Article
Oncology
Na Su, Wubulitalifu Dawuti, Yan Hu, Hui Zhao
Summary: This study explored the feasibility of serum Raman spectroscopy for rapid screening of cholangitis and cholangiocarcinoma (CCA). The results demonstrate that this method, combined with the support vector machine algorithm, can effectively identify cholangitis and CCA, indicating its potential as a noninvasive screening tool.
PHOTODIAGNOSIS AND PHOTODYNAMIC THERAPY
(2022)
Article
Microbiology
Thomas J. Tewes, Mario Kerst, Frank Platte, Dirk P. Bockmuehl
Summary: In this study, predictive models based on Raman spectroscopy combined with support vector machines were developed for accurate identification of microorganisms. The method showed promising results for various applications such as medical diagnostics and the food industry.
Article
Computer Science, Artificial Intelligence
Shili Peng, Wenwu Wang, Yinli Chen, Xueling Zhong, Qinghua Hu
Summary: This article presents a new idea for addressing the challenge of unifying classification and regression in machine learning. It proposes converting the classification problem into a regression problem and using regression methods to solve key problems in classification. Experimental results demonstrate that the proposed method outperforms existing algorithms in terms of prediction accuracy and model uncertainty.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Mathematics
Akash Saxena, Ahmad M. Alshamrani, Adel Fahad Alrasheedi, Khalid Abdulaziz Alnowibet, Ali Wagdy Mohamed
Summary: This paper presents the application of Support Vector Machine (SVM) in classifying power quality events using well-known signal processing techniques. The results show that SVM with attributes from both signal-processing techniques gives satisfactory results.
Article
Medicine, General & Internal
Chao-Wei Wu, Hsin-Yi Chen, Jui-Yu Chen, Ching-Hung Lee
Summary: This study aimed to analyze the diagnostic capability of Spectralis OCT parameters on glaucoma detection using the Support Vector Machine (SVM) classification method. The results showed good performance in detecting glaucomatous eyes and discriminating different degrees of glaucoma. Combining peripapillary and macular parameters provided better results.
Article
Health Care Sciences & Services
Jong-Uk Park, Yeewoong Kim, Yerin Lee, Erdenebayar Urtnasan, Kyoung-Joung Lee
Summary: This study proposes an algorithm that predicts hypoglycemic events using glucose levels and electrocardiogram data. The results show that the algorithm performs better than previous studies in predicting hypoglycemia, and has high sensitivity, specificity, and accuracy.
JOURNAL OF MEDICAL SYSTEMS
(2022)
Article
Spectroscopy
Caroleny Villalba-Hernandez, Maria de los Angeles Moyaho-Bernal, Freddy Narea-Jimenez, Hector Nahum Chavarria-Lizarraga, Maria Cecilia Galeazzi-Minutti, Rosendo Carrasco-Gutierrez, Jorge Castro-Ramos
Summary: This study predicts periodontitis by analyzing Raman spectra and biomarkers in saliva, such as albumin and alanine aminotransferase (ALT). The study uses MATLAB for data processing and analysis, ORCA software to predict fundamental frequencies and intensities, and support vector machines for spectral distinction.
JOURNAL OF RAMAN SPECTROSCOPY
(2022)
Article
Oncology
Guohua Wu, Chenchen Li, Longfei Yin, Jing Wang, Xiangxiang Zheng
Summary: In this study, a minimally invasive test method for cervical cancer in vitro was proposed by comparing Raman spectroscopy with support vector machine (SVM) model and deep belief network (DBN) model. The serum Raman spectra of cervical cancer, hysteromyoma, and healthy people were collected. The experimental results showed that the DBN classification model achieved accurate division of the sample test set and the result of cross-validation was ideal. This method improved the accuracy by 13.93%+/- 2.47% compared with the SVM method, based on 445 collected samples, providing a new direction and idea for the in vitro diagnosis of cervical diseases.
PHOTODIAGNOSIS AND PHOTODYNAMIC THERAPY
(2023)
Article
Oncology
M. Saleem, Safdar Ali, M. Bilal, Khushbakht Safdar, Mehdi Hassan
Summary: This study develops a method for diagnosing DENV infection in human blood sera using Raman spectroscopy and multivariate classification models. By analyzing the significant differences and characteristic patterns in Raman spectra, a multivariate model is built using PCA and SVM, achieving high accuracy and robustness.
PHOTODIAGNOSIS AND PHOTODYNAMIC THERAPY
(2022)
Article
Spectroscopy
Geer Teng, Qianqian Wang, Xutai Cui, Kai Wei, Wenting Xiangli, Guoyan Chen
Summary: In the clinical field, the detection and diagnosis of many diseases rely on identifying the corresponding bacteria. Raman spectroscopy and laser-induced breakdown spectroscopy (LIBS) are novel technologies used for clinical bacteria identification, combined with optimized support vector machine (SVM). Two feature-level fusion methods were proposed to improve SVM classifier performance, reduce analyzing time, and achieve 95.67% correct classification rate with the most important 10 feature lines.
