AI models for automated segmentation of engineered polycystic kidney tubules.
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
03 Feb 2024
03 Feb 2024
Historique:
received:
26
06
2023
accepted:
21
01
2024
medline:
4
2
2024
pubmed:
4
2
2024
entrez:
3
2
2024
Statut:
epublish
Résumé
Autosomal dominant polycystic kidney disease (ADPKD) is a monogenic, rare disease, characterized by the formation of multiple cysts that grow out of the renal tubules. Despite intensive attempts to develop new drugs or repurpose existing ones, there is currently no definitive cure for ADPKD. This is primarily due to the complex and variable pathogenesis of the disease and the lack of models that can faithfully reproduce the human phenotype. Therefore, the development of models that allow automated detection of cysts' growth directly on human kidney tissue is a crucial step in the search for efficient therapeutic solutions. Artificial Intelligence methods, and deep learning algorithms in particular, can provide powerful and effective solutions to such tasks, and indeed various architectures have been proposed in the literature in recent years. Here, we comparatively review state-of-the-art deep learning segmentation models, using as a testbed a set of sequential RGB immunofluorescence images from 4 in vitro experiments with 32 engineered polycystic kidney tubules. To gain a deeper understanding of the detection process, we implemented both pixel-wise and cyst-wise performance metrics to evaluate the algorithms. Overall, two models stand out as the best performing, namely UNet++ and UACANet: the latter uses a self-attention mechanism introducing some explainability aspects that can be further exploited in future developments, thus making it the most promising algorithm to build upon towards a more refined cyst-detection platform. UACANet model achieves a cyst-wise Intersection over Union of 0.83, 0.91 for Recall, and 0.92 for Precision when applied to detect large-size cysts. On all-size cysts, UACANet averages at 0.624 pixel-wise Intersection over Union. The code to reproduce all results is freely available in a public GitHub repository.
Identifiants
pubmed: 38310171
doi: 10.1038/s41598-024-52677-1
pii: 10.1038/s41598-024-52677-1
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
2847Informations de copyright
© 2024. The Author(s).
Références
Willey, C. J. et al. Prevalence of autosomal dominant polycystic kidney disease in the European union. Nephrol. Dial. Transplant. 32, 1356–1363 (2017).
pubmed: 27325254
Lanktree, M. B. et al. Prevalence estimates of polycystic kidney and liver disease by population sequencing. J. Am. Soc. Nephrol. 29, 2593–2600 (2018).
doi: 10.1681/ASN.2018050493
pubmed: 30135240
pmcid: 6171271
Bergmann, C. et al. Polycystic kidney disease. Nat. Rev. Dis. Primers 4, 1–24 (2018).
doi: 10.1038/s41572-018-0047-y
Vasileva, V. Y., Sultanova, R. F., Sudarikova, A. V. & Ilatovskaya, D. V. Insights into the molecular mechanisms of polycystic kidney diseases. Front. Physiol. 12, 693130 (2021).
doi: 10.3389/fphys.2021.693130
pubmed: 34566674
pmcid: 8456103
Ghata, J. & Cowley, B. D. Jr. Polycystic kidney disease. Compr. Physiol. 7, 945–975 (2011).
Perico, N. et al. Octreotide-lar in later-stage autosomal dominant polycystic kidney disease (Aladin 2): A randomized, double-blind, placebo-controlled, multicenter trial. PLoS Med. 16, e1002777 (2019).
doi: 10.1371/journal.pmed.1002777
pubmed: 30951521
pmcid: 6450618
Benedetti, V. et al. Engineered kidney tubules for modeling patient-specific diseases and drug discovery. EBioMedicine 33, 253–268 (2018).
doi: 10.1016/j.ebiom.2018.06.005
pubmed: 30049385
pmcid: 6085557
Brizi, V., Benedetti, V., Lavecchia, A. M. & Xinaris, C. Engineering kidney tissues for polycystic kidney disease modeling and drug discovery. In Methods in Cell Biology Vol. 153 (eds Brizi, V. et al.) 113–132 (Elsevier, 2019).
Hesamian, M. H., Jia, W., He, X. & Kennedy, P. Deep learning techniques for medical image segmentation: Achievements and challenges. J. Digit. Imaging 32, 582–596 (2019).
doi: 10.1007/s10278-019-00227-x
pubmed: 31144149
pmcid: 6646484
Ronneberger, O., Fischer, P. & Brox, T. U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015: 18th International Conference, Munich, Germany, October 5–9, 2015, Proceedings, Part III 18 (eds Ronneberger, O. et al.) 234–241 (Springer, 2015).
Chen, H., Qi, X., Yu, L. & Heng, P.-A. Dcan: deep contour-aware networks for accurate gland segmentation. In Proc. IEEE Conf. Comput. Vision Pattern Recogn., 2487–2496 (2016).
