{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T16:28:24Z","timestamp":1775838504040,"version":"3.50.1"},"reference-count":40,"publisher":"MDPI AG","issue":"12","license":[{"start":{"date-parts":[[2023,12,18]],"date-time":"2023-12-18T00:00:00Z","timestamp":1702857600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Tuscany Region Project"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J. Imaging"],"abstract":"<jats:p>Imaging plays a key role in the clinical management of Coronavirus disease 2019 (COVID-19) as the imaging findings reflect the pathological process in the lungs. The visual analysis of High-Resolution Computed Tomography of the chest allows for the differentiation of parenchymal abnormalities of COVID-19, which are crucial to be detected and quantified in order to obtain an accurate disease stratification and prognosis. However, visual assessment and quantification represent a time-consuming task for radiologists. In this regard, tools for semi-automatic segmentation, such as those based on Convolutional Neural Networks, can facilitate the detection of pathological lesions by delineating their contour. In this work, we compared four state-of-the-art Convolutional Neural Networks based on the encoder\u2013decoder paradigm for the binary segmentation of COVID-19 infections after training and testing them on 90 HRCT volumetric scans of patients diagnosed with COVID-19 collected from the database of the Pisa University Hospital. More precisely, we started from a basic model, the well-known UNet, then we added an attention mechanism to obtain an Attention-UNet, and finally we employed a recurrence paradigm to create a Recurrent\u2013Residual UNet (R2-UNet). In the latter case, we also added attention gates to the decoding path of an R2-UNet, thus designing an R2-Attention UNet so as to make the feature representation and accumulation more effective. We compared them to gain understanding of both the cognitive mechanism that can lead a neural model to the best performance for this task and the good compromise between the amount of data, time, and computational resources required. We set up a five-fold cross-validation and assessed the strengths and limitations of these models by evaluating the performances in terms of Dice score, Precision, and Recall defined both on 2D images and on the entire 3D volume. From the results of the analysis, it can be concluded that Attention-UNet outperforms the other models by achieving the best performance of 81.93%, in terms of 2D Dice score, on the test set. Additionally, we conducted statistical analysis to assess the performance differences among the models. Our findings suggest that integrating the recurrence mechanism within the UNet architecture leads to a decline in the model\u2019s effectiveness for our particular application.<\/jats:p>","DOI":"10.3390\/jimaging9120283","type":"journal-article","created":{"date-parts":[[2023,12,18]],"date-time":"2023-12-18T08:55:29Z","timestamp":1702889729000},"page":"283","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Enhancing COVID-19 CT Image Segmentation: A Comparative Study of Attention and Recurrence in UNet Models"],"prefix":"10.3390","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-5190-0223","authenticated-orcid":false,"given":"Rossana","family":"Buongiorno","sequence":"first","affiliation":[{"name":"Institute of Information Science and Technologies, National Research Council of Italy (ISTI-CNR), 56124 Pisa, PI, Italy"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4604-2006","authenticated-orcid":false,"given":"Giulio","family":"Del Corso","sequence":"additional","affiliation":[{"name":"Institute of Information Science and Technologies, National Research Council of Italy (ISTI-CNR), 56124 Pisa, PI, Italy"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7814-5280","authenticated-orcid":false,"given":"Danila","family":"Germanese","sequence":"additional","affiliation":[{"name":"Institute of Information Science and Technologies, National Research Council of Italy (ISTI-CNR), 56124 Pisa, PI, Italy"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6217-4933","authenticated-orcid":false,"given":"Leonardo","family":"Colligiani","sequence":"additional","affiliation":[{"name":"Department