Review
Ecology
Marc Besson, Jamie Alison, Kim Bjerge, Thomas E. Gorochowski, Toke T. Hoye, Tommaso Jucker, Hjalte M. R. Mann, Christopher F. Clements
Summary: High-resolution monitoring is crucial for understanding ecosystem dynamics in a time of global change and biodiversity declines. While real-time and automated monitoring of physical components has been possible, monitoring ecological communities has been more challenging. Recent technological advancements offer potential solutions to this challenge.
Article
Chemistry, Multidisciplinary
Yoshitaka Isobe, Atsushi Teramoto, Fujio Morita, Kuniaki Saito, Hiroshi Fujita
Summary: In computed tomography colonography (CTC), an electronic cleansing method using CycleGAN was developed to assist diagnosis and accurately visualize the colon.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Artificial Intelligence
Pablo J. Blanco, Paulo G. P. Ziemer, Carlos A. Bulant, Yasushi Ueki, Ronald Bass, Lorenz Raber, Pedro A. Lemos, Hector M. Garcia-Garcia
Summary: The study proposes a machine learning approach for automatically extracting lumen and vessel boundaries from IVUS datasets, delivering accurate segmentations and potentially useful clinical applications.
MEDICAL IMAGE ANALYSIS
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Faizan Ahmad, Wenguo Hou, Jing Xiong, Zeyang Xia
Summary: This article proposes a fully automatic pipeline for segmenting the right ventricle (RV), myocardium (MYO), and left ventricle (LV) using short-axis CMRI sequence images. A dilated residual network (DRN) is introduced to capture features at full resolution in the bottleneck of UNet, significantly increasing spatial and temporal information and maintaining localization accuracy. The method achieves superior performance and outperforms state-of-the-art methods in terms of accuracy and segmentation.
Article
Engineering, Biomedical
Xindi Liu, Kai Jin, Zehua Yang, Yan Yan, Shuai Wang, Yaqi Wang, Juan Ye
Summary: This study developed an automated deep learning system for quantitative analysis of the choroid with pathological changes in high myopia patients. The system achieved high accuracy in choroid segmentation and demonstrated correlation between choroidal structure and clinical characteristics. The findings suggest the potential of this system in exploring the pathogenesis of various fundus diseases.
PHYSICS IN MEDICINE AND BIOLOGY
(2022)
Article
Computer Science, Interdisciplinary Applications
Ryoji Hirano, Takuto Emura, Otoichi Nakata, Toshiharu Nakashima, Miyako Asai, Kuriko Kagitani-Shimono, Haruhiko Kishima, Masayuki Hirata
Summary: Magnetoencephalography (MEG) is a useful tool for clinical evaluation of interictal spike localization. Manual analysis by neurophysiologists is time-consuming and may not be cost-effective. This study proposes a fully automated AI-based approach called FAMED for spike identification and equivalent current dipole (ECD) estimation using deep learning and semantic segmentation techniques. FAMED achieved comparable performance to neurophysiologists and can contribute to the efficiency and consistency of MEG ECD analysis.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Multidisciplinary Sciences
Esraa A. Mohamed, Essam A. Rashed, Tarek Gaber, Omar Karam
Summary: This study proposes a fully automatic breast cancer detection system that utilizes infrared technology for detection and applies deep learning models for breast tissue classification. The experimental results show that the system achieves high accuracy and effectiveness, and is expected to be a helpful tool for clinical physicians.
Article
Radiology, Nuclear Medicine & Medical Imaging
Yu-Chun Lin, Gigin Lin, Sumit Pandey, Chih-Hua Yeh, Jiun-Jie Wang, Chien-Yu Lin, Tsung-Ying Ho, Sheung-Fat Ko, Shu-Hang Ng
Summary: This study utilized convolutional neural networks for fully automated segmentation and radiomics features extraction of hypopharyngeal cancer on MRI. DeepLab V3 + and U-Net models were trained and evaluated using various metrics. Results showed that DeepLab V3 + outperformed U-Net in terms of dice similarity coefficient, especially for small tumor volumes. Additionally, the radiomics features extracted by DeepLab V3 + exhibited higher reliability compared to U-Net for several parameters.
EUROPEAN RADIOLOGY
(2023)
Article
Computer Science, Software Engineering
Zhe Liu, Chunyang Chen, Junjie Wang, Yuekai Huang, Jun Hu, Qing Wang
Summary: Graphical User Interface (GUI) serves as a visual link between software application and users, enabling interaction between them. However, the complexity of GUI poses challenges to its implementation, as display issues often occur due to software or hardware compatibility. To address this, a fully automated approach called Nighthawk is proposed, which uses deep learning to detect and locate display issues in GUI screenshots for developers to fix. Additionally, a heuristic-based training data auto-generation method is introduced to generate labeled training data. Evaluation shows that Nighthawk achieves high precision and recall in detecting UI display issues and successfully uncovers previously-undetected issues in popular Android apps.
