Article
Forestry
Chongyuan Cai, Hao Xu, Sheng Chen, Laibang Yang, Yuhui Weng, Siqi Huang, Chen Dong, Xiongwei Lou
Summary: Tree crown width is directly related to wood quality and tree growth. The traditional method of measuring crown width is labor-intensive and time-consuming. This study proposes a novel Faster-RCNN algorithm for tree crown identification and crown width extraction in a high-density loblolly pine forest environment.
Article
Computer Science, Artificial Intelligence
Liye Guo
Summary: The advancement of information technology is crucial to the reform of teaching, and the requirements for interaction and convenience in art teaching are getting higher. This paper presents an automatic classification method of art teaching works based on improved deep learning model resnet101 and multimodal information fusion. The proposed method effectively facilitates the construction of art teaching platform and makes contributions to art interactive teaching.
Article
Ophthalmology
Mo Tiwari, Chris Piech, Medina Baitemirova, Namperumalsamy Prajna, Muthiah Srinivasan, Prajna Lalitha, Natacha Villegas, Niranjan Balachandar, Janice T. Chua, Travis Redd, Thomas M. Lietman, Sebastian Thrun, Charles C. Lin
Summary: This study developed and evaluated an automated, portable algorithm that can differentiate active corneal ulcers from healed scars using only external photographs. The algorithm utilized a convolutional neural network trained on a large dataset of corneal ulcer and scar photographs. The results showed that the algorithm achieved high accuracy in classifying corneal ulcers and scars in diverse patient populations. It demonstrated potential as a cost-effective diagnostic approach that can aid triage in communities with limited access to eye care.
Article
Medicine, Research & Experimental
Wuteng Cao, Huabin Hu, Jirui Guo, Qiyuan Qin, Yanbang Lian, Jiao Li, Qianyu Wu, Junhong Chen, Xinhua Wang, Yanhong Deng
Summary: This study aimed to develop and validate a deep learning model based on pre-treatment CT images for predicting DNA mismatch repair (MMR) status in patients with colorectal cancer (CRC). The results showed that the deep learning model had good predictive ability and could serve as a noninvasive tool for predicting MMR status in patients with CRC.
JOURNAL OF TRANSLATIONAL MEDICINE
(2023)
Article
Computer Science, Hardware & Architecture
Saif Ur Rehman Khan, Ming Zhao, Sohaib Asif, Xuehan Chen, Yusen Zhu
Summary: This article introduces a novel global-local convolution technique based on pre-trained ResNet101 for image classification in computer vision. The proposed model utilizes global and local feature compositions to extract features from histopathological slides and achieves high accuracy in multi-label cancer category prediction.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Neurosciences
Tingting Wang, Meng Wang, Weifang Zhu, Lianyu Wang, Zhongyue Chen, Yuanyuan Peng, Fei Shi, Yi Zhou, Chenpu Yao, Xinjian Chen
Summary: In this paper, a novel semi-supervised multi-scale self-transformer generative adversarial network (Semi-MsST-GAN) is proposed for corneal ulcer segmentation in fluorescein staining of slit-lamp images. By introducing a multi-scale self-transformer network and a semi-supervised approach, the performance of corneal ulcer segmentation is improved.
FRONTIERS IN NEUROSCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Dilbag Singh, Vijay Kumar, Manjit Kaur, Rajani Kumari
Summary: In this paper, a method using deep transfer learning and deep forest model to diagnose COVID-19 infection is proposed. The ResNet101 model is used to extract features from chest X-ray images, and the deep forest model is employed to predict COVID-19 infected patients. Experimental results show that the proposed model effectively diagnoses COVID-19 infection with an accuracy of 99.4%.
JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Mujahid Hussain, Makhmoor Fiza, Aiman Khalil, Asad Ali Siyal, Fayaz Ali Dharejo, Waheeduddin Hyder, Antonella Guzzo, Moez Krichen, Giancarlo Fortino
Summary: Skin cancer, especially melanoma, is a serious problem due to its increasing incidence. Early attention and timely detection are crucial for effective treatment and patient survival. In this study, a new dataset called Nailmelonma is introduced to train and evaluate deep learning models for nail melanoma detection.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Hiam Alquran, Wan Azani Mustafa, Isam Abu Qasmieh, Yasmeen Mohd Yacob, Mohammed Alsalatie, Yazan Al-Issa, Ali Mohammad Alqudah
Summary: This paper presents a computer-aided diagnostic system for classifying cervical smear images with high accuracy, achieving early detection and classification of cervical cancer with seven classes. By extracting features and utilizing SVM classifier, the system successfully distinguishes abnormality levels with high accuracy.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Multidisciplinary Sciences
Zhongwen Li, Jiewei Jiang, Kuan Chen, Qianqian Chen, Qinxiang Zheng, Xiaotian Liu, Hongfei Weng, Shanjun Wu, Wei Chen
Summary: Keratitis is the main cause of corneal blindness worldwide, early detection and treatment are key to avoiding vision loss. A deep learning system for automated classification of keratitis and other cornea abnormalities has been developed to help address the shortage of ophthalmologists in resource-limited settings.
