Article
Chemistry, Multidisciplinary
Luca Brunese, Francesco Mercaldo, Alfonso Reginelli, Antonella Santone
Summary: This paper proposes an automatic method to analyze respiratory sounds and demonstrates the effectiveness of machine learning techniques in detecting and characterizing lung diseases. The experimental analysis shows promising results, with the neural network model achieving the best performance.
APPLIED SCIENCES-BASEL
(2022)
Review
Health Care Sciences & Services
Peng Xue, Jiaxu Wang, Dongxu Qin, Huijiao Yan, Yimin Qu, Samuel Seery, Yu Jiang, Youlin Qiao
Summary: This study conducted a meta-analysis to evaluate the diagnostic performance of deep learning algorithms for early breast and cervical cancer identification. The results showed that these algorithms performed acceptably well across all subgroups, comparable to human clinicians. However, the relatively poor design and reporting of the included studies may have caused bias in the results.
NPJ DIGITAL MEDICINE
(2022)
Article
Chemistry, Multidisciplinary
Permatasari Silitonga, Alhadi Bustamam, Hengki Muradi, Wibowo Mangunwardoyo, Beti E. Dewi
Summary: The study developed models using Artificial Neural Network (ANN) and Discriminant Analysis (DA) to predict the severity level of dengue based on laboratory test results, achieving high accuracy of 90.91%, sensitivity of 91.11%, and specificity of 95.51%. The proposed model can assist physicians in timely predicting and treating dengue patients to prevent fatal cases.
APPLIED SCIENCES-BASEL
(2021)
Article
Chemistry, Analytical
Yuanming Lu, Di Wang, Die Liu, Xianyi Yang
Summary: This paper introduces a damage detection method based on vibration signals, using stochastic configuration networks and stochastic convolutional feature extraction approach. Experimental results show significant improvements in accuracy and efficiency compared to traditional methods.
Article
Computer Science, Artificial Intelligence
Fath U. Min Ullah, Mohammad S. Obaidat, Khan Muhammad, Amin Ullah, Sung Wook Baik, Fabio Cuzzolin, Joel J. P. C. Rodrigues, Victor Hugo C. Albuquerque
Summary: The paper proposes a computationally intelligent violence detection approach, which precisely detects violent scenes through deep analysis of surveillance video sequential patterns. The method utilizes convolutional neural networks, optical flow feature extraction, and long short-term memory networks to learn violence patterns, achieving a 2% increase in accuracy over surveillance fight data set.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Chemistry, Multidisciplinary
Syed Ijaz Ul Haq, Muhammad Naveed Tahir, Yubin Lan
Summary: In this study, device visualization and deep learning were used to detect weeds in real time in the wheat crop system. Two different frameworks, TensorFlow and PyTorch, were used to apply deep learning algorithms, with PyTorch performing better in terms of speed and accuracy. The weed detection models achieved accuracies of 0.89 and 0.91 with inference times of 9.43 ms and 12.38 ms on the NVIDIA RTX2070 GPU for each picture (640 x 640).
APPLIED SCIENCES-BASEL
(2023)
Article
Multidisciplinary Sciences
Tiankuang Zhou, Wei Wu, Jinzhi Zhang, Shaoliang Yu, Lu Fang
Summary: We propose a spatiotemporal photonic computing architecture to achieve dynamic processing, matching highly parallel spatial computing with high-speed temporal computing. A unified training framework is devised to optimize the physical system and the network model. The proposed architecture paves the way for ultrafast advanced machine vision and will find applications in unmanned systems, autonomous driving, ultrafast science, etc.
Article
Engineering, Marine
Jizhong Wu, Bo Liu, Hao Zhang, Shumei He, Qianqian Yang
Summary: Researchers developed a FCN method for fault detection in continental sandstone reservoirs, trained using synthetic seismic data to achieve accuracy, and optimized the model parameters using a balanced crossentropy loss function. The method was validated on real field data, demonstrating its accuracy and efficiency in fault prediction compared to conventional methods.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2021)
Article
Agronomy
Muhammad Hammad Saleem, Johan Potgieter, Khalid Mahmood Arif
Summary: This research aims to achieve accurate detection of various classes of weeds and a negative class using deep learning, and thoroughly analyze the architectural details of the Faster RCNN model. The results show improved classification and localization performance, validating the effectiveness and robustness of the approach.
Article
Computer Science, Artificial Intelligence
Mohammad Kazim Hooshmand, Doreswamy Hosahalli
Summary: This study proposes a model using one-dimensional CNN architecture for network anomaly detection. The model achieves good performance by dividing network traffic data into protocol categories and independently processing each category. Before training, feature selection and oversampling techniques are applied to handle class imbalance. Experimental results demonstrate the effectiveness of the approach in detecting anomalies of different protocol categories.
CAAI TRANSACTIONS ON INTELLIGENCE TECHNOLOGY
(2022)
Article
Computer Science, Hardware & Architecture
Kutub Thakur, Hamed Alqahtani, Gulshan Kumar
Summary: The intelligent system IDGADS is capable of quickly detecting algorithmically generated domains with 99% accuracy based on easy-to-compute features of real domain name system (DNS) traffic. It can serve as the first line of defense in a security stack for validating DNS queries.