IEEE TRANSACTIONS ON SOFTWARE ENGINEERING
(2023)
Article
Multidisciplinary Sciences
Daniel Fernandez-Llaneza, Andrea Gondova, Harris Vince, Arijit Patra, Magdalena Zurek, Peter Konings, Patrik Kagelid, Leif Hultin
Summary: This paper presents a successful application of deep architectures 3D cardiac segmentation for rats in the preclinical context, which achieves performance comparable to human performance and provides an automated phase selection strategy.
SCIENTIFIC REPORTS
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Kristina I. Ringe, Van Dai Vo Chieu, Frank Wacker, Henrike Lenzen, Michael P. Manns, Christian Hundt, Bertil Schmidt, Hinrich B. Winther
Summary: The study demonstrates the feasibility of using transfer learning and extensive image augmentation to detect PSC-compatible cholangiographic changes on 3D-MRCP images with high sensitivity and low mean absolute error. Further validation with more and multicentric data is needed to address potential overfitting issues with neural networks.
EUROPEAN RADIOLOGY
(2021)
Review
Radiology, Nuclear Medicine & Medical Imaging
Nikita Sushentsev, Nadia Moreira Da Silva, Michael Yeung, Tristan Barrett, Evis Sala, Michael Roberts, Leonardo Rundo
Summary: This study systematically reviewed the current literature to evaluate the ability of fully-automated deep learning (DL) and semi-automated traditional machine learning (TML) MRI-based artificial intelligence (AI) methods in differentiating clinically significant prostate cancer from indolent prostate cancer and benign conditions. The study found comparable performance of the two classes of AI methods and identified common methodological limitations and biases.
INSIGHTS INTO IMAGING
(2022)
Review
Chemistry, Analytical
Giacomo Baccolo, Beatriz Quintanilla-Casas, Stefania Vichi, Dillen Augustijn, Rasmus Bro
Summary: Gas chromatography-mass spectrometry (GC-MS) is crucial in contemporary untargeted chemical analysis, but software uncertainty can lead to reproducibility issues. By utilizing tensor-based modeling and machine learning, a fully automated method was developed to convert GC-MS data into peak tables, improving analysis efficiency and reproducibility in comparison to current methods.
TRAC-TRENDS IN ANALYTICAL CHEMISTRY
(2021)
Article
Computer Science, Information Systems
Sujata Kulkarni, Rinku Rabidas
Summary: In this paper, a deep learning architecture based on U-Net is proposed for the detection and characterization of breast masses. The proposed architecture achieves a true positive rate of 99.64% with 0.25 false positives per image for the INbreast dataset, and 97.36% with 0.38 false positives per image for the DDSM dataset in the detection task. In the mass characterization task, the proposed architecture achieves an accuracy of 97.39% with an AUC of 0.97 for INbreast, and 96.81% with an AUC of 0.96 for DDSM. The introduced scheme outperforms state-of-the-art techniques in both tasks.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Environmental Sciences
Sean Krisanski, Mohammad Sadegh Taskhiri, Susana Gonzalez Aracil, David Herries, Allie Muneri, Mohan Babu Gurung, James Montgomery, Paul Turner
Summary: The research focuses on developing a robust, sensor-agnostic and fully automated forest point cloud measurement tool called the Forest Structural Complexity Tool (FSCT), which aims to improve the reliability and accuracy of forest measurement in research and industry.
Article
Multidisciplinary Sciences
Tomoki Ota, Akinobu Senoo, Masumi Shirakawa, Hiroshi Nonaka, Yutaro Saito, Sho Ito, Go Ueno, Satoru Nagatoishi, Kouhei Tsumoto, Shinsuke Sando
Summary: Bile acids play a crucial role in lipid digestion and absorption in the small bowel, as well as regulating their own metabolism and impacting other biological systems. The study suggests that secondary bile acid conjugates may act as modulators of serine hydroxymethyltransferase activity.
Article
Instruments & Instrumentation
Seiki Baba, Hiroaki Matsuura, Takashi Kawamura, Naoki Sakai, Yuki Nakamura, Yoshiaki Kawano, Nobuhiro Mizuno, Takashi Kumasaka, Masaki Yamamoto, Kunio Hirata
Summary: This study demonstrates the importance of selecting a radiation dose lower than 10 MGy for de novo structure determination using SWSX. Data collection at a dose of 5 MGy is shown to be optimal for balancing signal availability while minimizing damage.
