LungRads+AI: Automatização do Índice Lung-RADS em Laudos de TC de Tórax
Resumo
O câncer do pulmão é o segundo câncer mais frequentemente diagnosticado. Representa a forma mais mortal de neoplasia maligna, resultando em cerca de 1,8 milhão de mortes em 2020. O Lung-RADS é uma diretriz utilizada para o rastreio e o acompanhamento de lesões pulmonares suspeitas. Neste contexto, o principal objetivo deste trabalho é avaliar a eficácia de três técnicas de Reconhecimento de Entidades Nomeadas, CNN, BiLSTM e BERT, para extrair características de nódulos pulmonares em relatórios de TC de tórax e calcular o índice de probabilidade de malignidade usando a diretriz Lung-RADS. O nosso modelo com melhor desempenho foi o BiLSTM-CRF, que obteve uma precisão de 96%, uma revocação de 88% e um F1-score de 90%.
Referências
(2023). Lung ct screening reporting data system. [link]. Accessed: 2023-05-01.
Beyer, S. E., McKee, B. J., Regis, S. M., McKee, A. B., Flacke, S., El Saadawi, G., and Wald, C. (2017). Automatic Lung-RADS™ classification with a natural language processing system. J Thorac Dis, 9(9):3114–3122.
da Rocha, N. C., Barbosa, A. M. P., Schnr, Y. O., Machado-Rugolo, J., de Andrade, L. G. M., Corrente, J. E., and de Arruda Silveira, L. V. (2023). Natural language processing to extract information from portuguese-language medical records. Data, 8(1).
Devlin, J., Chang, M., Lee, K., and Toutanova, K. (2018). BERT: pre-training of deep bidirectional transformers for language understanding. CoRR, abs/1810.04805.
Fei, X., Chen, P., Wei, L., Huang, Y., Xin, Y., and Li, J. (2022). Quality management of pulmonary nodule radiology reports based on natural language processing. Bioengineering (Basel), 9(6).
Gershanik, E. F., Lacson, R., and Khorasani, R. (2011). Critical finding capture in the impression section of radiology reports. AMIA Annu Symp Proc, 2011:465–469.
Gould, M. K., Donington, J., Lynch, W. R., Mazzone, P. J., Midthun, D. E., Naidich, D. P., and Wiener, R. S. (2013). Evaluation of individuals with pulmonary nodules: When is it lung cancer?: Diagnosis and management of lung cancer, 3rd ed: American college of chest physicians evidence-based clinical practice guidelines. Chest, 143(5, Supplement):e93S–e120S.
Kang, S. K., Garry, K., Chung, R., Moore, W. H., Iturrate, E., Swartz, J. L., Kim, D. C., Horwitz, L. I., and Blecker, S. (2019). Natural language processing for identification of incidental pulmonary nodules in radiology reports. J Am Coll Radiol, 16(11):1587–1594.
Liu, H., Zhang, Z., Xu, Y., Wang, N., Huang, Y., Yang, Z., Jiang, R., and Chen, H. (2021). Use of bert (bidirectional encoder representations from transformers)-based deep learning method for extracting evidences in chinese radiology reports: Development of a computer-aided liver cancer diagnosis framework. J Med Internet Res, 23(1):e19689.
Lopes, F., Teixeira, C., and Gonçalo Oliveira, H. (2019). Contributions to clinical named entity recognition in Portuguese. In Demner-Fushman, D., Cohen, K. B., Ananiadou, S., and Tsujii, J., editors, Proceedings of the 18th BioNLP Workshop and Shared Task, pages 223–233, Florence, Italy. Association for Computational Linguistics.
MacMahon, H. and Naidich, D. P. (2017). Guidelines for management of incidental pulmonary nodules detected on ct images: From the fleischner society 2017. Radiology, 284(1):228–243. PMID: 28240562.
Mendoza, D. P., Petranovic, M., Som, A., Wu, M. Y., and Digumarthy, S. R. (2022). Lung-rads category 3 and 4 nodules on lung cancer screening in clinical practice. American Journal of Roentgenology, 219(1):55–65. PMID: 35080453.
Nakayama, H., Kubo, T., Kamura, J., Taniguchi, Y., and Liang, X. (2018). doccano: Text annotation tool for human. Software available from [link].
