OriLoc: Unlimited-FoV and Orientation-Free Cross-View Geolocalization by Boni Hu, Haowei Li, Shuhui Bu, Lin Chen, Pengcheng Han

“…Our approach employs a dual-weighted soft-margin triplet loss with hard sample mining to extract discriminative features. Additionally, we develop an orientation estimation module using convolution-based sliding windows to assess similarity between satellite-view and query embeddings. …”
Get full text

Automated diagnosis of respiratory diseases from lung ultrasound videos ensuring XAI: an innovative hybrid model approach by Arefin Ittesafun Abian, Mohaimenul Azam Khan Raiaan, Asif Karim, Sami Azam, Nur Mohammad Fahad, Niusha Shafiabady, Kheng Cher Yeo, Friso De Boer

“…The eleven-ablation case study reduced training costs and redundancy. K-fold cross-validation and accuracy-loss curves demonstrated model generalization. …”
Get full text

Ultralow Saturation Intensity Topological Insulator Saturable Absorber for Gigahertz Mode-Locked Solid-State Lasers by Yi-Ran Wang, Wei-Heng Sung, Xian-Cui Su, Yue Zhao, Bai-Tao Zhang, Chung-Lung Wu, Guan-Bai He, Yuan-Yao Lin, Hong Liu, Jing-Liang He, Chao-Kuei Lee

“…The successful demonstration of Q-switched and mode-locked fiber-laser operations using a topological insulator (TI) as saturable absorber (SA) has opened an application window besides TI's originally expected features. However, to date, all-solid-state mode-locked lasers base on TISAs are still unavailable and became a desired goal not only due to their application as light sources, but also because of providing a way for deeper investigation of the nature of ultrafast dynamics present in TISAs. …”
Get full text

Validation of an AI-enabled exome/transcriptome liquid biopsy platform for early detection, MRD, disease monitoring, and therapy selection for solid tumors by J. Abraham, V. Domenyuk, N. Perdigones, S. Klimov, S. Antani, T. Yoshino, E. I. Heath, E. Lou, S. V. Liu, J. L. Marshall, W. S. El-Deiry, A. F. Shields, M. F. Dietrich, Y. Nakamura, T. Fujisawa, G. D. Demetri, A. Barker, J. Xiu, D. A. Sacchetti, S. Stahl, R. Hahn-Lowry, A. Stark, J. Swensen, G. Poste, D. D. Halbert, M. Oberley, M. Radovich, G. W. Sledge, David B. Spetzler

“…Caris Assure for therapy selection was CLIA validated using 1,910 samples. 376,197 tissue profiles along with 7,061 paired blood and tissue profiles were used to engineer features for three machine learning models. The MCED model was trained on 1,013 patients and validated on an independent set of 2,675 patients. …”
Get full text

Enhanced U-Net for Underwater Laser Range-Gated Image Restoration: Boosting Underwater Target Recognition by Peng Liu, Shuaibao Chen, Wei He, Jue Wang, Liangpei Chen, Yuguang Tan, Dong Luo, Wei Chen, Guohua Jiao

“…Built upon the U-Net architecture with added residual connections, our network combines a VGG16-based perceptual loss with Mean Squared Error (MSE) as the loss function, effectively capturing high-level semantic features while preserving critical target details during reconstruction. …”
Get full text

Machine-learning model for predicting left atrial thrombus in patients with paroxysmal atrial fibrillation by Wanli Xiong, Qiqi Cao, Lu Jia, Min Chen, Tao Liu, Qingyan Zhao, Yanhong Tang, Bo Yang, Li Li, Shaobo Shi, He Huang, Congxin Huang, China Atrial Fibrillation Center Project Team

“…The final model, refined to include nine essential features, achieved an AUC of 0.787 (95%CI 0.670–0.883). …”
Get full text

Machine learning-aided hybrid technique for dynamics of rail transit stations classification: a case study by Ahad Amini Pishro, Shiquan Zhang, Alain L’Hostis, Yuetong Liu, Qixiao Hu, Farzad Hejazi, Maryam Shahpasand, Ali Rahman, Abdelbacet Oueslati, Zhengrui Zhang

“…Accuracy in rail transit station classification is critical, as it not only strengthens the model’s predictive capabilities but also ensures more reliable data-driven decisions for transit planning and development, allowing for more precise ridership forecasts and evidence-based strategies for optimizing TOD. This classification model provides stakeholders with valuable insights into the dynamics and features of rail transit stations, supporting sustainable urban development planning.…”
Get full text

Coastal salt marsh vegetation classification using hybrid convolutional neural networks and spectral index time series images by Bolu Sun, Dong Zhang, Zhengqing Lai

“…The hybrid model integrates 2D-3D convolutions to extract spatial features and fully capture temporal dynamics, enabling accurate classification. …”
Get full text

An Efficient and Hybrid Deep Learning-Driven Model to Enhance Security and Performance of Healthcare Internet of Things by Muhammad Babar, Muhammad Usman Tariq, Basit Qureshi, Zabeeh Ullah, Fahim Arif, Zahid Khan

“…Secondly, a security module based on Bidirectional Long Short-Term Memory (BiLSTM) is implemented to recognize various forms of attacks within the H-CIoT network. …”
Get full text

Detection and classification of long terminal repeat sequences in plant LTR-retrotransposons and their analysis using explainable machine learning by Jakub Horvath, Pavel Jedlicka, Marie Kratka, Zdenek Kubat, Eduard Kejnovsky, Matej Lexa

