed a higher proportion of acylation, over 63%, but with the exception of Camellia azalea. The latter group showed less than 52% acylation, but with the exception of C. hongkongensis and C. boreali-yunnanica. PCA and WMVCA indicated that the greater the amount of di-O-glycosides and acylation, the more primitive anthocyanin traits the species possess. Based on these results, in conjunction with geographical and literary information, the data suggest that the Xinan district is the site/center of origin for the red-flowered Camellia species of which both C. saluenensis and C. reticulata have the most primitive anthocyanin traits.

Oolong tea theasinensins attenuate cyclooxygenase-2 expression in lipopolysaccharide (LPS)-activated mouse macrophages: structure-activity relationship and molecular mechanisms

Oolong tea theasinensins are a group of tea polyphenols different from green tea catechins and black tea theaflavins. The present study reports the inhibitory effects of oolong tea theasinensins on the expression of cyclooxygenase-2 (COX-2) and underlying molecular mechanisms in lipopolysaccharide (LPS)-activated murine macrophage RAW264 cells. The structure-activity data revealed that the galloyl moiety of theasinensins played an important role in the inhibitory actions. Theasinensin A, a more potent inhibitor, caused a dose-dependent inhibition of mRNA, protein, and promoter activity of COX-2. An electrophoretic mobility shift assay (EMSA) revealed that theasinensin A reduced the complex of NF-κB- and AP-1-DNA in the promoter of COX-2. Signaling analysis demonstrated that theasinensin A attenuated IκB-α degradation, nuclear p65 accumulation, and c-Jun phosphorylation. Furthermore, theasinensin A suppressed the phosphorylation of MAPKs, IκB kinase α/β (IKKα/β), and TGF-β activated kinase (TAK1). These data demonstrated that the down-regulation of TAK1-mediated MAPKs and NF-κB signaling pathways might be involved in the inhibition of COX-2 expression by theasinensin A. These findings provide the first molecular basis for the anti-inflammatory properties of oolong tea theasinensins.

Chemosystematics of tea trees based on tea leaf polyphenols as phenetic markers

This study examined the polyphenols of tea leaves as chemotaxonomic markers to investigate the phenetic relationship between 89 wild (the small-leaved C.sinensis var. sinensis and large-leaved C. sinensis var. assamica), hybrid, and cultivated tea trees from China and Japan. (-)-Epigallocatechin 3-O-gallate, EGCG (1); (-)-epigallocatechin, EGC (2); (-)-epicatechin 3-O-gallate, ECG (3); (-)-epicatechin, EC (4); (+)-catechin, CA (5); strictinin, STR (6); and gallic acid, GA (7) were used as polyphenolic markers. Of the 13 polyphenol patterns observed, Principal Component Analysis (PCA) indicated that the structure-types of the flavonoid B-rings, such as the pyrogallol-(EGCG (1) and EGC (2)) and catechol-(ECG (3) and EC (4)) types, greatly influenced the classification. Ward's minimum-variance cluster analysis was used to produce a dendrogram that consisted of three sub-clusters. One sub-cluster (A) was composed of old tea trees 'Gushu' cha (C. sinensis var. assamica) and cv 'Taidi' cha, suggesting that relatively primitive tea trees contain greater amounts of compounds 3 and 4 and lower amounts of compounds 1 and 2. The other two sub-clusters B and C, made up of Chinese hybrids (sub-cluster B) and Japanese and Taiwanese tea trees (sub-cluster C), had lower contents of 3 and 4 than sub-cluster A. Therefore, PCA and cluster analysis indicated that the greater the amounts of 1 and 2 (and the lower of 3 and 4), the more recent the origin of the tea line. Based on morphological characteristics, geographical information, and the historical information on tea trees, these results show good agreement with the current theory of tea tree origins, and this suggests that the Xishuangbanna district and Puer City are among the original sites of the tea tree species.

