Segmentation and Fractional Coverage Estimation of Soil, Illuminated Vegetation, and Shaded Vegetation in Corn Canopy Images Using CCSNet and UAV Remote Sensing by Shanxin Zhang, Jibo Yue, Xiaoyan Wang, Haikuan Feng, Yang Liu, Meiyan Shu

“…The model was evaluated using Pixel Accuracy (PA), mean Intersection over Union (mIoU), and Recall, and was benchmarked against U-Net, PSPNet and UNetFormer. …”
Get full text

DS-YOLOv7: Dense Small Object Detection Algorithm for UAV by Tao Sun, Haonan Chen, Haiying Liu, Lixia Deng, Lida Liu, Shuang Li

“…Dimensionality reduction techniques focus on reducing model parameters to facilitate the deployment of algorithms on lightweight devices. …”
Get full text

Deep Learning–Based Precision Cropping of Eye Regions in Strabismus Photographs: Algorithm Development and Validation Study for Workflow Optimization by Dawen Wu, Yanfei Li, Zeyi Yang, Teng Yin, Xiaohang Chen, Jingyu Liu, Wenyi Shang, Bin Xie, Guoyuan Yang, Haixian Zhang, Longqian Liu

“…The control experiment reduced image preparation time from 10 hours for manual cropping of 900 photos to 30 seconds with the automated model. Downstream strabismus screening task validation showed our model (with head tilt correction) achieving an area under the curve of 0.917 (95% CI 0.901‐0.933), surpassing Dlib-toolkit and faster R-CNN (both without head tilt correction) with an area under the curve of 0.856 (PP ConclusionsThis study delivers an AI-driven platform featuring a preprocessing algorithm that automates eye region cropping, correcting head tilt variations to improve image quality for AI development and clinical use. …”
Get full text

The potential of carbon emission reduction through the transition to battery electric vehicles by W. Kongwee, W. Achariyaviriya, D. Rinchumphu, P. Suttakul, A. Achariyaviriya, N. Chanpichaigosol, V. Vichiensan, M. Kii, P. Iamtrakul, Y. Hayashi

“…Furthermore, thorough survey data encompassing population activity times, home addresses, and workplace coordinates would significantly boost the model's performance. Even though the model can provide any probability of battery electric vehicle proportion. …”
Get full text

Intelligent rehabilitation in an aging population: empowering human-machine interaction for hand function rehabilitation through 3D deep learning and point cloud by Zhizhong Xing, Zhizhong Xing, Zhizhong Xing, Zhijun Meng, Gengfeng Zheng, Guolan Ma, Lin Yang, Lin Yang, Xiaojun Guo, Li Tan, Yuanqiu Jiang, Huidong Wu

“…In terms of experimental results, this research validated the superior performance of the proposed model in recognizing hand surface point clouds, with an average accuracy of 88.72%. …”
Get full text

Measurement and prediction of small molecule retention by Gram-negative bacteria based on a large-scale LC/MS screen by François Le Goff, Julien Hazemann, Lukas Christen, Geoffroy Bourquin, Gabin Pierlot, Roland Lange, Philippe Panchaud, Cornelia Zumbrunn, Oliver Peter, Georg Rueedi, Daniel Ritz

“…The ML model demonstrated robust performance across similar and dissimilar molecule subsets, showcasing its strong generalization and real-world predictive potential. …”
Get full text

Enhancing Dongba Pictograph Recognition Using Convolutional Neural Networks and Data Augmentation Techniques by Shihui Li, Lan Thi Nguyen, Wirapong Chansanam, Natthakan Iam-On, Tossapon Boongoen

“…Experimental results demonstrate that the proposed model achieves a classification accuracy of 99.43% and consistently outperforms other conventional methods, with its performance peaking at 99.84% under optimized training conditions—specifically, with 75 training epochs and a batch size of 512. …”
Get full text

Transformer-based multiple instance learning network with 2D positional encoding for histopathology image classification by Bin Yang, Lei Ding, Jianqiang Li, Yong Li, Guangzhi Qu, Jingyi Wang, Qiang Wang, Bo Liu

“…Furthermore, TMIL divides histopathological images into pseudo-bags and trains patch-level feature vectors with deep metric learning to enhance classification performance. …”
Get full text

Harnessing synergy of machine learning and nature-inspired optimization for enhanced compressive strength prediction in concrete by Abba Bashir, Esar Ahmad, Shashivendra Dulawat, Sani I. Abba

“…The visual evidence highlights several advantages, including superior quality control, cost savings, increased safety, and environmental sustainability, which underscore the effectiveness of these models. In addition, feature analysis was performed using SHAP analysis, age and cement are identified as the dominant inputs exacting influence on the CS of SCMC.…”
Get full text

An insightful analysis of CNN-based dietary medicine recognition by Mohammad Didarul Alam, Tanjir Ahmed Niloy, Aurnob Sarker Aurgho, Mahady Hasan, Md. Tarek Habib