Çiçek, Ö., Abdulkadir, A., Lienkamp, S. S., Brox, T. & Ronneberger, O. 3d u-net: Learning dense volumetric segmentation from sparse annotation. In Medical Image Computing and Computer-Assisted Intervention-MICCAI 2016: 19th International Conference, Athens, Greece, October 17–21, 2016, Proceedings, Part II 19, 424–432 (eds Çiçek, Ö. et al.) (Springer, 2016).
Gordienko, Y. et al. Deep learning with lung segmentation and bone shadow exclusion techniques for chest x-ray analysis of lung cancer. In Advances in Computer Science for Engineering and Education 13 (eds Gordienko, Y. et al.) 638–647 (Springer, 2019).
doi: 10.1007/978-3-319-91008-6_63
Zhou, Z., Siddiquee, M. M. R., Tajbakhsh, N. & Liang, J. Unet++: A nested u-net architecture for medical image segmentation. In Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support (eds Zhou, Z. et al.) 3–11 (Springer, 2018).
doi: 10.1007/978-3-030-00889-5_1
Kim, T., Lee, H. & Kim, D. Uacanet: Uncertainty augmented context attention for polyp segmentation. In Proc. 29th ACM Int. Conf. Multim., 2167–2175 (2021).
Yao, L. et al. Application of artificial intelligence in renal disease. Clin. eHealth 4, 54–61. https://doi.org/10.1016/j.ceh.2021.11.003 (2021).
doi: 10.1016/j.ceh.2021.11.003
Shin, T. Y. et al. Expert-level segmentation using deep learning for volumetry of polycystic kidney and liver. Investig. Clin. Urol. 61, 555 (2020).
doi: 10.4111/icu.20200086
pubmed: 33135401
pmcid: 7606119
Athavale, A. M. et al. Development and validation of a deep learning model to quantify interstitial fibrosis and tubular atrophy from kidney ultrasonography images. JAMA Netw. Open 4, e2111176 (2021).
doi: 10.1001/jamanetworkopen.2021.11176
pubmed: 34028548
pmcid: 8144924
Park, S. et al. Artificial intelligence with kidney disease: A scoping review with bibliometric analysis. Medicine 100, e25422 (2021).
doi: 10.1097/MD.0000000000025422
pubmed: 33832141
pmcid: 8036048
Chorley, C. Healx and the pkd charity collaborate to apply ai to the discovery of novel treatments for rare kidney diseases. https://healx.ai/healx-and-the-pkd-charity-collaborate-to-apply-ai-to-the-discovery-of-novel-treatments-for-rare-kidney-diseases/ (2021).
Goel, A. et al. Deployed deep learning kidney segmentation for polycystic kidney disease MRI. Radiol. Artif. Intell. 4, e210205 (2022).
doi: 10.1148/ryai.210205
pubmed: 35391774
pmcid: 8980881
Raj, A. et al. Deep learning-based total kidney volume segmentation in autosomal dominant polycystic kidney disease using attention, cosine loss, and sharpness aware minimization. Diagnostics 12, 1159 (2022).
doi: 10.3390/diagnostics12051159
pubmed: 35626314
pmcid: 9139731
Rizwan, I., Haque, I. & Neubert, J. Deep learning approaches to biomedical image segmentation. Inform. Med. Unlocked 18, 100297. https://doi.org/10.1016/j.imu.2020.100297 (2020).
doi: 10.1016/j.imu.2020.100297
Xiangbin, L., Song, L., Liu, S. & Zhang, Y. A review of deep-learning-based medical image segmentation methods. Sustainability 13, 1224. https://doi.org/10.3390/su13031224 (2021).
doi: 10.3390/su13031224
Dong, K. et al. Renal plasticity revealed through reversal of polycystic kidney disease in mice. Nat. Genet. 53, 1649–63 (2021).
doi: 10.1038/s41588-021-00946-4
pubmed: 34635846
pmcid: 9278957
Bae, K. T. et al. Growth pattern of kidney cyst number and volume in autosomal dominant polycystic kidney disease. Clin. J. Am. Soc. Nephrol. 14, 823–833 (2019).
doi: 10.2215/CJN.10360818
pubmed: 31088850
pmcid: 6556721
Magistroni, R., Corsi, C., Martí, T. & Torra, R. A review of the imaging techniques for measuring kidney and cyst volume in establishing autosomal dominant polycystic kidney disease progression. Nephrology 48, 67–78 (2018).