of Translational Research, Academic Radiology, University of Pisa, 56124 Pisa, PI, Italy"}]},{"given":"Lorenzo","family":"Python","sequence":"additional","affiliation":[{"name":"2nd Radiology Unit, Pisa University Hospital, 56124 Pisa, PI, Italy"}]},{"given":"Chiara","family":"Romei","sequence":"additional","affiliation":[{"name":"2nd Radiology Unit, Pisa University Hospital, 56124 Pisa, PI, Italy"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2022-0804","authenticated-orcid":false,"given":"Sara","family":"Colantonio","sequence":"additional","affiliation":[{"name":"Institute of Information Science and Technologies, National Research Council of Italy (ISTI-CNR), 56124 Pisa, PI, Italy"}]}],"member":"1968","published-online":{"date-parts":[[2023,12,18]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"989","DOI":"10.1016\/j.chest.2021.10.011","article-title":"Complications of Critical COVID-19: Diagnostic and Therapeutic Considerations for the Mechanically Ventilated Patient","volume":"161","author":"Maslove","year":"2022","journal-title":"Chest"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"50","DOI":"10.2807\/1560-7917.ES.2020.25.50.2000568","article-title":"Estimating false-negative detection rate of SARS-CoV-2 by RT-PCR","volume":"25","author":"Wikramaratna","year":"2020","journal-title":"Eurosurveillance"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"E177","DOI":"10.1148\/radiol.2021203153","article-title":"Six-month Follow-up Chest CT Findings after Severe COVID-19 Pneumonia","volume":"299","author":"Han","year":"2021","journal-title":"Radiology"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"16","DOI":"10.1109\/RBME.2020.2990959","article-title":"The Role of Imaging in the Detection and Management of COVID-19: A Review","volume":"14","author":"Dong","year":"2021","journal-title":"IEEE Rev. Biomed. Eng."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"E32","DOI":"10.1148\/radiol.2020200642","article-title":"Correlation of Chest CT and RT-PCR Testing for Coronavirus Disease 2019 (COVID-19) in China: A Report of 1014 Cases","volume":"296","author":"Ai","year":"2020","journal-title":"Radiology"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"4381","DOI":"10.1007\/s00330-020-06801-0","article-title":"Chest CT manifestations of new coronavirus disease 2019 (COVID-19): A pictorial review","volume":"30","author":"Ye","year":"2020","journal-title":"Eur. Radiol."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"E41","DOI":"10.1148\/radiol.2020200343","article-title":"Chest CT for typical coronavirus disease 2019 (COVID-19) pneumonia: Relationship to negative RT-PCR testing","volume":"296","author":"Xie","year":"2020","journal-title":"Radiology"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Laino, M.E., Ammirabile, A., Posa, A., Cancian, P., Shalaby, S., Savevski, V., and Neri, E. (2021). The Applications of Artificial Intelligence in Chest Imaging of COVID-19 Patients: A Literature Review. Diagnostics, 11.","DOI":"10.3390\/diagnostics11081317"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"110028","DOI":"10.1016\/j.ejrad.2021.110028","article-title":"Artificial intelligence for imaging-based COVID-19 detection: Systematic review comparing added value of AI versus human readers","volume":"145","author":"Kriza","year":"2021","journal-title":"Eur. J. Radiol."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"612914","DOI":"10.3389\/frai.2021.612914","article-title":"AI-Empowered Computational Examination of Chest Imaging for COVID-19 Treatment: A Review","volume":"4","author":"Deng","year":"2021","journal-title":"Front. Artif. Intell."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1109\/RBME.2020.2987975","article-title":"Review of Artificial Intelligence Techniques in Imaging Data Acquisition, Segmentation, and Diagnosis for COVID-19","volume":"14","author":"Shi","year":"2021","journal-title":"IEEE Rev. Biomed. Eng."