Article
Multidisciplinary Sciences
Marwa Helmy, Eman Eldaydamony, Nagham Mekky, Mohammed Elmogy, Hassan Soliman
Summary: This study proposes a prediction system to identify protein and lncRNA genes related to Parkinson's disease (PD), aiding in early diagnosis. The system achieved promising results in terms of performance compared to other systems.
SCIENTIFIC REPORTS
(2022)
Article
Mathematical & Computational Biology
Chengkang Li, Ran Wei, Yishen Mao, Yi Guo, Ji Li, Yuanyuan Wang
Summary: In this study, a CAD system based on radiomics and clinical indices was proposed to differentiate benign from malignant IPMN and MCN. By utilizing a novel feature selection algorithm and Support Vector Machine model, the CAD system achieved significant diagnostic performance with AUC of 0.83 and 0.92 in the cross-validation and independent testing cohorts, respectively.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Syed Fawad Hussain, Saeed Mian Qaisar
Summary: Epilepsy, characterized by seizures, requires constant monitoring, and EEG signals are commonly used for diagnosis. A new framework for EEG-based epilepsy detection has been proposed to reduce power consumption and improve accuracy in multiclass classification. This framework involves data preprocessing and a novel classification paradigm, achieving high accuracy in testing on different datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ming Ni, Lili Wang, Haiyang Yu, Xiaoyi Wen, Yinghua Yang, Guangzhen Liu, Yabin Hu, Zhiming Li
Summary: The LASSO-SVM model is effective in predicting liver fibrosis in a rodent model using nonenhanced T1-weighted imaging, showing high diagnostic performance for different stages of liver fibrosis. The study results demonstrate that this model exhibits high accuracy and AUC in liver fibrosis prediction.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2021)
Article
Engineering, Biomedical
Deepa Kumari, Pavan Kumar Reddy Yannam, Isha Nilesh Gohel, Mutyala Venkata Sai, Subhash Naidu, Yash Arora, B. S. A. S. Rajita, Subhrakanta Panda, Jabez Christopher
Summary: This paper proposes a novel hybrid feature extraction and selection method to classify mammograms into benign and malignant images. It compares combinations of existing feature extraction methods and selects the most relevant features using a hybrid feature selection approach. The performance of the classifiers is improved through hyperparameter tuning and pipeline optimization techniques. Experimental results show that the proposed framework achieves high accuracy, specificity, sensitivity, and F1-score on artificial neural networks.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Artificial Intelligence
Salah Eddine Bekhouche, Yassine Ruichek, Fadi Dornaika
Summary: Monitoring driver's drowsiness is crucial for road safety. This paper presents a computer vision-based framework for driver drowsiness detection, which detects the driver's face, extracts deep features, applies temporal feature aggregation and feature selection, and uses a binary classifier to determine drowsiness.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Engineering, Multidisciplinary
Asmaa M. Khalid, Wael Said, Mahmoud Elmezain, Khalid M. Hosny
Summary: This paper proposes a novel feature selection algorithm for high-dimensional datasets, using the Object-Oriented Programming Optimization Algorithm (OOPOA) to solve the FS problem. Experimental results show that the proposed algorithm outperforms existing methods in terms of classification accuracy and computational performance.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Engineering, Biomedical
Dezhong Bi, Dongxia Zhu, Fatima Rashid Sheykhahmad, Mingqi Qiao
Summary: This study develops a computer-aided diagnosis system for accurate diagnosis of skin cancer, which outperforms traditional methods and other new algorithms. By optimizing feature selection and using support vector machine classifier, the proposed method achieves higher detection rates and lower false acceptance and rejection rates.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Computer Science, Interdisciplinary Applications
Jingkun Wang, Haotian Sun, Ke Jiang, Weiwei Cao, Shuangqing Chen, Jianbing Zhu, Xiaodong Yang, Jian Zheng
Summary: This study proposes a novel method for MC detection in digital breast tomosynthesis (DBT). The method improves the feature sharing and extraction in CNN models to achieve accurate and rapid detection of small and low-contrast MCs. Experimental results on a clinical dataset demonstrate impressive performance of the method in MC detection, providing valuable diagnostic suggestions for early breast cancer screening.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2023)
Article
Multidisciplinary Sciences
Abeer Elkhouly, Allan Melvin Andrew, Hasliza A. Rahim, Nidhal Abdulaziz, Mohd Fareq Abd Malek, Shafiquzzaman Siddique
Summary: In this study, a machine learning solution based on unsupervised spectral clustering is introduced to classify audiograms according to their shapes. The proposed ML algorithm outperforms existing models with higher accuracy, precision, recall, specificity, and F-score values. This work presents a novel ML technique that can potentially change the existing practices in classifying audiograms.
