sapiСистемный анализ и прикладная информатика«System analysis and applied information science»2309-49232414-0481Belarusian National Technical University10.21122/2309-4923-2025-1-44-53sapi-730Research ArticleОбработка информации и принятие решенийData processing and decision–makingLANet для сегментации медицинских изображенийLANet for medical image segmentationЧжаоДиZhaoDi

Ди Чжао, аспирант кафедры информационных технологий в автоматизированных системах

г. Минск

Di Zhao, Postgraduate at the Department of Information Technologies in Automated Systems

Minsk

3189124246@qq.comТанИ.TangYi

И Тан, аспирант кафедры информационных технологий в автоматизированных системах

г. Минск

Yi Tang, Postgraduate at the Department of Information Technologies in Automated Systems

Minsk

tangyijcb@163.comПерцевД. Ю.PertsauD. Y.

Перцев Д.Ю., кандидат технических наук, доцент, доцент кафедры электронных вычислительных машин

г. Минск

Pertsau D., PhD, Associate Professor, Associate Professor of Electronic Computing Machines Department

Minsk

pertsev@bsuir.byГуриновичА. Б.GourinovitchA. B.

Гуринович А.Б., доцент кафедры информационных технологий автоматизированных систем

г. Минск

Gourinovitch A.B., Assistant Professor at the Department of Information Technologies in Automated Systems

Minsk

gurinovich@bsuir.byКуприяноваД. В.KupryianavaD. V.

Куприянова Д.В., старший преподаватель, исследователь, кафедра электронных вычислительных машин

г. Минск

Kupryianava D., Senior Lecturer, Electronic Computing Machines Department

Minsk

kupriuanova@bsuir.byБелорусский государственный университет информатики и радиоэлектроникиБеларусьBelarusian State University of Informatics and RadioelectronicsBelarus202507042025014453Copyright © Чжао Д., Тан И., Перцев Д.Ю., Гуринович А.Б., Куприянова Д.В., 20252025Чжао Д., Тан И., Перцев Д.Ю., Гуринович А.Б., Куприянова Д.В.Zhao D., Tang Y., Pertsau D.Y., Gourinovitch A.B., Kupryianava D.V.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://sapi.bntu.by/jour/article/view/730

В данной работе представлена оригинальная модель LANet, предназначенная для улучшения результатов сегментации медицинских изображений и основанная на нейронной сети MobileViT. Разработанные и интегрированные блоки Efficient Fusion Attention и Adaptive Feature Fusion улучшают качество извлечения признаков и уменьшают избыточность данных. Эффективность представленных блоков подтверждена множеством экспериментов, включая оценку точности на различных наборах данных, на основе таких метрик, как Dice, Precision, Recall, mIoU, оценку производительности модели, а также исследование абляции.

The paper presents an original LANet model for improving medical image segmentation results based on MobileViT neural network. The developed and integrated Efficient Fusion Attention and Adaptive Feature Fusion blocks improve the quality of feature extraction and reduce data redundancy. The effectiveness of the presented blocks is validated by multiple experiments, including accuracy evaluation on different datasets, based on metrics such as Dice, Precision, Recall, mIoU, model performance evaluation, and ablation study.