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Wang, J., Yang, X., Zhou, B., Sohn, J.J., Zhou, J., Jacob, J.T., Higgins, K.A., Bradley, J.D., and Liu, T. (2022). Review of Machine Learning in Lung Ultrasound in COVID-19 Pandemic. J. Imaging, 8.","DOI":"10.3390\/jimaging8030065"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2626","DOI":"10.1109\/TMI.2020.2996645","article-title":"Inf-Net: Automatic COVID-19 Lung Infection Segmentation From CT Images","volume":"39","author":"Fan","year":"2020","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"102205","DOI":"10.1016\/j.media.2021.102205","article-title":"COVID-19 lung infection segmentation with a novel two-stage cross-domain transfer learning framework","volume":"74","author":"Liu","year":"2021","journal-title":"Med. Image Anal."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Kumar Singh, V., Abdel-Nasser, M., Pandey, N., and Puig, D. (2021). LungINFseg: Segmenting COVID-19 Infected Regions in Lung CT Images Based on a Receptive-Field-Aware Deep Learning Framework. Diagnostics, 11.","DOI":"10.3390\/diagnostics11020158"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"6566982","DOI":"10.1155\/2022\/6566982","article-title":"Automatic COVID-19 Lung Infection Segmentation through Modified Unet Model","volume":"2022","author":"Shamim","year":"2022","journal-title":"J. Healthc. Eng."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"3090","DOI":"10.1038\/s41598-022-06931-z","article-title":"Cascaded 3D UNet architecture for segmenting the COVID-19 infection from lung CT volume","volume":"12","author":"Aswathy","year":"2022","journal-title":"Sci. Rep."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Saeedizadeh, N., Minaee, S., Kafieh, R., Yazdani, S., and Sonka, M. (2021). COVID TV-Unet: Segmenting COVID-19 chest CT images using connectivity imposed Unet. Comput. Methods Programs Biomed. Update, 1.","DOI":"10.1016\/j.cmpbup.2021.100007"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"102605","DOI":"10.1016\/j.media.2022.102605","article-title":"et al. Rapid Artificial Intelligence Solutions in a Pandemic\u2014The COVID-19-20 Lung CT Lesion Segmentation Challenge","volume":"82","author":"Roth","year":"2022","journal-title":"Med. Image Anal."},{"key":"ref_20","unstructured":"Oktay, O., Schlemper, J., Folgoc, L.L., Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Hammerla, N.Y., and Kainz, B. (2018). Attention U-Net: Learning Where to Look for the Pancreas. arXiv."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Alom, M.Z., Hasan, M., Yakopcic, C., Taha, T., and Asari, V. (2018). Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image Segmentation. arXiv.","DOI":"10.1109\/NAECON.2018.8556686"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"6625688","DOI":"10.1155\/2021\/6625688","article-title":"R2AU-Net: Attention Recurrent Residual Convolutional Neural Network for Multimodal Medical Image Segmentation","volume":"2021","author":"Zuo","year":"2021","journal-title":"Secur. Commun. Netw."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Nodirov, J., Abdusalomov, A.B., and Whangbo, T.K. (2022). Attention 3D U-Net with Multiple Skip Connections for Segmentation of Brain Tumor Images. Sensors, 22.","DOI":"10.3390\/s22176501"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Yuan, W., Peng, Y., Guo, Y., Ren, Y., and Xue, Q. (2022). DCAU-Net: Dense convolutional attention U-Net for segmentation of intracranial aneurysm images. Vis. Comput. Ind. Biomed. Art, 5.","DOI":"10.1186\/s42492-022-00105-4"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Chen, Y., Zheng, C., Zhou, T., Feng, L., Liu, L., Zeng, Q., and Wang, G. (2023). A deep residual attention-based U-Net with a biplane joint method for liver segmentation from CT scans. Comput. Biol. Med., 152.","DOI":"10.1016\/j.compbiomed.2022.106421"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"e349","DOI":"10.7717\/peerj-cs.349","article-title":"ADID-UNET\u2014A segmentation model for COVID-19 infection from lung CT scans","volume":"7","author":"Zhu","year":"2021","journal-title":"PeerJ Comput. Sci."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"16","DOI":"10.1002\/ima.22527","article-title":"Automatic COVID-19 CT segmentation using U-Net integrated spatial and channel attention mechanism","volume":"31","author":"Zhou","year":"2021","journal-title":"Int. J. Imaging Syst. Technol."