“…These collective efforts yield highly competitive performance metrics, with accuracy metrics consistently surpassing 96 % across all deep learning models. …”
Get full text

CUNet-CLSTM: A Novel Fusion of CUNet and CLSTM for Superior Liver Cancer Detection in CT Scans by K. Vijayaprabakaran, Padmanaban Ramalingam, Rajakumar Ramalingam, A. Ilavendhan, R. Vedhapriyavadhana

“…Compared to state-of-the-art models, the proposed CUNet-CLSTM not only improves segmentation accuracy, but also mitigates overfitting through sequential feature learning and optimizes computational efficiency by providing an end-to-end framework, reducing the complexity of multi-stage pipelines. …”
Get full text

Machine Learning Approach for Estimating Magnetic Field Strength in Galaxy Clusters from Synchrotron Emission by Jiyao Zhang, Yue Hu, Alex Lazarian

“…Additionally, we have confirmed that our CNN model remains robust against noise and variations in viewing angles with sufficient training, ensuring reliable performance under a wide range of observational conditions. …”
Get full text

SwinCNet leveraging Swin Transformer V2 and CNN for precise color correction and detail enhancement in underwater image restoration by Chun Yang, Liwei Shao, Yi Deng, Jiahang Wang, Hexiang Zhai

“…Experimental results on the EUVP dataset demonstrate that SwinCNet achieves PSNR values of 24.1075 dB and 28.1944 dB on the EUVP-UI and EUVP-UD subsets, respectively. Furthermore, the model demonstrates competitive performance in reference-free evaluation metrics compared to existing methods while processing 512×512 resolution images in merely 30.32 ms—a significant efficiency improvement over conventional approaches, confirming its practical applicability in real-world underwater scenarios.…”
Get full text

BrainTumNet: multi-task deep learning framework for brain tumor segmentation and classification using adaptive masked transformers by Cheng Lv, Xu-Jun Shu, Xu-Jun Shu, Quan Liang, Jun Qiu, Zi-Cheng Xiong, Jing bo Ye, Shang bo Li, Cheng Qing Liu, Jing Zhen Niu, Sheng-Bo Chen, Hong Rao

“…We designed and implemented BrainTumNet, a deep learning-based multi-task framework featuring an improved encoder-decoder architecture, adaptive masked Transformer, and multi-scale feature fusion strategy to simultaneously perform tumor region segmentation and pathological type classification. …”
Get full text

Efficient tuna detection and counting with improved YOLOv8 and ByteTrack in pelagic fisheries by Yuanchen Cheng, Zichen Zhang, Yuqing Liu, Jie Li, Zhou Fu

“…The method uses YOLOv8n as the base model, enhanced with detail-enhanced convolution and a multi-scale feature fusion pyramid network, which significantly improves detection accuracy in complex marine environments. …”
Get full text

P-GELU: A Novel Activation Function to Optimize Whisper for Darija Speech Translation by Maria Labied, Abdessamad Belangour, Mouad Banane

“…The proposed P-GELU offers a promising balance between computational efficiency and performance gains, presenting a viable solution for enhancing Transformer-based speech models in low-resource language scenarios.…”
Get full text

A Novel Five-Level Knight Multilevel Inverter With Phase Disposition Modulation Technique by Y. Vijaya Sambhavi, R. Vijayapriya

“…The proposed 5L MLI topology is explored thoroughly using the MATLAB simulation model. The evaluation of results is also demonstrated concerning the proposed PD-PWM technique by comparing its performance with the conventional sinusoidal PWM method. …”
Get full text

Development of recognition algorithms in the control system of robot drives for plucking tea by Yan Yang, Chicherin I.V., Lijun Zhao, Chanjuan Long, Ignatieva E.A.

“…Secondly, The training is conducted in Pycharm with the processed images divided into batches, and the respective performance indicators are obtained. Finally, all the candidate boxes with confidence greater than 0.65 are retained without any missed detections, The grayscale conversion makes the processing effect more obvious when extracting feature maps during the YOLOv5 training process, because grayscale images only have black and white gradients. …”
Get full text

Robust lung segmentation in Chest X-ray images using modified U-Net with deeper network and residual blocks by Wiley Tam, Paul Babyn, Javad Alirezaie

“…This study aims to develop and evaluate a lung segmentation model based on a modified U-Net architecture. The architecture leverages techniques such as transfer learning with DenseNet201 as a feature extractor alongside dilated convolutions and residual blocks. …”
Get full text

Empowering Healthcare: TinyML for Precise Lung Disease Classification by Youssef Abadade, Nabil Benamar, Miloud Bagaa, Habiba Chaoui

“…We applied quantization techniques to ensure model efficiency. The custom CNN model achieved the highest performance, with 96% accuracy and 97% precision, recall, and F1-scores, while maintaining moderate resource usage. …”
Get full text