Nobel, J. M., Puts, S., Bakers, F. C. H., Robben, S. G. F., and Dekker, A. L. A. J. (2020). Natural language processing in dutch free text radiology reports: Challenges in a small language area staging pulmonary oncology. Journal of Digital Imaging, 33(4):1002–1008.
Patwardhan, N., Marrone, S., and Sansone, C. (2023). Transformers in the real world: A survey on nlp applications. Information, 14(4).
Pinsky, P. F., Gierada, D. S., Black, W., Munden, R., Nath, H., Aberle, D., and Kazerooni, E. (2015). Performance of Lung-RADS in the national lung screening trial: a retrospective assessment. Ann Intern Med, 162(7):485–491.
Siegel, R. L. and Miller (2023). Cancer statistics, 2023. CA: A Cancer Journal for Clinicians, 73(1):17–48.
Sugimoto, K., Takeda, T., Oh, J.-H., Wada, S., Konishi, S., Yamahata, A., Manabe, S., Tomiyama, N., Matsunaga, T., Nakanishi, K., and Matsumura, Y. (2021). Extracting clinical terms from radiology reports with deep learning. Journal of Biomedical Informatics, 116:103729.
Sung, H. and Ferlay (2021). Global cancer statistics 2020: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians, 71(3):209–249.
Vykoukal, J., Fahrmann, J. F., and Patel (2022). Contributions of circulating micrornas for early detection of lung cancer. Cancers, 14(17).
Wang, Y., Wang, L., Rastegar-Mojarad, M., Moon, S., Shen, F., Afzal, N., Liu, S., Zeng, Y., Mehrabi, S., Sohn, S., and Liu, H. (2018). Clinical information extraction applications: A literature review. Journal of Biomedical Informatics, 77:34–49.
Wolf, T. and Debuth, L. (2020). Transformers: State-of-the-art natural language processing. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 38–45, Online. Association for Computational Linguistics.
Zhang, K., Ren, W., and Zhang, Y. (2018). Attention-based bi-lstm for chinese named entity recognition. In Hong, J.-F., Su, Q., and Wu, J.-S., editors, Chinese Lexical Semantics, pages 643–652, Cham. Springer International Publishing.
Zheng, C. and Z., B. (2021). Natural language processing to identify pulmonary nodules and extract nodule characteristics from radiology reports. Chest, 160(5):1902–1914.
Beyer, S. E., McKee, B. J., Regis, S. M., McKee, A. B., Flacke, S., El Saadawi, G., and Wald, C. (2017). Automatic Lung-RADS™ classification with a natural language processing system. J Thorac Dis, 9(9):3114–3122.
da Rocha, N. C., Barbosa, A. M. P., Schnr, Y. O., Machado-Rugolo, J., de Andrade, L. G. M., Corrente, J. E., and de Arruda Silveira, L. V. (2023). Natural language processing to extract information from portuguese-language medical records. Data, 8(1).
Devlin, J., Chang, M., Lee, K., and Toutanova, K. (2018). BERT: pre-training of deep bidirectional transformers for language understanding. CoRR, abs/1810.04805.
Fei, X., Chen, P., Wei, L., Huang, Y., Xin, Y., and Li, J. (2022). Quality management of pulmonary nodule radiology reports based on natural language processing. Bioengineering (Basel), 9(6).
Gershanik, E. F., Lacson, R., and Khorasani, R. (2011). Critical finding capture in the impression section of radiology reports. AMIA Annu Symp Proc, 2011:465–469.
Gould, M. K., Donington, J., Lynch, W. R., Mazzone, P. J., Midthun, D. E., Naidich, D. P., and Wiener, R. S. (2013). Evaluation of individuals with pulmonary nodules: When is it lung cancer?: Diagnosis and management of lung cancer, 3rd ed: American college of chest physicians evidence-based clinical practice guidelines. Chest, 143(5, Supplement):e93S–e120S.
Kang, S. K., Garry, K., Chung, R., Moore, W. H., Iturrate, E., Swartz, J. L., Kim, D. C., Horwitz, L. I., and Blecker, S. (2019). Natural language processing for identification of incidental pulmonary nodules in radiology reports. J Am Coll Radiol, 16(11):1587–1594.