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Zhao, X., Zhang, P., Song, F., Fan, G., Sun, Y., Wang, Y., Tian, Z., Zhang, L., and Zhang, G. (2021). D2A U-Net: Automatic Segmentation of COVID-19 Lesions from CT Slices with Dilated Convolution and Dual Attention Mechanism. arXiv.","DOI":"10.1016\/j.compbiomed.2021.104526"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Ahmed, I., Chehri, A., and Jeon, G. (2022). A Sustainable Deep Learning-Based Framework for Automated Segmentation of COVID-19 Infected Regions: Using U-Net with an Attention Mechanism and Boundary Loss Function. Electronics, 11.","DOI":"10.3390\/electronics11152296"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"102797","DOI":"10.1016\/j.media.2023.102797","article-title":"PDAtt-Unet: Pyramid Dual-Decoder Attention Unet for COVID-19 infection segmentation from CT-scans","volume":"86","author":"Bougourzi","year":"2023","journal-title":"Med. Image Anal."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Yin, S., Deng, H., Xu, Z., Zhu, Q., and Cheng, J. (2022). SD-UNet: A Novel Segmentation Framework for CT Images of Lung Infections. Electronics, 11.","DOI":"10.3390\/electronics11010130"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"17723","DOI":"10.1007\/s00521-022-07419-7","article-title":"R2U++: A multiscale recurrent residual U-Net with dense skip connections for medical image segmentation","volume":"34","author":"Mubashar","year":"2022","journal-title":"Neural Comput. Appl."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"11","DOI":"10.2174\/1874347102012010011","article-title":"ProgNet: COVID-19 Prognosis Using Recurrent and Convolutional Neural Networks","volume":"12","author":"Fakhfakh","year":"2020","journal-title":"Open Med. Imaging J."},{"key":"ref_34","unstructured":"Chen, X., Yao, L., and Zhang, Y. (2020). Residual Attention U-Net for Automated Multi-Class Segmentation of COVID-19 Chest CT Images. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"7127","DOI":"10.1002\/mp.15231","article-title":"CARes-UNet: Content-aware residual UNet for lesion segmentation of COVID-19 from chest CT images","volume":"48","author":"Xu","year":"2021","journal-title":"Med. Phys."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"9580991","DOI":"10.1155\/2022\/9580991","article-title":"Deep Neural Networks for Medical Image Segmentation","volume":"2022","author":"Malhotra","year":"2022","journal-title":"J. Healthc. Eng."},{"key":"ref_37","unstructured":"Bertels, J., Robben, D., Lemmens, R., and Vandermeulen, D. (2022). Convolutional neural networks for medical image segmentation. arXiv."},{"key":"ref_38","unstructured":"Tilborghs, S., Dirks, I., Fidon, L., Willems, S., Eelbode, T., Bertels, J., Ilsen, B., Brys, A., Dubbeldam, A., and Buls, N. (2020). Comparative study of deep learning methods for the automatic segmentation of lung, lesion and lesion type in CT scans of COVID-19 patients. arXiv."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Buongiorno, R., Germanese, D., Romei, C., Tavanti, L., Liperi, A., and Colantonio, S. (2021, January 10\u201315). UIP-Net: A Decoder-Encoder CNN for the Detection and Quantification of Usual Interstitial Pneumoniae Pattern in Lung CT Scan Images. Proceedings of the Pattern Recognition. ICPR International Workshops and Challenges: Virtual Event.","DOI":"10.1007\/978-3-030-68763-2_30"},{"key":"ref_40","unstructured":"Reinke, A., Maier-Hein, L., Christodoulou, E., Glocker, B., Scholz, P., Isensee, F., Kleesiek, J., Kozubek, M., Reyes, M., and Riegler, M.A. (2022, January 6\u20138). Metrics reloaded-a new recommendation framework for biomedical image analysis validation. Proceedings of the Medical Imaging with Deep Learning, Zurich, Switzerland."}],"container-title":["Journal of Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2313-433X\/9\/12\/283\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T21:40:55Z","timestamp":1760132455000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2313-433X\/9\/12\/283"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,18]]},"references-count":40,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["jimaging9120283"],"URL":"https:\/\/doi.org\/10.3390\/jimaging9120283","relation":{},"ISSN":["2313-433X"],"issn-type":[{"value":"2313-433X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,12,18]]}}}