LANetLightweight Attention Networkблок эффективного слияния вниманияблок адаптивного слияния признаковмеханизм вниманияLANetLightweight Attention NetworkEfficient Fusion Attention BlockAdaptive Feature Fusion Decoding BlockAttention MechanismДанная работа выполнена при поддержке Китайского стипендиального совета (CSC).ReferencesTaghanaki, S.A. Deep semantic segmentation of natural and medical images: a review / S.A.Taghanaki, [etc.] // Artificial Intelligence Review. – 2021. – Vol. 54 – P. 137–178. – DOI: 10.1007/s10462-020-09854-1Taghanaki, S.A. Deep semantic segmentation of natural and medical images: a review / S.A.Taghanaki, [etc.] // Artificial Intelligence Review. – 2021. – Vol. 54 – P. 137–178. – DOI: 10.1007/s10462-020-09854-1Tong, L. Aclustering-aidedapproachfordiagnosisprediction: Acasestudyofelderlyfall/ L. Tong,[etc.]// 2022 IEEE 46th Annual Computers, Software, and Applications Conference. – 2022. – P. 337–342. – DOI: 10.1109/COMPSAC54236.2022.00054Tong, L. Aclustering-aidedapproachfordiagnosisprediction: Acasestudyofelderlyfall/ L. Tong,[etc.]// 2022 IEEE 46th Annual Computers, Software, and Applications Conference. – 2022. – P. 337–342. – DOI: 10.1109/COMPSAC54236.2022.00054Yu, H. A scalable region-based level set method using adaptive bilateral filter for noisy image segmentation / H. Yu, F. He, Y. Pan // Multimedia Tools and Applications. – 2020. – Vol. 79. – P.5743-5765. – DOI: 10.1007/s11042-019-08493-1Yu, H. A scalable region-based level set method using adaptive bilateral filter for noisy image segmentation / H. Yu, F. He, Y. Pan // Multimedia Tools and Applications. – 2020. – Vol. 79. – P.5743-5765. – DOI: 10.1007/s11042-019-08493-1Mehta, S. MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer / S. Mehta, M. Rastegari // arXiv preprint. – 2021. – arXiv: 2110.02178.Mehta, S. MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer / S. Mehta, M. Rastegari // arXiv preprint. – 2021. – arXiv: 2110.02178.Li, J. SCConv: Spatial and Channel Reconstruction Convolution for Feature Redundancy / J. Li, Y. Wen, L. He // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. – 2023. – P. 6153-6162. – DOI: 10.1109/CVPR52729.2023.00596Li, J. SCConv: Spatial and Channel Reconstruction Convolution for Feature Redundancy / J. Li, Y. Wen, L. He // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. – 2023. – P. 6153-6162. – DOI: 10.1109/CVPR52729.2023.00596Qiu, J. SlimConv: Reducing Channel Redundancy in Convolutional Neural Networks by Features Recombining / J. Qiu, [etc.] // IEEE Transactions on Image Processing. – 2021. – Vol. 30. – P. 6434-6445. – DOI: 10.1109/TIP.2021.3093795Qiu, J. SlimConv: Reducing Channel Redundancy in Convolutional Neural Networks by Features Recombining / J. Qiu, [etc.] // IEEE Transactions on Image Processing. – 2021. – Vol. 30. – P. 6434-6445. – DOI: 10.1109/TIP.2021.3093795Oktay, O. Attention U-Net: Learning Where to Look for the Pancreas / O. Oktay [etc.] // arXiv preprint. – 2018. – arXiv:1804.03999.Oktay, O. Attention U-Net: Learning Where to Look for the Pancreas / O. Oktay [etc.] // arXiv preprint. – 2018. – arXiv:1804.03999.Jha, D. ResUNet++: An Advanced Architecture for Medical Image Segmentation / D. Jha, [etc.] // 2019 IEEE International Symposium on Multimedia (ISM). – 2019. – P. 225-235. – DOI: 10.1109/ISM46123.2019.00049Jha, D. ResUNet++: An Advanced Architecture for Medical Image Segmentation / D. Jha, [etc.] // 2019 IEEE International Symposium on Multimedia (ISM). – 2019. – P. 225-235. – DOI: 10.1109/ISM46123.2019.00049Zhou, Z. UNet++: A Nested U-Net Architecture for Medical Image Segmentation / Z. Zhou, [etc.] // Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. – 2018. – P. 3-11. – DOI: 10.1007/978-3-030-00889-5_1Zhou, Z. UNet++: A Nested U-Net Architecture for Medical Image Segmentation / Z. Zhou, [etc.] // Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support. – 2018. – P. 3-11. – DOI: 10.1007/978-3-030-00889-5_1Jha, D. DoubleU-Net: A Deep Convolutional Neural Network for Medical Image Segmentation / D. Jha [etc.] // 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems (CBMS). – 2020. – P. 558-564. – DOI: 10.1109/CBMS49503.2020.00111Jha, D. DoubleU-Net: A Deep Convolutional Neural Network for Medical Image Segmentation / D. Jha [etc.] // 2020 IEEE 33rd International Symposium on Computer-Based Medical Systems (CBMS). – 2020. – P. 558-564. – DOI: 10.1109/CBMS49503.2020.00111Berman, M. The Lovasz-Softmax Loss: A Tractable Surrogate for the Optimization of the Intersection-Over-Union Measure