JOURNAL OF RAMAN SPECTROSCOPY
(2021)
Article
Multidisciplinary Sciences
Jing Zhu, Siyu Zhang, Ruting Wang, Ruhua Fang, Lan Lei, Ji Zheng, Zhongjian Chen
Summary: This study explores an effective, noninvasive, and convenient diagnostic tool for cancer using urine based near-infrared spectroscopy (NIRS) analysis combined with machine learning algorithm. The results show that the combination of urine based NIRS analysis and machine learning can achieve high prediction accuracy, providing encouragement for further evaluation in large-scale multi-center studies.
Article
Chemistry, Analytical
Sherif Abdelfattah, Mohamed Baza, Mohamed Mahmoud, Mostafa M. Fouda, Khalid Abualsaud, Elias Yaacoub, Maazen Alsabaan, Mohsen Guizani
Summary: Machine learning, especially SVM, has been widely used in medical diagnosis. However, privacy protection and intellectual property preservation are still challenging. This paper proposes a modified encryption cryptosystem to address these issues and successfully fulfills security, privacy, and accuracy objectives in medical diagnosis.
Article
Biochemical Research Methods
Amir Nakar, Aikaterini Pistiki, Oleg Ryabchykov, Thomas Bocklitz, Petra Roesch, Juergen Popp
Summary: Raman spectroscopy was successfully used to differentiate different pathogens of the Enterobacteriaceae family, with UV-Resonance Raman spectroscopy outperforming 532 nm excitation in genus-level classification and achieving species-level classification for Klebsiella oxytoca for the first time.
JOURNAL OF BIOPHOTONICS
(2022)
Article
Computer Science, Artificial Intelligence
Xiaochen Zhou, Xudong Wang
Summary: Fed-KSVM is a federated learning scheme designed for training low-memory-consumption kernel SVM models. By decomposing the training process into subproblems and using an incremental learning algorithm, it achieves reduced memory consumption on edge devices. Additionally, by constructing a global model after training the local models, the scheme reduces communication costs while maintaining high accuracy.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Information Systems
Tiep M. Hoang, Trung Q. Duong, Hoang Duong Tuan, Sangarapillai Lambotharan, Lajos Hanzo
Summary: This article presents a framework for converting wireless signals into structured datasets for detecting active eavesdropping attacks at the physical layer using machine learning algorithms.
Article
Medicine, General & Internal
Saddam Hussain Khan, Anabia Sohail, Asifullah Khan, Yeon-Soo Lee
Summary: COVID-19 is a respiratory illness with devastating consequences globally. A new CNN architecture called STM-RENet is developed to interpret radiographic patterns from X-ray images, and a CB-STM-RENet is proposed to enhance the detection accuracy by exploiting channel boosting and learning textural variations.
Article
Computer Science, Artificial Intelligence
Umme Zahoora, Muttukrishnan Rajarajan, Zahoqing Pan, Asifullah Khan
Summary: This study introduces a Deep Contractive Autoencoder based Attribute Learning technique and an Inference Stage method based on Heterogeneous Voting Ensemble, which can effectively handle unseen classes and detect zero-day attacks.
APPLIED INTELLIGENCE
(2022)
Article
Oncology
Muhammad Mohsin Zafar, Zunaira Rauf, Anabia Sohail, Abdul Rehman Khan, Muhammad Obaidullah, Saddam Hussain Khan, Yeon Soo Lee, Asifullah Khan
Summary: This study presents a deep convolutional neural network based lymphocyte counter, which improves accuracy through a two-phase correction. Experimental results show its performance outperforms existing models and demonstrates good generalization ability.
PHOTODIAGNOSIS AND PHOTODYNAMIC THERAPY
(2022)
Article
Chemistry, Analytical
Mirza Mumtaz Zahoor, Shahzad Ahmad Qureshi, Sameena Bibi, Saddam Hussain Khan, Asifullah Khan, Usman Ghafoor, Muhammad Raheel Bhutta
Summary: A novel two-phase deep learning-based framework is proposed for the detection and categorization of brain tumors in MRIs. The framework achieves effective tumor detection through deep convolutional neural networks and machine learning classifiers, and categorizes different tumor types using a fusion of static and dynamic features.
Article
Physics, Multidisciplinary
Mirza Mumtaz Zahoor, Shahzad Ahmad Qureshi, Asifullah Khan, Aziz ul Rehman, Muhammad Rafique
Summary: In this study, a dual-channel brain tumor detection framework is proposed to improve the detection performance by using dynamic and static features. Computer experiments on a public brain tumor dataset show that the proposed framework outperforms other existing methods with high accuracy and F-score.
WAVES IN RANDOM AND COMPLEX MEDIA
(2022)
Article
Chemistry, Multidisciplinary
Aqsa Kiran, Shahzad Ahmad Qureshi, Asifullah Khan, Sajid Mahmood, Muhammad Idrees, Aqsa Saeed, Muhammad Assam, Mohamad Reda A. Refaai, Abdullah Mohamed
Summary: This paper proposes a new deep learning-based methodology that achieves a more efficient feature database through the collaboration of two effective models without using image metadata. Experiments show that the retrieval accuracy of this method is generally 97% under different types of noise.