“…Results We used machine learning methods suitable for DNA sequence classification and applied them to a large dataset of plant LTR retrotransposon sequences. We trained three machine learning models using (i) traditional model ensembles (Gradient Boosting), (ii) hybrid convolutional/long and short memory network models, and (iii) a DNA pre-trained transformer-based model using k-mer sequence representation. …”
Get full text

A method to retrieve mixed-phase cloud vertical structure from airborne lidar by E. Crosbie, E. Crosbie, J. W. Hair, A. R. Nehrir, R. A. Ferrare, C. Hostetler, T. Shingler, D. Harper, M. Fenn, M. Fenn, J. Collins, J. Collins, R. Barton-Grimley, B. Collister, K. L. Thornhill, K. L. Thornhill, C. Voigt, C. Voigt, S. Kirschler, S. Kirschler, A. Sorooshian, A. Sorooshian

“…Clouds known to be liquid-only based on ambient temperature were used to train an empirical model of the multiple-scattering depolarization that results at different ranges from the lidar. …”
Get full text

From Sedentary to Success: How Physical Activity Transforms Diabetes Management: A Systematic Review by Sorina Ispas, Andreea Nelson Twakor, Nicoleta Mihaela Mindrescu, Viorel Ispas, Doina Ecaterina Tofolean, Emanuela Mercore Hutanu, Adina Petcu, Sorin Deacu, Ionut Eduard Iordache, Cristina Ioana Bica, Lucian Cristian Petcu, Florentina Gherghiceanu, Mihaela Simona Popoviciu, Anca Pantea Stoian

“…Studies included in this review were selected based on specific criteria: randomized controlled trials involving adults aged 18 and older, published in English between January 2018 and May 2024, with full-text availability and quantifiable outcome results. …”
Get full text

Cationic substitution, dynamical stability, thermal stability, electronic and thermoelectric properties in 2D dialkali metal monoxides via DFT and ML approach by S. Chellaiya, Thomas Rueshwin, R. D. Eithiraj

“…Based on the phonon studies, both 1T-KNaO and 1T-KRbO are dynamically unstable with a slightly visible imaginary frequency. …”
Get full text

Enhancing personalized learning: AI-driven identification of learning styles and content modification strategies by Md. Kabin Hasan Kanchon, Mahir Sadman, Kaniz Fatema Nabila, Ramisa Tarannum, Riasat Khan

“…A custom dataset has been constructed in this research comprising approximately 506 samples and 22 features utilizing the Moodle learning management system (LMS), successfully categorizing students into their respective learning styles. …”
Get full text

SkinEHDLF a hybrid deep learning approach for accurate skin cancer classification in complex systems by Umesh Kumar Lilhore, Yogesh Kumar Sharma, Sarita Simaiya, Roobaea Alroobaea, Abdullah M. Baqasah, Majed Alsafyani, Afnan Alhazmi

“…., ConvNeXt, EfficientNetV2, and Swin Transformer, while integrating an adaptive attention-based feature fusion mechanism to enhance the synthesis of acquired features. …”
Get full text

A Novel Method for 3D Lung Tumor Reconstruction Using Generative Models by Hamidreza Najafi, Kimia Savoji, Marzieh Mirzaeibonehkhater, Seyed Vahid Moravvej, Roohallah Alizadehsani, Siamak Pedrammehr

“…Our solution employs a GAN model trained with a reinforcement learning (RL)-based algorithm to mitigate this imbalance and enhance segmentation accuracy. …”
Get full text

Hybrid Deep Learning Approach for Accurate Detection and Multiclass Classification of Broken Conductor Faults in Power Distribution Systems by Firas Saadoon Mohammed Al-Jumaili, Mustafa Onat

“…Moreover, we use an appropriate DenseNet pre-trained model to extract features and estimate the topology state with reliable data representation. …”
Get full text

Deep Neural Emulation of the Supermassive Black Hole Binary Population by Nima Laal, Stephen R. Taylor, Luke Zoltan Kelley, Joseph Simon, Kayhan Gültekin, David Wright, Bence Bécsy, J. Andrew Casey-Clyde, Siyuan Chen, Alexander Cingoranelli, Daniel J. D’Orazio, Emiko C. Gardiner, William G. Lamb, Cayenne Matt, Magdalena S. Siwek, Jeremy M. Wachter

“…Our analyses conclude that the NF-based emulator not only outperforms GPs in the ease and computational cost of training but also outperforms in the fidelity of the emulated GWB strain ensemble distributions.…”
Get full text

Time-Domain Versus Frequency-Embedded EEG Sequences for Sensorimotor BCI Using 1D-CNN by Simanto Saha, Mathias Baumert, Alistair Mcewan

“…The proposed time/frequency feature representation techniques with 1D-CNN outperformed CSP-based algorithms (p-value <0.05). …”
Get full text

Oil Spill Detection using Convolutional Neural Networks and Sentinel-1 SAR Imagery by E. Kalogirou, E. Kalogirou, K. Christofi, K. Christofi, D. Makri, D. Makri, M. A. Iqbal, V. La Pegna, M. Tzouvaras, M. Tzouvaras, C. Mettas, C. Mettas, D. Hadjimitsis, D. Hadjimitsis

“…Preprocessing involved a thresholding technique to enhance feature extraction and improve classification precision. …”
Get full text