Kline, T. L. et al. Automatic semantic segmentation of kidney cysts in MR images of patients affected by autosomal-dominant polycystic kidney disease. Abdom. Radiol. 46, 1053–1061 (2021).
doi: 10.1007/s00261-020-02748-4
Gregory, A. V. et al. Semantic instance segmentation of kidney cysts in MR images: A fully automated 3D approach developed through active learning. J. Digit. Imaging 34, 773–787 (2021).
doi: 10.1007/s10278-021-00452-3
pubmed: 33821360
pmcid: 8455788
Cordido, A., Cernadas, E., Fernández-Delgado, M. & García-González, M. A. CystAnalyser: A new software tool for the automatic detection and quantification of cysts in Polycystic Kidney and Liver Disease, and other cystic disorders. PLoS Comput. Biol. 16, e1008337 (2020).
doi: 10.1371/journal.pcbi.1008337
pubmed: 33090995
pmcid: 7608985
Asawa, R. R. et al. A high-throughput screening platform for polycystic kidney disease (pkd) drug repurposing utilizing murine and human adpkd cells. Sci. Rep. 10, 4203. https://doi.org/10.1038/s41598-020-61082-3 (2020).
doi: 10.1038/s41598-020-61082-3
pubmed: 32144367
pmcid: 7060218
Kan, A. Machine learning applications in cell image analysis. Immuno. Cell Biol. 95, 525–530. https://doi.org/10.1038/icb.2017.16 (2017).
doi: 10.1038/icb.2017.16
Dima, A. A. et al. Comparison of segmentation algorithms for fluorescence microscopy images of cells. Cytometry A 79A, 545–559. https://doi.org/10.1002/cyto.a.21079 (2011).
doi: 10.1002/cyto.a.21079
Moen, E. et al. Deep learning for cellular image analysis. Nat. Methods 16, 1233–1246. https://doi.org/10.1038/s41592-019-0403-1 (2019).
doi: 10.1038/s41592-019-0403-1
pubmed: 31133758
pmcid: 8759575
Zhang, H., Botler, M. & Kooman, J. P. Deep learning for image analysis in kidney care. Adv. Kidney Dis. Health 30, 25–32. https://doi.org/10.1053/j.akdh.2022.11.003 (2023).
doi: 10.1053/j.akdh.2022.11.003
pubmed: 36723278
Deng, S. et al. Deep learning in digital pathology image analysis: A survey. Front. Med. 14, 470–487. https://doi.org/10.1007/s11684-020-0782-9 (2020).
doi: 10.1007/s11684-020-0782-9
pubmed: 32728875
Kar, A. et al. Assessment of deep learning algorithms for 3d instance segmentation of confocal image datasets. bioRxiv https://doi.org/10.1101/2021.06.09.447748 (2021).
doi: 10.1101/2021.06.09.447748
pubmed: 34729558
pmcid: 8562540
Wada, K. Labelme: Image polygonal annotation with python. https://github.com/wkentaro/labelme , https://doi.org/10.5281/zenodo.5711226 .
Shorten, C. & Khoshgoftaar, T. M. A survey on image data augmentation for deep learning. J. Big Data 6, 1–48 (2019).
doi: 10.1186/s40537-019-0197-0
Deng, J. et al. Imagenet: A large-scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition (eds Deng, J. et al.) 248–255 (IEEE, 2009).
doi: 10.1109/CVPR.2009.5206848
Liu, X., Song, L., Liu, S. & Zhang, Y. A review of deep-learning-based medical image segmentation methods. Sustainability 13, 1224 (2021).
doi: 10.3390/su13031224
Kromp, F. et al. Evaluation of deep learning architectures for complex immunofluorescence nuclear image segmentation. IEEE Trans. Med. Imaging 40, 1934–1949 (2021).
doi: 10.1109/TMI.2021.3069558
pubmed: 33784615
Huang, C.-H., Wu, H.-Y. & Lin, Y.-L. Hardnet-mseg: A simple encoder-decoder polyp segmentation neural network that achieves over 0.9 mean dice and 86 fps. Preprint at arXiv:2101.07172 (2021).
Chao, P., Kao, C.-Y., Ruan, Y.-S., Huang, C.-H. & Lin, Y.-L. Hardnet: A low memory traffic network. Proc. IEEE/CVF Int. Conf. Comput. Vision, 3552–3561 (2019).
Fan, D.-P. et al. Pranet: Parallel reverse attention network for polyp segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention (eds Fan, D.-P. et al.) 263–273 (Springer, 2020).
Loshchilov, I. & Hutter, F. Sgdr: Stochastic gradient descent with warm restarts. Preprint at arXiv:1608.03983 (2016).
Monaco, S. et al. Cyst segmentation on kidney tubules by means of u-net deep-learning models. In 2021 IEEE International Conference on Big Data (Big Data) (eds Monaco, S. et al.) 3923–3926 (IEEE, 2021).
doi: 10.1109/BigData52589.2021.9671669
Isensee, F., Jaeger, P. F., Kohl, S. A., Petersen, J. & Maier-Hein, K. H. nnu-net: A self-configuring method for deep learning-based biomedical image segmentation. Nat. Methods 18, 203–211 (2021).
doi: 10.1038/s41592-020-01008-z
pubmed: 33288961