APPLIED SCIENCES-BASEL
(2022)
Article
Multidisciplinary Sciences
Muhammad Asam, Saddam Hussain Khan, Altaf Akbar, Sameena Bibi, Tauseef Jamal, Asifullah Khan, Usman Ghafoor, Muhammad Raheel Bhutta
Summary: The interaction between devices, people, and the Internet has led to the emergence of IoT, introducing security challenges. This study proposes a CNN-based IoT malware detection architecture to address the malware detection challenge in IoT devices. The proposed architecture achieves promising malware detection capacity with 97.93% accuracy, demonstrating potential for extended applications in the future.
SCIENTIFIC REPORTS
(2022)
Article
Oncology
M. Saleem, Safdar Ali, M. Bilal, Khushbakht Safdar, Mehdi Hassan
Summary: This study develops a method for diagnosing DENV infection in human blood sera using Raman spectroscopy and multivariate classification models. By analyzing the significant differences and characteristic patterns in Raman spectra, a multivariate model is built using PCA and SVM, achieving high accuracy and robustness.
PHOTODIAGNOSIS AND PHOTODYNAMIC THERAPY
(2022)
Article
Biochemical Research Methods
Muhammad Saleem, Hina Ali, M. Bilal, Babar M. Atta, Naveed Ahmad
Summary: The potential of fluorescence spectroscopy for quality analysis of canola and mustard oil, and the effect of heating on their molecular composition, has been investigated. Carotenoids, isomers of vitamin E, and chlorophylls were identified as markers for quality assurance. The deterioration of carotenoids and vitamin E isomers increased with temperature, but both oils can be safely used for cooking/frying up to 150 degrees C.
JOURNAL OF FLUORESCENCE
(2023)
Article
Chemistry, Analytical
Anum Mushtaq, Irfan Ul Haq, Muhammad Azeem Sarwar, Asifullah Khan, Wajeeha Khalil, Muhammad Abid Mughal
Summary: Intelligent traffic management systems have gained significant attention in Intelligent Transportation Systems (ITS), and Reinforcement Learning (RL) based control methods have become increasingly popular in applications such as autonomous driving and traffic management solutions in ITS. This paper proposes an approach based on Multi-Agent Reinforcement Learning (MARL) and smart routing to improve the flow of autonomous vehicles on road networks. The evaluation of Multi-Agent Advantage Actor-Critic (MA2C) and Independent Advantage Actor-Critical (IA2C) shows the potential of these recently suggested techniques for traffic signal optimization. The effectiveness and reliability of the method are demonstrated through simulations using SUMO, a software modeling tool for traffic simulations.
Article
Computer Science, Information Systems
Faiza Babar Khan, Muhammad Hanif Durad, Asifullah Khan, Farrukh Aslam Khan, Sajjad Hussain Chauhdary, Mohammed Alqarni
Summary: Malware is a significant threat to information security, and efficient anti-malware software is crucial for protection. However, identifying malware remains challenging, especially with unknown samples. In this paper, a novel architecture based on the Relation Network is proposed for Few-Shot Learning (FSL) implementation, achieving improved classification accuracy by up to 94% with only one training instance.
Article
Optics
Muhammad Bilal, Tian Zhenyu
Summary: PIV is an important experimental technique for measuring the velocity fields of fluid flows, producing quantitative representations of instantaneous flow patterns. It has applications in various fields and has developed significantly in terms of techniques and trends. PIV is widely used in medical research, energy fuels, combustion, flow field measurement, and many other fields.
ACTA PHOTONICA SINICA
(2023)
Article
Biochemistry & Molecular Biology
Babar Manzoor Atta, M. Saleem, M. Bilal, Aziz ul Rehman, M. Fayyaz
Summary: This study investigated the application of fluorescence spectroscopy, advanced statistical techniques, and confocal microscopy for the early detection of stripe rust infection in wheat. The results showed that fluorescence emission spectra can differentiate between healthy and infected leaf samples, and the change in chlorophyll fluorescence bands could be used for the early prediction. The combination of fluorescence emission spectroscopy and chemometrics aided in the effective and timely diagnosis of plant diseases and provided the basis for remote sensing.
PHOTOCHEMICAL & PHOTOBIOLOGICAL SCIENCES
(2023)
Article
Computer Science, Information Systems
Muhammad Arif Arshad, Saddam Hussain Khan, Suleman Qamar, Muhammad Waleed Khan, Iqbal Murtza, Jeonghwan Gwak, Asifullah Khan
Summary: This article presents a novel strategy for drone navigation in complex and dynamic environments using a deep Convolutional Neural Network (CNN). The proposed method effectively navigates drones and helps them avoid obstacles. The experimental results show promising performance and suggest that the approach can be applied to real-time drone navigation and real-world flights.
Article
Computer Science, Information Systems
Suleman Qamar, Saddam Hussain Khan, Muhammad Arif Arshad, Maryam Qamar, Jeonghwan Gwak, Asifullah Khan
Summary: This study introduces an autonomous approach utilizing deep reinforcement learning for swarm navigation in complex environments, with the novel island policy optimization model and new reward functions for handling multiple dynamic targets to enhance swarm dynamics.