To stimulate progress in automating the reconstruction of neural circuits, we organized the first international challenge on 2D segmentation of electron microscopic (EM) images of the brain. Participants submitted boundary maps predicted for a test set of images, and were scored based on their agreement with a consensus of human expert annotations. The winning team had no prior experience with EM images, and employed a convolutional network. This “deep learning” approach has since become accepted as a standard for segmentation of EM images. The challenge has continued to accept submissions, and the best so far has resulted from cooperation between two teams. The challenge has probably saturated, as algorithms cannot progress beyond limits set by ambiguities inherent in 2D scoring and the size of the test dataset. Retrospective evaluation of the challenge scoring system reveals that it was not sufficiently robust to variations in the widths of neurite borders. We propose a solution to this problem, which should be useful for a future 3D segmentation challenge.

No abstract

Text detection and recognition in natural images have long been considered as two separate tasks that are processed sequentially. Training of two tasks in a unified framework is non-trivial due to significant differences in optimisation difficulties. In this work, we present a conceptually simple yet efficient framework that simultaneously processes the two tasks in one shot. Our main contributions are three-fold: 1) we propose a novel text-alignment layer that allows it to precisely compute convolutional features of a text instance in arbitrary orientation, which is the key to boost the performance; 2) a character attention mechanism is introduced by using character spatial information as explicit supervision, leading to large improvements in recognition; 3) two technologies, together with a new RNN branch for word recognition, are integrated seamlessly into a single model which is end-to-end trainable. This allows the two tasks to work collaboratively by sharing convolutional features, which is critical to identify challenging text instances. Our model achieves impressive results in end-to-end recognition on the IC-DAR2015 [1] dataset, significantly advancing most recent results [2], with improvements of F-measure from (0.54, 0.51, 0.47) to (0.82, 0.77, 0.63), by using a strong, weak and generic lexicon respectively. Thanks to joint training, our method can also serve as a good detector by achieving a new state-of-the-art detection performance on two datasets.

Bioassay-directed fractionation of an EtOH extract of Curcuma zedoaria led to isolation of an active curcuminoid, which was identified as demethoxycurcumin (2) by comparison of its 1H and 13C NMR spectra with literature data and by direct comparison with synthetic material. Curcumin (1) and bisdemethoxycurcumin (3) were also obtained. Curcuminoids (1-3) were synthesized and demonstrated to be cytotoxic against human ovarian cancer OVCAR-3 cells. The observed CD50 values of 1, 2, and 3 were 4.4, 3.8, and 3.1 microg/mL, respectively. Three additional novel compounds, 3, 7-dimethylindan-5-carboxylic acid (4), curcolonol (5), and guaidiol (6), were also isolated from the EtOH extract. The structures and relative stereochemistry of 4-6 were determined by spectroscopic methods and X-ray crystallographic analysis.

Automatic extraction of liver and tumor from CT volumes is a challenging task due to their heterogeneous and diffusive shapes. Recently, 2D deep convolutional neural networks have become popular in medical image segmentation tasks because of the utilization of large labeled datasets to learn hierarchical features. However, few studies investigate 3D networks for liver tumor segmentation. In this paper, we propose a 3D hybrid residual attention-aware segmentation method, i.e., RA-UNet, to precisely extract the liver region and segment tumors from the liver. The proposed network has a basic architecture as U-Net which extracts contextual information combining low-level feature maps with high-level ones. Attention residual modules are integrated so that the attention-aware features change adaptively. This is the first work that an attention residual mechanism is used to segment tumors from 3D medical volumetric images. We evaluated our framework on the public MICCAI 2017 Liver Tumor Segmentation dataset and tested the generalization on the 3DIRCADb dataset. The experiments show that our architecture obtains competitive results.

