Article
Optics
Kensei Morijiri, Kento Takehana, Takatomo Mihana, Kazutaka Kanno, Makoto Naruse, Atsushi Uchida
Summary: Photonic accelerators have gained attention for use in AI applications, but the scalability of photonic decision making has not been demonstrated in experiments. A parallel photonic decision-making system using optical spatiotemporal chaos is proposed to solve large-scale multi-armed bandit problems. The experimental demonstration shows the superiority of the proposed parallel principle for correct decision making, with an exponent of 0.86. This facilitates photonic decision making for future photonic accelerators.
Review
Microbiology
Matus Dohal, Igor Porvaznik, Ivan Solovic, Juraj Mokry
Summary: Tuberculosis is a global health issue with drug-resistant strains posing a significant challenge to treatment. Treatment ineffectiveness in some individuals with pulmonary tuberculosis has hindered global tuberculosis control efforts. Approaches such as artificial intelligence, genetic screening, and whole genome sequencing can help identify the causes of treatment failure and adjust treatment accordingly.
FRONTIERS IN MICROBIOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Henry Heberle, Linlin Zhao, Sebastian Schmidt, Thomas Wolf, Julian Heinrich
Summary: Explainable artificial intelligence (XAI) methods are increasingly applicable in chemistry, and visualization techniques can highlight the influence of molecule regions on predicted properties. However, some XAI techniques face challenges in representing attribution scores for non-atom tokens in SMILES strings. The proposed XSMILES tool provides an interactive visualization technique to address this issue and support the interpretation of SMILES.
JOURNAL OF CHEMINFORMATICS
(2023)
Article
Urology & Nephrology
N. Heller, R. Tejpaul, F. Isensee, T. Benidir, M. Hofmann, P. Blake, Z. Rengal, K. Moore, N. Sathianathen, A. Kalapara, J. Rosenberg, S. Peterson, E. Walczak, A. Kutikov, R. G. Uzzo, D. A. Palacios, E. M. Remer, S. C. Campbell, N. Papanikolopoulos, Christopher J. Weight
Summary: The purpose of this study was to develop an artificial intelligence algorithm for automated scoring of R.E.N.A.L. nephrometry scores and evaluate its predictive ability for oncologic and perioperative outcomes. The results showed that AI-scores were comparable to H-scores and predicted a wide variety of meaningful patient outcomes.
JOURNAL OF UROLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Andreas M. Hoetker, Raffaele Da Mutten, Anja Tiessen, Ender Konukoglu, Olivio F. Donati
Summary: The study developed an artificial intelligence algorithm to determine the necessity of dynamic contrast-enhanced sequences in prostate MRI. The CNN showed a sensitivity of 94.4% and specificity of 68.8%, with a potential increase in sensitivity of 30.5% compared to a radiology technician. In examinations from a different vendor, the CNN achieved an AUC of 0.73.
INSIGHTS INTO IMAGING
(2021)
Article
Environmental Sciences
Victor O. K. Li, Jacqueline C. K. Lam, Jiahuan Cui
Summary: This article discusses the role and challenges of AI and big data technologies in environmental decision-making, raises a series of important questions, and summarizes the significance and innovation of the articles included in the special issue. It also highlights the important principles of AI for social good.
ENVIRONMENTAL SCIENCE & POLICY
(2021)
Article
Humanities, Multidisciplinary
Liu Yu-cheng
Summary: This article explores the connection between artificial intelligence and transhumanist posthumanism, proposing a distinction between the logic of illustration and the logic of exhaustion. The author uses ethnomethodology to understand how AI makes its accountings and accounting practices observable and accountable. The distinction represents two different ways of approaching and understanding the world. The logic of illustration creates distance between humans and their world, while the logic of exhaustion seeks to eliminate this distance and integrate humans and machines. It is important to not overlook the logic of illustration in evaluating the relationship between humans and machines.