NATURE COMMUNICATIONS
(2021)
Article
Neurosciences
Linquan Lv, Mengle Peng, Xuefeng Wang, Yuanjun Wu
Summary: Corneal ulcer, a common symptom of corneal disease, can lead to corneal blindness. This study proposes a deep learning method based on multi-scale information fusion and label smoothing strategy, achieving accurate classification of corneal ulcer and assisting ophthalmologists in diagnosis.
FRONTIERS IN NEUROSCIENCE
(2022)
Article
Chemistry, Analytical
Lindsay Stern, Atena Roshan Fekr
Summary: This study proposes 2D and 3D Convolutional Neural Networks to detect three main body positions in bed. The best-performing models achieved accuracies of 98.90% and 97.80% for 5-fold and leave-one-subject-out cross-validation, respectively. These models provide promising results for in-bed posture recognition and can be used to further distinguish postures into more detailed subclasses.
Article
Computer Science, Hardware & Architecture
Xiaozhang Liu, Lang Li, Xueyang Wang, Li Hu
Summary: This paper studies the nature of attacks from adversarial samples from the perspective of the main and minor features, finding that deep learning models mainly learn the main features and proposing a method to generate adversarial samples in the sample subspace.
COMPUTER STANDARDS & INTERFACES
(2022)
Article
Multidisciplinary Sciences
Bernard Tiddeman, Morteza Ghahremani
Summary: The paper introduces a novel learning-based wavelet transform method, which combines 1x1 convolution filters learnt from PCA with invertible wavelet filter-bank to create a separable CNN-like architecture, avoiding overfitting issues. By applying the network to linear inverse problems using ADMM, it achieves promising results in compressive sensing, in-painting, denoising, and super-resolution tasks, closing the performance gap with GAN-based methods.
Review
Medicine, General & Internal
Hsu-Heng Yen, Ping-Yu Wu, Mei-Fen Chen, Wen-Chen Lin, Cheng-Lun Tsai, Kang-Ping Lin
Summary: With a decreasing incidence of peptic ulcer bleeding, the experience of managing patients with PUB among young endoscopists has declined accordingly. Artificial Intelligence (AI) has shown great potential in the field of gastroenterology, particularly in enhancing human performance. The introduction of AI technologies may soon impact endoscopists' clinical practice by improving the quality of care for patients with PUB.
JOURNAL OF CLINICAL MEDICINE
(2021)
Article
Medicine, General & Internal
Yasmin Mohd Yacob, Hiam Alquran, Wan Azani Mustafa, Mohammed Alsalatie, Harsa Amylia Mat Sakim, Muhamad Safiih Lola
Summary: Atrophic gastritis (AG) is a chronic condition caused by H. pylori infection and can lead to gastric cancer if left untreated. Early detection of AG is crucial to prevent such cases.
Article
Medicine, General & Internal
Ala'a Zyout, Hiam Alquran, Wan Azani Mustafa, Ali Mohammad Alqudah
Summary: This study used two different spectrum representations, iris-spectrogram and scalogram, to extract deep features and classify different ECG beat waves using two deep convolutional neural networks (CNN), ResNet101 and ShuffleNet. The results showed that using ResNet101 and scalogram of T-wave achieved the highest accuracy of 98.3% for beat rhythm detection, while using iris-spectrogram and ResNet101 for QRS-wave achieved an accuracy of 94.4%. In conclusion, deep features from time-frequency representation of ECG beat waves can accurately detect basic rhythms such as normal, tachycardia, and bradycardia.
Article
Biotechnology & Applied Microbiology
Shefa Tawalbeh, Hiam Alquran, Mohammed Alsalatie
Summary: This paper proposes six feature fusion techniques to improve the classification accuracy of cervical cancer. After comparing ten feature datasets, it is found that canonical correlation analysis technique achieves the best performance with 99.7% accuracy. Independent component analysis and least absolute shrinkage and selection operator techniques are also effective with 98.3% accuracy.
BIOENGINEERING-BASEL
(2023)
Article
Multidisciplinary Sciences
Wan Imanul Aisyah Wan Mohamad Nawi, Abdul Aziz K. Abdul Hamid, Muhamad Safiih Lola, Syerrina Zakaria, Elayaraja Aruchunan, R. U. Gobithaasan, Nurul Hila Zainuddin, Wan Azani Mustafa, Mohd Lazim Abdullah, Nor Aieni Mokhtar, Mohd Tajuddin Abdullah
Summary: To improve the accuracy and efficiency of COVID-19 forecasting, a hybrid ARIMA-SVM model is proposed and empirically shown to outperform ARIMA and SVM models in reducing error percentages.