JOURNAL OF SYNCHROTRON RADIATION
(2021)
Article
Biology
Akinobu Senoo, Sho Ito, Satoru Nagatoishi, Yutaro Saito, Go Ueno, Daisuke Kuroda, Kouhei Yoshida, Takumi Tashima, Shota Kudo, Shinsuke Sando, Kouhei Tsumoto
Summary: The study identified a chemical fragment specific to P-cadherin that inhibits P-cadherin-mediated cell adhesion. Despite being a fragment compound, it indirectly prevents the formation of hydrogen bonds necessary for X dimer formation by binding to a cavity of P-cadherin, thereby modulating the process of X dimerization.
COMMUNICATIONS BIOLOGY
(2021)
Article
Multidisciplinary Sciences
Nicholas Pearce, Katherine E. A. Reynolds, Surajit Kayal, Xue Z. Sun, E. Stephen Davies, Ferdinando Malagreca, Christian J. Schuermann, Sho Ito, Akihito Yamano, Stephen P. Argent, Michael W. George, Neil R. Champness
Summary: This study demonstrates the ability to control photoinduced charge transfer within homo- and hetero-rotaxanes through the organization of the components of mechanically interlocked molecules, introducing alternative pathways for electron donation.
NATURE COMMUNICATIONS
(2022)
Article
Biochemical Research Methods
Yoshiaki Kawano, Masahide Hikita, Naohiro Matsugaki, Masaki Yamamoto, Toshiya Senda
Summary: Native SAD phasing is a promising method for next-generation crystallography, but it requires high-quality diffraction data using long-wavelength X-rays. Optimizing crystal size and shape, as well as removing noncrystalline sample portions, can improve data quality. A crystal-processing machine using a deep-UV laser has been developed and significantly improved diffraction data quality and the success rate of experimental phasing.
ACTA CRYSTALLOGRAPHICA SECTION F-STRUCTURAL BIOLOGY COMMUNICATIONS
(2022)
Correction
Instruments & Instrumentation
Seiki Baba, Hiroaki Matsuura, Takashi Kawamura, Naoki Sakai, Yuki Nakamura, Yoshiaki Kawano, Nobuhiro Mizuno, Takashi Kumasaka, Masaki Yamamoto, Kunio Hirata
JOURNAL OF SYNCHROTRON RADIATION
(2022)
Article
Chemistry, Inorganic & Nuclear
Felix Steinke, Tobias Otto, Sho Ito, Stephan Wohlbrandt, Norbert Stock
Summary: In this study, eight new phosphonate-based metal-organic frameworks were synthesized using tetraphosphonic acid 1,1,2,2-tetra kis(4-phosphonophenypethylene (H8L) as a linker. The compounds were synthesized under hydrothermal reaction conditions using the corresponding metal nitrates as starting materials. The crystal structure was determined by electron and powder X-ray diffraction.
EUROPEAN JOURNAL OF INORGANIC CHEMISTRY
(2022)
Article
Chemistry, Inorganic & Nuclear
Thomas Mies, Christian Schurmann, Sho Ito, Andrew J. P. White, Mark R. Crimmin, Anthony G. M. Barrett
Summary: The synthesis and characterization of a calcium pyrazolonato complex was studied using X-ray diffraction (XRD) and microcrystal electron diffraction (micro-ED) methods. The XRD analysis revealed a octahedral geometry with two solvent ligands, while the micro-ED measurement showed a polymeric pentagonal bipyramidal coordination geometry joined via a central Ca2O2 ring with water ligands. The loss of water was observed during the micro-ED measurement, resulting in a secondary structure with a polymeric distorted octahedral coordination geometry. These findings highlight the complementary data provided by XRD and micro-ED methods in structural determination and the susceptibility of electrophilic metals to desolvation under micro-ED measurements.
ZEITSCHRIFT FUR ANORGANISCHE UND ALLGEMEINE CHEMIE
(2023)
Article
Chemistry, Multidisciplinary
Mohana Shivanna, Jia-Jia Zheng, Keith G. G. Ray, Sho Lto, Hirotaka Ashitani, Yoshiki Kubota, Shogo Kawaguchi, Vitalie Stavila, Ming-Shui Yao, Takao Fujikawa, Ken-ichi Otake, Susumu Kitagawa
Summary: Incorporating electron donor functionalities into porous coordination frameworks allows for strong and reversible binding of electron-withdrawing guests. A structurally dynamic 2D coordination network is found to selectively bind oxygen and nitrous oxide, while also undergoing large structural deformations. The combination of flexible frameworks and electron donor groups induces strong interactions with electron-withdrawing species, making it intriguing for sorption applications.