COMPUTERS & ELECTRICAL ENGINEERING
(2021)
Article
Mathematics
Hanting Wei, Bo Yu, Wei Wang, Chenghong Zhang
Summary: This paper proposes a risk factor detection model based on deep learning in low illumination scenarios and tests the optimization of low illumination image enhancement problems. The model has high detection accuracy and can overcome the impact of low lighting.
Article
Chemistry, Multidisciplinary
Raidan Ba-Hattab, Noha Barhom, Safa A. Azim Osman, Iheb Naceur, Aseel Odeh, Arisha Asad, Shahd Ali R. N. Al-Najdi, Ehsan Ameri, Ammar Daer, Renan L. B. Da Silva, Claudio Costa, Arthur R. G. Cortes, Faleh Tamimi
Summary: This study aimed to develop an artificial intelligence technology to detect periapical lesions that dentists could fail to notice on panoramic radiographs. By annotating and classifying 18618 periapical root areas on 713 panoramic radiographs, a two-stage convolutional neural network model was trained. The model, consisting of a detector and a classifier, successfully localized and classified periapical lesions with an accuracy of 84.6%, sensitivity of 72.2%, and specificity of 85.6%.
APPLIED SCIENCES-BASEL
(2023)
Review
Computer Science, Artificial Intelligence
Jia-Chi Wang, Yi-Chung Shu, Che-Yu Lin, Wei-Ting Wu, Lan-Rong Chen, Yu-Cheng Lo, Hsiao-Chi Chiu, Levent Ozcakar, Ke-Vin Chang
Summary: This study aimed to explore and summarize the performance of deep learning algorithms in the automatic sonographic assessment of the median nerve at the carpal tunnel level. The results showed that the deep learning algorithm enables automated localization and segmentation of the median nerve in ultrasound imaging with acceptable accuracy and precision. Future research should validate the performance of deep learning algorithms in detecting and segmenting the median nerve along its entire length and across datasets obtained from various ultrasound manufacturers.
ARTIFICIAL INTELLIGENCE IN MEDICINE
(2023)
Article
Automation & Control Systems
Hongfeng Tao, Jier Qiu, Yiyang Chen, Vladimir Stojanovic, Long Cheng
Summary: This paper proposes an unsupervised cross-domain fault diagnosis method based on time-frequency information fusion to address the challenges of label scarcity and data distribution differences in bearing fault diagnosis. The method utilizes wavelet packet decomposition and reconstruction to extract fault features in the form of a 2-D time-frequency map, constructs an unsupervised cross-domain fault diagnosis model, and calculates the joint distribution distance using the improved maximum mean discrepancy algorithm and pseudo-labels. Experimental results on motor bearings demonstrate the high diagnosis accuracy and strong robustness of the proposed method.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2023)
Article
Robotics
Pieter M. Blok, Frits K. van Evert, Antonius P. M. Tielen, Eldert J. van Henten, Gert Kootstra
Summary: The study aims to develop a robot capable of detecting broccoli heads using a deep learning algorithm to reduce labor costs. Data augmentation was found to improve the algorithm's generalization performance, with geometric transformations proving more effective than photometric transformations. The algorithm achieved successful generalization after including 5% of images from a specific cultivar in the training dataset.
JOURNAL OF FIELD ROBOTICS
(2021)
Article
Computer Science, Interdisciplinary Applications
Simon van Mourik, Rik van der Tol, Raphael Linker, Daniel Reyes-Lastiri, Gert Kootstra, Peter Groot Koerkamp, Eldert J. van Henten
Summary: The challenge in modern agriculture is to find a sustainable way to achieve sufficient production by precisely controlling input resources. Utilizing technology and data to develop model-based management support and automation can improve agricultural productivity.
ENVIRONMENTAL MODELLING & SOFTWARE
(2021)
Article
Agricultural Engineering
Wouter J. P. Kuijpers, Duarte J. Antunes, Silke Hemming, Eldert J. van Henten, Marinus J. G. van de Molengraft
Summary: This paper introduces a RHOC method with an economic objective function to balance resource cost and income through yield, considering yield and product price as the two main factors. Simulations show that the new approach for predicting future income through yield is accurate and unaffected by model assumptions.
BIOSYSTEMS ENGINEERING
(2021)
Article
Agricultural Engineering
Pieter M. Blok, Eldert J. van Henten, Frits K. van Evert, Gert Kootstra
Summary: This study aimed to improve the estimation accuracy of broccoli head size by using deep learning algorithms to deal with occlusions. The ORCNN method outperformed the Mask R-CNN method in sizing occluded broccoli heads, showing better sizing performance.