Editorial Material
Medicine, General & Internal
Wan Azani Mustafa, Hiam Alquran
Article
Medicine, General & Internal
Abdul Aziz K. Abdul Hamid, Wan Imanul Aisyah Wan Mohamad Nawi, Muhamad Safiih Lola, Wan Azani Mustafa, Siti Madhihah Abdul Malik, Syerrina Zakaria, Elayaraja Aruchunan, Nurul Hila Zainuddin, R. U. Gobithaasan, Mohd Tajuddin Abdullah
Summary: Improving the accuracy and efficiency of time-series forecasts is crucial for authorities to predict and prevent the spread of the Coronavirus disease. The dataset contains both linear and non-linear patterns, which makes it inefficient to use linear models for prediction. A hybrid approach is proposed to achieve a more accurate and efficient predictive value of COVID-19.
Review
Medicine, General & Internal
Marina Yusoff, Toto Haryanto, Heru Suhartanto, Wan Azani Mustafa, Jasni Mohamad Zain, Kusmardi Kusmardi
Summary: Breast cancer diagnosis relies on histopathological imaging, which is time-consuming due to image complexity and volume. Deep learning has become popular for diagnosing cancerous images but achieving high precision and minimizing overfitting remains challenging for classification solutions.
Article
Computer Science, Interdisciplinary Applications
Shefa M. Tawalbeh, Ahmed Al-Omari, Lina M. K. Al-Ebbini, Hiam Alquran
Summary: Jordanian healthcare institutes have launched programs to establish health information systems since 2009. Medical doctors may face barriers in using HIS resources due to lack of knowledge and training. To address this issue, a survey was conducted to evaluate the need for HIS training among specialized medical doctors in Jordan, as well as their current usage of HIS resources and areas of required training.
Article
Medicine, General & Internal
Mohammed Alsalatie, Hiam Alquran, Wan Azani Mustafa, Ala'a Zyout, Ali Mohammad Alqudah, Reham Kaifi, Suhair Qudsieh
Summary: This study focuses on using Pap smear images to detect cervical cancer and compares different scenarios and features for classification. The paper explores the use of convolutional neural networks and optimization algorithms to create an efficient computer-aided diagnosis system. By relying on tissue analysis rather than just the nucleus, this method improves accuracy in diagnosing precancerous and early-stage cervical cancer.
Review
Medicine, General & Internal
Wan Azani Mustafa, Shahrina Ismail, Fahirah Syaliza Mokhtar, Hiam Alquran, Yazan Al-Issa
Summary: Cervical cancer is a significant global health problem with high mortality and incidence rates. Advances in cervical cancer detection techniques, particularly machine learning-based CAD systems, have shown promise in improving accuracy and sensitivity.
Article
Biotechnology & Applied Microbiology
Ateka Khader, Hiam Alquran
Summary: This study aims to automatically classify histopathological images of cartilage specimens using artificial intelligence algorithms in order to understand the progression of osteoarthritis (OA). Histopathology scoring systems were used to evaluate OA progress and mechanisms. The results show that by using convolutional neural networks and feature extraction algorithms, early-stage OA can be diagnosed instantly with high accuracy and F1 scores.
BIOENGINEERING-BASEL
(2023)
Article
Computer Science, Information Systems
Mohd Aminudin Jamlos, Nur Amirah Othman, Wan Azani Mustafa, Mohd Faizal Jamlos, Mohamad Nur Khairul Hafizi Rohani
Summary: This paper discusses the use of metamaterial elements combined with ultra-wideband (UWB) antenna for microwave imaging systems to detect tumors. Through detailed designs and performance analysis, it is shown that the 7 x 4 metamaterial element shape has higher gain and is a better choice compared to the 10 x 5 metamaterial element for realizing metamaterial superstrate antennas.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Information Systems
Mohd Annuar Isa, Mohamad Nur Khairul Hafizi Rohani, Baharuddin Ismail, Mohamad Kamarol Jamil, Muzamir Isa, Afifah Shuhada Rosmi, Mohd Aminudin Jamlos, Wan Azani Mustafa, Nurulbariah Idris, Abdullahi Abubakar Mas'ud
Summary: Electrical trees are an aging mechanism in crosslinked polyethylene insulation of high-voltage cables, and the accurate segmentation of tree structures in 2D images is crucial for developing new insulation materials. This study proposes a new method based on the multi-scale line tracking algorithm for segmenting electrical tree images, achieving better performance compared to established techniques.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Biotechnology & Applied Microbiology
Noor Ilanie Nordin, Wan Azani Mustafa, Muhamad Safiih Lola, Elissa Nadia Madi, Anton Abdulbasah Kamil, Marah Doly Nasution, Abdul Aziz K. Abdul Hamid, Nurul Hila Zainuddin, Elayaraja Aruchunan, Mohd Tajuddin Abdullah
Summary: This study proposes a novel hybrid model based on SVM and LR for predicting small events per variable. The hybrid model outperforms SVM and LR in terms of accuracy, mean squared error, and root mean squared error. This hybrid model is particularly important for medical institutions and practitioners in the face of future pandemics.
BIOENGINEERING-BASEL
(2023)