COMMUNICATIONS CHEMISTRY
(2023)
Article
Chemistry, Inorganic & Nuclear
Mirjam P. M. Poschmann, Oezge Alan, Sho Ito, Christian Naether, Gernot Friedrichs, Norbert Stock
Summary: Three new zirconium chelidamates, including a complex, a porous metal-containing hydrogen-bonded network (M-HOF), and a metal-organic framework (MOF), were synthesized and characterized. The structures were determined using various techniques and the sorption properties were investigated. These compounds exhibit high structural flexibility and stability in organic solvents. The M-HOF shows high selectivity towards water, while the MOF exhibits porosity towards multiple adsorbates.
INORGANIC CHEMISTRY
(2023)
Article
Biochemistry & Molecular Biology
Akinobu Senoo, Satoru Nagatoishi, Daisuke Kuroda, Sho Ito, Go Ueno, Jose M. M. Caaveiro, Kouhei Tsumoto
Summary: Small molecules that regulate protein-protein interactions have potential as valuable drugs. In this study, we propose that modulating the conformational ensemble of the proteins involved in an interaction can interfere with protein-protein interactions. We applied this concept to P-cadherin and found that a small molecule inhibitor modulates the conformational ensemble of P-cadherin, resulting in the inhibition of cell adhesion.
Article
Biochemical Research Methods
Hiroaki Matsuura, Naoki Sakai, Sachiko Toma-Fukai, Norifumi Muraki, Koki Hayama, Hironari Kamikubo, Shigetoshi Aono, Yoshiaki Kawano, Masaki Yamamoto, Kunio Hirata
Summary: In macromolecular structure determination using X-ray diffraction from multiple crystals, hierarchical clustering analysis (HCA) has been effective in classifying polymorphous data sets with a certain threshold, allowing for the identification of polymorphs in data sets with unknown structural heterogeneity. Additionally, polymorphs can also be detected in single crystals using the continuous helical scheme.
ACTA CRYSTALLOGRAPHICA SECTION D-STRUCTURAL BIOLOGY
(2023)
Article
Materials Science, Multidisciplinary
Carl Emil Andersen, James Neill McPherson, Monica Gimenez-Marques, Jingguo Li, Mariusz Kubus, Sho Ito, Christian R. R. Goeb, Sascha Ott, Rene Wugt Larsen, Guillermo Minguez Espallargas, Kasper S. S. Pedersen
Summary: Incorporating low-valent metals into metal-organic frameworks (MOFs) is a novel approach that defies the commonly accepted strategy to create robust and porous structures. The unique electronic structures of low-valent metals, which have driven the success of organometallic chemistry, can now be translated into MOFs through the use of metal carbonyls as synthons. This study demonstrates the formation of CO2 adsorbing MOFs through direct vapor-phase substitution of CO by ditopic ligands, allowing for easy coating of low-valent MOFs onto substrates.
JOURNAL OF MATERIALS CHEMISTRY C
(2023)
Article
Chemistry, Multidisciplinary
Tharushi A. Perera, William V. Taylor, M. Brenton Gildner, Eric W. Reinheimer, Sho Ito, Anna Nelson, Shane R. Yost, Todd W. Hudnall
Summary: We report unprecedented photochemistry of diamidocarbene 1, including double cyclopropanation of 1-bromonaphthalene, double addition to pyridine, and remarkable insertion into unactivated sp(3) C-H bonds of cyclohexane, tetramethylsilane, and n-pentane to yield compounds 2-6, respectively. All compounds were fully characterized, and the solid state structure of 4 was determined using single crystal electron diffraction.
Article
Chemistry, Multidisciplinary
Sho Ito, Fraser J. White, Eiji Okunishi, Yoshitaka Aoyama, Akihito Yamano, Hiroyasu Sato, Joseph D. Ferrara, Michal Jasnowski, Mathias Meyer
Summary: 3D electron diffraction/Micro electron diffraction techniques have expanded the possibilities of crystallography for determining three-dimensional molecular structures from sub-micrometer microcrystals. However, current measurements using advanced electron microscopes require expertise in both electron microscopy and crystallography, making it challenging for researchers. The newly developed electron diffractometer offers an easier path for many researchers to determine crystal structures smaller than 1 micrometer in size.