BIOSYSTEMS ENGINEERING
(2021)
Article
Agricultural Engineering
Wouter J. P. J. Kuijpers, Duarte Antunes, Simon van Mourik, Eldert J. van Henten, Marinus J. G. van de Molengraft
Summary: This research investigates the impact of weather forecast errors on the performance of controlled greenhouse systems. Findings show that while forecast errors do have some effect on system performance, the impact is not significant. The study suggests that by using an optimal control algorithm and similar forecast errors, the influence of weather forecast errors on greenhouse system performance can be mitigated.
BIOSYSTEMS ENGINEERING
(2022)
Article
Agricultural Engineering
Frans P. Boogaard, Eldert J. van Henten, Gert Kootstra
Summary: This paper focuses on measuring plant architecture of cucumber plants using 3D point clouds, spectral data, and deep learning. Results show that spectral data can improve segmentation accuracy, and the effect of uncertainty in ground truth data collection was analyzed.
BIOSYSTEMS ENGINEERING
(2021)
Review
Agriculture, Multidisciplinary
David Katzin, Eldert J. van Henten, Simon van Mourik
Summary: This study provides an overview of the current state of greenhouse modelling, identifying the key processes and common approaches used in process-based greenhouse models. The analysis reveals the variation and overlap in model design and complexity. The study suggests that increased transparency, code availability, shared datasets, and evaluation benchmarks will enhance model reuse, extension, evaluation, and comparison.
AGRICULTURAL SYSTEMS
(2022)
Article
Agriculture, Multidisciplinary
Pieter M. Blok, Gert Kootstra, Hakim Elchaoui Elghor, Boubacar Diallo, Frits K. van Evert, Eldert J. van Henten
Summary: The study aimed to train a CNN with fewer annotated images while maintaining its performance. An active learning method called MaskAL was developed to automatically select hard-to-classify images for annotation and retraining. The results showed that MaskAL outperformed random sampling on a broccoli dataset with visually similar classes.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Plant Sciences
Frans P. Boogaard, Eldert J. van Henten, Gert Kootstra
Summary: This paper presents a method for obtaining phenotypic datasets of traits related to plant architecture using 3D point cloud data. The authors address the issue of class imbalance in point cloud segmentation and propose a class-dependent sampling strategy to improve segmentation performance. The experiments show that using a class-dependent training set can increase segmentation quality, and different classes have different optimal neighbourhood sizes.
FRONTIERS IN PLANT SCIENCE
(2022)
Article
Chemistry, Analytical
Khan Wali, Haris Ahmad Khan, Mark Farrell, Eldert J. Van Henten, Erik Meers
Summary: This study investigates the potential of using near-infrared (NIR) and mid-infrared (MIR) techniques to characterize different properties of bio-based fertilizers. The results show that combining the model outcomes of NIR and MIR can improve the prediction performance, especially for heavy metals and elements.
Article
Agricultural Engineering
Henry J. Payne, Silke Hemming, Bram A. P. van Rens, Eldert J. van Henten, Simon van Mourik
Summary: The high energy demand in the Dutch greenhouse horticultural sector can be exacerbated by weather forecast errors. This study investigates the impact of weather forecast errors on energy prediction and trading uncertainty in greenhouse horticulture. The findings show that errors in temperature, wind speed, and radiation forecast contribute to overestimation of energy demand in greenhouses.
BIOSYSTEMS ENGINEERING
(2022)
Article
Automation & Control Systems
Eldert J. van Henten, Amy Tabb, John Billingsley, Marija Popovic, Mingcong Deng, John Reid
Summary: Agricultural production is crucial for providing resources to a growing population, but it also depletes various resources and has a significant impact on the environment. Despite population growth, participation in agriculture is decreasing. Technology is essential for the development of agriculture and will play a crucial role in sustainable agrifood production in the future.
IEEE ROBOTICS & AUTOMATION MAGAZINE
(2022)
Article
Agricultural Engineering
David Katzin, Leo F. M. Marcelis, Eldert J. van Henten, Simon van Mourik
Summary: This study presents a novel concept for greenhouses, where both lighting and heating are derived exclusively from lamps. Such greenhouses can be highly efficient, as light is used both for crop growth and for heating. By using model simulations, it was found that these greenhouses could achieve higher yields and energy consumption compared to traditional greenhouses.
BIOSYSTEMS ENGINEERING
(2023)
Article
Agricultural Engineering
David Rapado-Rincon, Eldert J. van Henten, Gert Kootstra
Summary: The ability to accurately represent and localise relevant objects is essential for robots to carry out effective tasks. This paper introduces a novel approach for building generic representations in occluded agro-food environments using multi-view perception and 3D multi-object tracking.
BIOSYSTEMS ENGINEERING
(2023)
Article
Agricultural Engineering
David Rapado-Rincon, Eldert J. van Henten, Gert Kootstra
Summary: This paper presents a novel method, MinkSORT, to improve the accuracy of world models in agro-food environments. By using a 3D sparse convolutional network to generate tracking features, robots can perform tasks in the agro-food industry with greater efficiency and accuracy. The proposed method was evaluated using real-world data collected in a tomato greenhouse, and it significantly improved the performance of the baseline model.
BIOSYSTEMS ENGINEERING
(2023)