SCIENTIFIC REPORTS
(2023)
Review
Computer Science, Artificial Intelligence
Divya Srivastava, B. Rajitha, Suneeta Agarwal
Summary: Content-based image retrieval faces the challenge of finding dissimilarity among similar objects, especially when the objects are highly similar. The proposed image retrieval approach achieves a high accuracy within the same category of objects.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Carlos Eiras-Franco, Bertha Guijarro-Berdinas, Amparo Alonso-Betanzos, Antonio Bahamonde
Summary: The ReliefF-LSH algorithm simplifies the costliest step of the ReliefF algorithm by approximating the nearest neighbor graph using locality-sensitive hashing. It can process large data sets and obtains better results and is more generally applicable than the original ReliefF.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2021)
Article
Computer Science, Information Systems
Pramoda Patro, Krishna Kumar, G. Suresh Kumar, Gandharba Swain
Summary: Function approximation is essential in various fields, and involves finding the suitable relationship between variables and responses from a dataset. This study proposes improved neural network methods, feature selection to handle noise in datasets, and classification using fuzzy rules. Performance evaluation is based on metrics such as precision, recall, accuracy, and error rate.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Hakan Ezgi Kiziloz
Summary: This study formally compares different classifier ensemble methods in the feature selection domain and finds that ensemble methods outperform single classifiers, albeit with longer execution time, and are more effective in minimizing the number of features.
Article
Biology
Akshata K. Naik, Venkatanareshbabu Kuppili
Summary: Gene selection is crucial for classifying high-dimensional microarray gene expression data. This paper proposes a neural network-based embedded feature selection method called Weighted GCNN (WGCNN), which can capture non-linear interactions and solve multi-class problems. The WGCNN incorporates feature weighting and statistical guided dropout to avoid overfitting. Experimental validation demonstrates that the WGCNN performs well in terms of F1 score and number of features selected.
COMPUTERS IN BIOLOGY AND MEDICINE
(2024)
Letter
Radiology, Nuclear Medicine & Medical Imaging
Paulo Mazzoncini de Azevedo-Marques, Jose Raniery Ferreira Jr
ACADEMIC RADIOLOGY
(2022)
Article
Computer Science, Information Systems
Daniel Jasbick, Lucio Santos, Paulo M. Azevedo-Marques, Agma J. M. Traina, Daniel de Oliveira, Marcos Bedo
Summary: Diversified similarity queries expand similarity searches by retrieving data elements similar to a given object but dissimilar to other elements in the result set. The quality and performance of diversity-related querying routines for exploring high-dimensional datasets are still open issues.
INFORMATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Silvio Ricardo Rodrigues Sanches, Cleber Gimenez Correa, Beatriz Regina Brum, Pedro Henrique Bugatti, Priscila Tiemi Maeda Saito, Claudinei Moreira da Silva, Elton Custodio Junior
Summary: The evaluation of a change detection algorithm involves executing it to segment videos and comparing the results with the ground truth. A metric based on difficulty maps was developed to evaluate algorithm performance. The results showed promising algorithms that classify pixels most state-of-the-art algorithms cannot classify.
IEEE LATIN AMERICA TRANSACTIONS
(2023)
Article
Computer Science, Information Systems
Marcelo Souza, William C. Horikoshi, Priscila T. M. Saito, Pedro H. Bugatti
Summary: To achieve higher productivity in soybean crops, the use of high-quality seeds is crucial. The quality of the seeds directly affects the plant's growth process. Traditional methods of seed vigor analysis, performed by human specialists, are labor-intensive and prone to failures due to the requirement of seed anatomy understanding. This paper proposes a learning-based expert approach and pipeline that simplifies the seed vigor analysis process, providing efficient and accurate results.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Natalia S. Chiari-Correia, Marcello H. Nogueira-Barbosa, Rodolfo Dias Chiari-Correia, Paulo M. Azevedo-Marques
Summary: This retrospective study trained an artificial neural network model using 3D radiomic features to differentiate benign from malignant vertebral compression fractures (VCFs) on MRI. The model achieved excellent performance with a ROC AUC of 0.98, an accuracy of 95%, a sensitivity of 93.5%, and a specificity of 96.3% in internal validation. In the validation with an independent test set, the model reached a ROC AUC of 0.97, an accuracy of 93.3%, a sensitivity of 93.3%, and a specificity of 93.3%. The proposed model has promising potential as an aid to radiologists in characterizing VCFs.
JOURNAL OF DIGITAL IMAGING
(2023)
Article
Audiology & Speech-Language Pathology
Maria Carolina Gironde Ataide, Filipe Andrade Bernardi, Paulo Mazzoncini de Azevedo Marques, Claudia Maria de Felicio
Summary: This study aims to develop, analyze, and improve a web version of the Orofacial Myofunctional Evaluation with Scores (OMES) protocol, and investigate the relationship between usability judgments and the evaluators' prior experience, as well as the impact of the interface on learning. The OMES-Web showed excellent usability levels and high satisfaction among participants, regardless of their level of experience.
Proceedings Paper
Computer Science, Artificial Intelligence
Cristiano N. de O. Bassani, Prisicla T. M. Saito, Pedro H. Bugatti
Summary: In this study, we addressed the issue of limited labeled samples in convolutional neural networks by applying the semi-supervised paradigm and evaluated the proposed approach on three public datasets. Our results showed that our method achieved up to 88% improvement in accuracy compared to the supervised paradigm.