Liu, H., Zhang, Z., Xu, Y., Wang, N., Huang, Y., Yang, Z., Jiang, R., and Chen, H. (2021). Use of bert (bidirectional encoder representations from transformers)-based deep learning method for extracting evidences in chinese radiology reports: Development of a computer-aided liver cancer diagnosis framework. J Med Internet Res, 23(1):e19689.
Lopes, F., Teixeira, C., and Gonçalo Oliveira, H. (2019). Contributions to clinical named entity recognition in Portuguese. In Demner-Fushman, D., Cohen, K. B., Ananiadou, S., and Tsujii, J., editors, Proceedings of the 18th BioNLP Workshop and Shared Task, pages 223–233, Florence, Italy. Association for Computational Linguistics.
MacMahon, H. and Naidich, D. P. (2017). Guidelines for management of incidental pulmonary nodules detected on ct images: From the fleischner society 2017. Radiology, 284(1):228–243. PMID: 28240562.
Mendoza, D. P., Petranovic, M., Som, A., Wu, M. Y., and Digumarthy, S. R. (2022). Lung-rads category 3 and 4 nodules on lung cancer screening in clinical practice. American Journal of Roentgenology, 219(1):55–65. PMID: 35080453.
Nakayama, H., Kubo, T., Kamura, J., Taniguchi, Y., and Liang, X. (2018). doccano: Text annotation tool for human. Software available from [link].
Nobel, J. M., Puts, S., Bakers, F. C. H., Robben, S. G. F., and Dekker, A. L. A. J. (2020). Natural language processing in dutch free text radiology reports: Challenges in a small language area staging pulmonary oncology. Journal of Digital Imaging, 33(4):1002–1008.
Patwardhan, N., Marrone, S., and Sansone, C. (2023). Transformers in the real world: A survey on nlp applications. Information, 14(4).
Pinsky, P. F., Gierada, D. S., Black, W., Munden, R., Nath, H., Aberle, D., and Kazerooni, E. (2015). Performance of Lung-RADS in the national lung screening trial: a retrospective assessment. Ann Intern Med, 162(7):485–491.
Siegel, R. L. and Miller (2023). Cancer statistics, 2023. CA: A Cancer Journal for Clinicians, 73(1):17–48.
Sugimoto, K., Takeda, T., Oh, J.-H., Wada, S., Konishi, S., Yamahata, A., Manabe, S., Tomiyama, N., Matsunaga, T., Nakanishi, K., and Matsumura, Y. (2021). Extracting clinical terms from radiology reports with deep learning. Journal of Biomedical Informatics, 116:103729.
Sung, H. and Ferlay (2021). Global cancer statistics 2020: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians, 71(3):209–249.
Vykoukal, J., Fahrmann, J. F., and Patel (2022). Contributions of circulating micrornas for early detection of lung cancer. Cancers, 14(17).
Wang, Y., Wang, L., Rastegar-Mojarad, M., Moon, S., Shen, F., Afzal, N., Liu, S., Zeng, Y., Mehrabi, S., Sohn, S., and Liu, H. (2018). Clinical information extraction applications: A literature review. Journal of Biomedical Informatics, 77:34–49.
Wolf, T. and Debuth, L. (2020). Transformers: State-of-the-art natural language processing. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 38–45, Online. Association for Computational Linguistics.
Zhang, K., Ren, W., and Zhang, Y. (2018). Attention-based bi-lstm for chinese named entity recognition. In Hong, J.-F., Su, Q., and Wu, J.-S., editors, Chinese Lexical Semantics, pages 643–652, Cham. Springer International Publishing.
Zheng, C. and Z., B. (2021). Natural language processing to identify pulmonary nodules and extract nodule characteristics from radiology reports. Chest, 160(5):1902–1914.
Publicado
25/06/2024
Como Citar
FERREIRA, Tarcísio Lima; OLIVEIRA, Marcelo Costa; VIEIRA, Thales Miranda de Almeida.
LungRads+AI: Automatização do Índice Lung-RADS em Laudos de TC de Tórax. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO APLICADA À SAÚDE (SBCAS), 24. , 2024, Goiânia/GO.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2024
.
p. 519-530.
ISSN 2763-8952.
DOI: https://doi.org/10.5753/sbcas.2024.2761.