HUMANITIES & SOCIAL SCIENCES COMMUNICATIONS
(2022)
Article
Integrative & Complementary Medicine
Musun Park, Min Hee Kim, So-Young Park, Inhwa Choi, Chang-Eop Kim
Summary: This study proposes a machine learning-based analysis tool to evaluate the clinical decision-making process of pattern identification in traditional medicine. The tool successfully identifies explicit and implicit knowledge in the process, and the analysis shows differences in the importance of explicit and implicit knowledge. This is the first study to evaluate the impact of explicit and implicit knowledge on the choice of traditional medicine doctors.
AMERICAN JOURNAL OF CHINESE MEDICINE
(2022)
Article
Automation & Control Systems
Yuming Li, Wei Zhang, Yanyan Liu, Rudong Jing, Changsong Liu
Summary: This paper proposes an object detection model based on DETR for fire and smoke detection, which simplifies the detection pipeline and builds an end-to-end detector. By adding a normalization-based attention module in the feature extraction stage and using multiscale deformable attention in the encoder-decoder structure, the model achieves improved detection performance while reducing complexity.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Cardiac & Cardiovascular Systems
Takeshi Nishi, Rikiya Yamashita, Shinji Imura, Kazuya Tateishi, Hideki Kitahara, Yoshio Kobayashi, Paul G. Yock, Peter J. Fitzgerald, Yasuhiro Honda
Summary: This study developed a deep learning method for automatic segmentation of IVUS images, including stent area along with lumen and vessel area. The DL-based segmentation showed good correlation and agreement with manual segmentation by experts, indicating the feasibility of AI-assisted IVUS assessment in patients undergoing coronary stent implantation.
INTERNATIONAL JOURNAL OF CARDIOLOGY
(2021)
Article
Cardiac & Cardiovascular Systems
Karima Benmohammed, Paul Valensi, Nabil Omri, Zeina Al Masry, Noureddine Zerhouni
Summary: This study validates the effectiveness of artificial intelligence-based scores (AI_METS) in screening metabolic syndrome (MetS) in adolescents without using specific percentiles. The results show that AI_METS has high accuracy and sensitivity in detecting MetS.
NUTRITION METABOLISM AND CARDIOVASCULAR DISEASES
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Jie Xu, Jia Liu, Ning Guo, Linli Chen, Weixiang Song, Dajing Guo, Yu Zhang, Zheng Fang
Summary: The study showed that the risk category performance of artificial intelligence-based coronary artery calcium score software in non-gated chest CT was consistent with the manual method, and the software exhibited good performance on CT machines from different manufacturers.
EUROPEAN JOURNAL OF RADIOLOGY
(2021)
Article
Psychology, Multidisciplinary
Alessandra Talamo, Silvia Marocco, Chiara Tricol
Summary: The application of artificial intelligence in the financial field is creating a new area of study called financial intelligence, aimed at assisting in complex decision-making processes. This is particularly crucial for venture capitalist organizations where different actors with varying decision-making behaviors are involved. This study proposes a modeling approach for financial AI-based services and suggests the integration of human/AI systems for better decision-making support.
FRONTIERS IN PSYCHOLOGY
(2021)
Article
Computer Science, Information Systems
P. Madhumathy, Digvijay Pandey
Summary: Biomedical image processing is a technique that uses artificial intelligence to extract complex data from internal human organs and evaluate them. This paper aims to use deep learning techniques to generate and classify artefacts in photo acoustic image datasets, and implement an effective artefact elimination strategy. These technologies have the potential to improve understanding of the human body and advance diagnostic and therapeutic procedures.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Editorial Material
Health Care Sciences & Services
Mirja Mittermaier, Marium M. Raza, Joseph C. Kvedar
Summary: Artificial intelligence is increasingly used in healthcare, particularly in surgery. While it holds promise in predicting outcomes and guiding surgeons, AI systems can also be biased, exacerbating existing inequalities. This impacts disadvantaged populations, who may receive less accurate algorithmic predictions or underestimate their need for care. Detecting and mitigating bias is crucial for creating fair and generalizable AI technology. This article discusses a recent study that developed a new strategy to address bias in surgical AI systems.
NPJ DIGITAL MEDICINE
(2023)