ARTIFICIAL INTELLIGENCE AND SOFT COMPUTING, ICAISC 2022, PT II
(2023)
Article
Biology
J. G. Maciel, C. E. G. Salmon, B. S. Hosseini, P. M. Azevedo-Marques, F. J. A. de Paula, M. H. Nogueira-Barbosa
Summary: This study evaluated the bone texture attributes (TA) extracted from routine lumbar spine MRI and their correlation with vertebral fragility fractures (VFF) and bone mineral density (BMD). The results showed that two TA (cluster tendency and variance) were significantly lower in the fracture group. A significant correlation was also found between BMD and several texture attributes.
BRAZILIAN JOURNAL OF MEDICAL AND BIOLOGICAL RESEARCH
(2022)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Lucas S. Rodrigues, Thiago G. Vespa, Igor A. R. Eleuterio, Willian D. Oliveira, Agma J. M. Traina, Caetano Traina Jr
Summary: The paper introduces MiDaS, a framework for handling missing data in KDD processes, which can be applied to similarity assessment for complex data. Experimental results show that MiDaS is well-suited for dealing with incompleteness, enhancing data analysis in various KDD scenarios.
COMPUTATIONAL SCIENCE, ICCS 2022, PT IV
(2022)
Proceedings Paper
Computer Science, Information Systems
Jonathan S. Ramos, Erikson J. de Aguiar, Ivar Belizario, Marcus V. L. Costa, Jamilly G. Maciel, Mirela T. Cazzolato, Caetano Traina, Marcello H. Nogueira-Barbosa, Agma J. M. Traina
Summary: BMD is not ideal for predicting VFF, thus a study using machine learning and DL techniques to assess VFF found that DL methods achieved better results.
2022 IEEE 35TH INTERNATIONAL SYMPOSIUM ON COMPUTER-BASED MEDICAL SYSTEMS (CBMS)
(2022)
Proceedings Paper
Computer Science, Information Systems
Marcos Bedo, Jonathan S. Ramos, Agma J. M. Traina, Caetano Traina Jr, Marcello H. Nogueira-Barbosa, Paulo M. Azevedo-Marques
Summary: Osteoporosis is a systemic disorder that reduces bone density and increases the risk of vertebral fractures. This study explores the use of radiomic features in magnetic resonance imaging (MRI) to identify and predict fragility fractures. The results show that a content-based image retrieval (CBIR) tool with embedded radiomic features accurately detects fragility fractures and predicts future fractures.
2022 IEEE 35TH INTERNATIONAL SYMPOSIUM ON COMPUTER-BASED MEDICAL SYSTEMS (CBMS)
(2022)
Proceedings Paper
Engineering, Biomedical
Karem Daiane Marcomini, Diego Armando Cardona Cardenas, Agma Juci Machado Traina, Jose Eduardo Krieger, Marco Antonio Gutierrez
Summary: This paper proposed a deep learning-based approach to simultaneously suggest a diagnosis for COVID-19 and localize lung opacity areas in CXR images. The classification task achieved high accuracy and AUC on the test set, while the opacity detector showed a decent mAP in positive results evaluation.
MEDICAL IMAGING 2022: COMPUTER-AIDED DIAGNOSIS
(2022)
Proceedings Paper
Engineering, Biomedical
Erikson J. de Aguiar, Karem D. Marcomini, Felipe A. Quirino, Marco A. Gutierrez, Caetano Traina, Agma J. M. Traina
Summary: This study investigates the impact of Adversarial Attacks on DL models for classifying X-ray images of COVID-19 cases. The research finds that DL models for COVID-19 are vulnerable to Adversarial Examples, with a significant reduction in performance when attacked.
MEDICAL IMAGING 2022: COMPUTER-AIDED DIAGNOSIS
(2022)
Article
Computer Science, Artificial Intelligence
Ammar N. Abbas, Georgios C. Chasparis, John D. Kelleher
Summary: Deep reinforcement learning has significant potential in industrial decision-making, but its lack of interpretability poses challenges for safety-critical systems. This paper introduces a novel approach that combines probabilistic modeling and reinforcement learning, addressing these challenges and achieving excellent results in predictive maintenance for turbofan engines.
DATA & KNOWLEDGE ENGINEERING
(2024)
Article
Computer Science, Artificial Intelligence
Tongzhao Xu, Turdi Tohti, Askar Hamdulla
Summary: This paper proposes a multi-hop KGQA model that combines global and item-by-item reasoning fusion. It introduces a convolutional attention reasoning mechanism and serial prediction of relations to form reasoning paths, effectively addressing the issues of ignoring intermediate path reasoning and information interaction. The proposed model achieves significant accuracy improvement on three datasets.
DATA & KNOWLEDGE ENGINEERING
(2024)
Article
Computer Science, Artificial Intelligence
Ilias Dimitriadis, George Dialektakis, Athena Vakali
Summary: The high growth of Online Social Networks (OSNs) has led to the emergence of social bots, which pose high-level security threats. This paper proposes an adaptive bot detection framework called CALEB based on CGAN and AC-GAN, which can simulate bot evolution and enhance detection performance. Experimental results show that the proposed approach outperforms previous methods in detecting new unseen bots.
DATA & KNOWLEDGE ENGINEERING
(2024)