in Neural Networks / M. Berman, T.A. Rannen, M.B. Blaschko // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). – 2018. – P. 4413-4421. – DOI: 10.1109/CVPR.2018.00464Berman, M. The Lovasz-Softmax Loss: A Tractable Surrogate for the Optimization of the Intersection-Over-Union Measure in Neural Networks / M. Berman, T.A. Rannen, M.B. Blaschko // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). – 2018. – P. 4413-4421. – DOI: 10.1109/CVPR.2018.00464Jham, D. Kvasir-SEG: A Segmented Polyp Dataset / D. Jham, [etc.] // International Conference on Multimedia Modeling. – 2019. – P. 451-462. – DOI: 10.1007/978-3-030-37734-2_37Jham, D. Kvasir-SEG: A Segmented Polyp Dataset / D. Jham, [etc.] // International Conference on Multimedia Modeling. – 2019. – P. 451-462. – DOI: 10.1007/978-3-030-37734-2_37Bernal, J. WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians / J. Bernal, [etc.] // Computerized Medical Imaging and Graphics. – 2015. – Vol. 43. – P. 99-111. – DOI: 10.1016/j.compmedimag.2015.02.007Bernal, J. WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians / J. Bernal, [etc.] // Computerized Medical Imaging and Graphics. – 2015. – Vol. 43. – P. 99-111. – DOI: 10.1016/j.compmedimag.2015.02.007Tajbakhsh, N. Automated Polyp Detection in Colonoscopy Videos Using Shape and Context Information / N. Tajbakhsh, S.R. Gurudu, J. Liang // IEEE Transactions on Medical Imaging. – 2016. – Vol. 35, Issue 2. – P. 630-644. – DOI: 10.1109/TMI.2015.2487997.Tajbakhsh, N. Automated Polyp Detection in Colonoscopy Videos Using Shape and Context Information / N. Tajbakhsh, S.R. Gurudu, J. Liang // IEEE Transactions on Medical Imaging. – 2016. – Vol. 35, Issue 2. – P. 630-644. – DOI: 10.1109/TMI.2015.2487997.Caicedo, J. C. Nucleus segmentation across imaging experiments: the 2018 Data Science Bowl / J.C. Caicedo, [etc.] // Nature methods. – 2019. – Vol.16. – P. 1247-1253. – DOI: 10.1038/s41592-019-0612-7Caicedo, J. C. Nucleus segmentation across imaging experiments: the 2018 Data Science Bowl / J.C. Caicedo, [etc.] // Nature methods. – 2019. – Vol.16. – P. 1247-1253. – DOI: 10.1038/s41592-019-0612-7Deng, J. ImageNet: A large-scale hierarchical image database / J. Deng, [etc.] // 2009 IEEE Conference on Computer Vision and Pattern Recognition. – 2009. – P. 248-255. – DOI: 10.1109/CVPR.2009.5206848Deng, J. ImageNet: A large-scale hierarchical image database / J. Deng, [etc.] // 2009 IEEE Conference on Computer Vision and Pattern Recognition. – 2009. – P. 248-255. – DOI: 10.1109/CVPR.2009.5206848T. Ronneberger, O. U-Net: Convolutional Networks for Biomedical Image Segmentation / O. Ronneberger, P. Fischer, Brox // International Conference on Medical Image Computing and Computer-Assisted Intervention. – 2015. – P. 234-241. – DOI: 10.1007/978-3-319-24574-4_28T. Ronneberger, O. U-Net: Convolutional Networks for Biomedical Image Segmentation / O. Ronneberger, P. Fischer, Brox // International Conference on Medical Image Computing and Computer-Assisted Intervention. – 2015. – P. 234-241. – DOI: 10.1007/978-3-319-24574-4_28Tomar, N. K. FANet: a feedback attention network for improved biomedical image segmentation / N.K. Tomar, [etc.] // IEEE Transactions on Neural Networks and Learning Systems. – 2022. – Vol. 34, Issue 11. – P. 9375-9388. – DOI: 10.1109/tnnls.2022.3159394Tomar, N. K. FANet: a feedback attention network for improved biomedical image segmentation / N.K. Tomar, [etc.] // IEEE Transactions on Neural Networks and Learning Systems. – 2022. – Vol. 34, Issue 11. – P. 9375-9388. – DOI: 10.1109/tnnls.2022.3159394Long, J. Fully convolutional networks for semantic segmentation / J. Long, E. Shelhamer, T. Darrell // 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). – 2015. – P. 3431-3440. – DOI: 10.1109/CVPR.2015.7298965Long, J. Fully convolutional networks for semantic segmentation / J. Long, E. Shelhamer, T. Darrell // 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). – 2015. – P. 3431-3440. – DOI: 10.1109/CVPR.2015.7298965Fan, D. PraNet: Parallel Reverse Attention Network for Polyp Segmentation / D. Fan, [etc.] // International conference on Medical Image Computing and Computer Assisted Intervention. – 2020. – P. 263-273. – DOI: 10.1007/978-3-030-59725-2_26Fan, D. PraNet: Parallel Reverse Attention Network for Polyp Segmentation / D. Fan, [etc.] // International conference on Medical Image Computing and Computer Assisted Intervention. – 2020. – P. 263-273. – DOI: 10.1007/978-3-030-59725-2_26

The authors declare that there are no conflicts of interest present.