A new acylated anthocyanin from the red flowers of Camellia hongkongensis and characterization of anthocyanins in the section Camellia species

Twelve anthocyanins (1-12) were isolated from the red flowers of Camellia hongkongensis Seem. by chromatography using open columns. Their structures were elucidated on the basis of spectroscopic analyses, that is, proton-nuclear magnetic resonance, carbon 13-nuclear magnetic resonance, heteronuclear multiple quantum correlation, heteronuclear multiple bond correlation, high resolution electrospray ionization mass and ultraviolet visible spectroscopies. Out of these anthocyanins, a novel acylated anthocyanin, cyanidin 3-O-(6-O-(Z)-p-coumaroyl)-beta-galactopyranoside (6), two known acylated anthocyanins, cyanidin 3-O-(6-O-(E)-p-coumaroyl)-beta-galactopyranoside (7) and cyanidin 3-O-(6-O-(E)-caffeoyl)-beta-galactopyranoside (8), and three known delphinidin glycosides (10-12) were for the first time isolated from the genus Camellia. Furthermore, pigment components in C. japonica L., C. chekiangoleosa Hu and C. semiserrata Chi were studied. The results indicated that the distribution of anthocyanins was differed among these species. Delphinidin glycoside was only detected in the flowers of C. hongkongensis, which is a special and important species in the section Camellia. Based on the characterization of anthocyanins in the section Camellia species, there is a close relationship among these species, and C. hongkongensis might be an important parent for creating new cultivars with bluish flower color.

Anthocyanins from red flowers of Camellia cultivar 'Dalicha'

Five anthocyanins, cyanidin 3-O-(2-O-beta-xylopyranosyl-6-O-(Z)-p-coumaroyl)-beta-galactopyranoside (2), cyanidin 3-O-(2-O-beta-xylopyranosyl-6-O-(E)-p-coumaroyl)-beta-galactopyranoside (3), cyanidin 3-O-(2-O-beta-xylopyranosyl-6-O-(E)-caffeoyl)-beta-galactopyranoside (4), cyanidin 3-O-(2-O-beta-xylopyranosyl-6-O-acetyl)-beta-galactopyranoside (5), and cyanidin 3-O-(2-O-beta-xylopyranosyl-6-O-acetyl)-beta-glucopyranoside (6), together with the known cyanidin 3-O-(2-O-beta-xylopyranosyl)-beta-galactopyranoside (1), were isolated from red flowers of Camellia cultivar 'Dalicha' (Camellia reticulata) by chromatography using open columns. Their structures were subsequently determined on the basis of spectroscopic analyses, i.e., 1H NMR, 13C NMR, HMQC, HMBC, HR ESI-MS and UV-vis.

Prodelphinidin B-4 3'-O-gallate, a tea polyphenol, is involved in the inhibition of COX-2 and iNOS via the downregulation of TAK1-NF-kappaB pathway

Much is known about the bioactive properties of green tea flavan-3-ol. However, very little work has been done to determine the properties of proanthocyanidins, another kind of polyphenols in green tea. In this study, we have investigated the anti-inflammatory effect of tea prodelphinidin B-4 3'-O-gallate (PDG) by demonstrating the inhibitory effects on cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (iNOS) in lipopolysaccharide (LPS)-activated murine macrophage RAW264 cells. PDG caused a dose-dependent inhibition of COX-2 and iNOS at both mRNA and protein levels with the attendant decrease of prostaglandin E2 (PGE2) and nitric oxide (NO) production. Molecular data revealed that PDG downregulated NF-kappaB signaling pathway. Electrophoretic mobility shift assay (EMSA) showed that PDG reduced the binding complex of NF-kappaB-DNA in the promoter of COX-2 and iNOS. Immunochemical analysis revealed that PDG suppressed LPS-induced phosphorylation and degradation of IkappaBalpha, and subsequent nuclear translocation of p65. Consequently, PDG suppressed phosphorylation of IkappaB kinase alpha/beta (IKKalpha/beta) and TGF-beta-activated kinase (TAK1). Taken together, our data indicated that PDG is involved in the inhibition of COX-2 and iNOS via the downregulation of TAK1-NF-kappaB pathway, revealing partial molecular basis for the anti-inflammatory properties of tea PDG.