This paper presents the evaluation results of the methods submitted to Challenge US: Biometric Measurements from Fetal Ultrasound Images, a segmentation challenge held at the IEEE International Symposium on Biomedical Imaging 2012. The challenge was set to compare and evaluate current fetal ultrasound image segmentation methods. It consisted of automatically segmenting fetal anatomical structures to measure standard obstetric biometric parameters, from 2D fetal ultrasound images taken on fetuses at different gestational ages (21 weeks, 28 weeks, and 33 weeks) and with varying image quality to reflect data encountered in real clinical environments. Four independent sub-challenges were proposed, according to the objects of interest measured in clinical practice: abdomen, head, femur, and whole fetus. Five teams participated in the head sub-challenge and two teams in the femur sub-challenge, including one team who tackled both. Nobody attempted the abdomen and whole fetus sub-challenges. The challenge goals were two-fold and the participants were asked to submit the segmentation results as well as the measurements derived from the segmented objects. Extensive quantitative (region-based, distance-based, and Bland-Altman measurements) and qualitative evaluation was performed to compare the results from a representative selection of current methods submitted to the challenge. Several experts (three for the head sub-challenge and two for the femur sub-challenge), with different degrees of expertise, manually delineated the objects of interest to define the ground truth used within the evaluation framework. For the head sub-challenge, several groups produced results that could be potentially used in clinical settings, with comparable performance to manual delineations. The femur sub-challenge had inferior performance to the head sub-challenge due to the fact that it is a harder segmentation problem and that the techniques presented relied more on the femur's appearance.

Bioassay-directed fractionation of extract of Arnebia euchroma led to the isolation of alkannin (1), shikonin (2), and their derivatives (3-8) as the active principles against methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant enterococci (VRE). The stereochemistry of alpha-methylbutyryl alkannin (8) is revealed for the first time, and the antimicrobial activity of 8 was compared with its corresponding diastereomer (9). The derivatives 3-9 showed stronger anti-MRSA activity [minimum inhibitory concentrations (MICs) ranged from 1.56 to 3.13 microg/mL] than alkannin or shikonin (MIC = 6.25 microg/mL). Anti-MRSA activity of derivatives was bactericidal with minimum bactericidal concentration (MBC)/MIC < or = 2. In a time-kill assay, the bactericidal activity against MRSA was achieved as rapidly as 2 h. The derivatives 3-9 were also active against vancomycin-resistant Enterococcus faecium (F935) and vancomycin-resistant Enterococcus faecalis (CKU-17) with MICs similar to those with MRSA. Aromatic ester derivatives were also synthesized for antimicrobial activity comparison. None of these compounds were active against Gram-negative bacteria tested. Their cytotoxicity was also evaluated on selected cancer cell lines, and they expressed their activity in the range 0.6-5.4 microg/mL (CD(50)). Our results indicate that the ester derivatives of alkannin are potential candidates of anti-MRSA and anti-VRE agents with antitumor activity.

Bioassay-directed fractionation of Saussurea lappa led to the isolation of a novel lappadilactone (1) and seven sesquiterpene lactones (2-8) as cytotoxic principles against selected human cancer cell lines. Lappadilactone (1), dehydrocostuslactone (2), and costunolide (5) exhibited the most potent cytotoxicity with CD50 values in the range 1.6-3.5 microg/mL in dose- and time-dependent manners. The cytotoxicities were not specific and showed similar activities against HepG2, OVCAR-3 and HeLa cell lines. The structure-activity relationship showed that the alpha-methylene-gamma-lactone moiety is necessary for cytotoxicity, and activity is reduced with the presence of a hydroxyl group. In addition, seven noncytotoxic compounds (9-15) were also isolated, including two novel sesquiterpenes, a guaianolide-type with a C17 skeleton, lappalone (13), and 1beta,6alpha-dihydroxycostic acid ethyl ester (14). The structures of the new compounds were elucidated from spectroscopic and/or X-ray data interpretations. Some representative compounds were also tested for antibacterial activity; however, only marginal activities were observed. Therefore, compounds 1-8 are potential cytotoxic agents but without significant antibacterial effect.