ABM-OCD: Advancing ovarian cancer diagnosis with attention-based models and 3D CNNs

Jenefa A.; Edward Naveen V.; Ebenezer Veemaraj; Lincy A.

doi:10.5935/jetia.v9i43.904

Autores

Jenefa A. Instituto Karunya de Tecnologia e Ciências, Coimbatore, Índia http://orcid.org/0000-0002-6697-1788 (não autenticado)
Edward Naveen V. Instituto Sri Shakthi de Engenharia e Tecnologia, Coimbatore, Índia http://orcid.org/0009-0003-5845-2173 (não autenticado)
Ebenezer Veemaraj Instituto Karunya de Tecnologia e Ciências, Coimbatore, Índia http://orcid.org/0000-0002-0801-6926 (não autenticado)
Lincy A. Faculdade Nacional de Engenharia, Kovilpatti, Índia http://orcid.org/0000-0002-4104-6529 (não autenticado)

DOI:

https://doi.org/10.5935/jetia.v9i43.904

Resumo

O câncer de ovário continua sendo uma das principais causas de mortalidade relacionada ao câncer entre as mulheres em todo o mundo. Os métodos diagnósticos tradicionais muitas vezes carecem da precisão necessária para a detecção precoce e classificação precisa dos subtipos. Neste estudo, abordamos o desafio de automatizar o diagnóstico de câncer de ovário, introduzindo Modelos Baseados em Atenção (ABMs) em combinação com Redes Neurais Convolucionais 3D (CNNs). Nossa pesquisa busca aumentar a precisão e a eficiência do diagnóstico do câncer de ovário, particularmente na distinção entre os subtipos seroso, mucinoso e endometrioide. As abordagens convencionais de diagnóstico são limitadas pela sua dependência da interpretação manual de imagens médicas e não conseguem explorar plenamente a rica informação presente nos exames de ressonância magnética. O trabalho proposto aproveita os ABMs para focar dinamicamente em regiões críticas em exames de ressonância magnética, permitindo extração aprimorada de recursos e maior precisão de classificação. Demonstramos nossa abordagem em um conjunto de dados bem selecionado, OvaCancerMRI-2023, mostrando o potencial para diagnóstico preciso e automatizado. Resultados experimentais indicam desempenho superior na classificação de subtipos de câncer em comparação aos métodos tradicionais, com precisão de 94% e pontuação F1 de 0,92. Nossas descobertas ressaltam o potencial dos ABMs e das CNNs 3D em revolucionar o diagnóstico do câncer de ovário, abrindo caminho para uma intervenção precoce e estratégias de tratamento mais eficazes. Em conclusão, esta investigação marca um avanço significativo no domínio do diagnóstico do cancro do ovário, oferecendo um caminho promissor para melhorar os resultados dos pacientes e reduzir o fardo desta doença devastadora. A integração de ABMs e CNNs 3D possui um potencial substancial para melhorar a precisão e a eficiência do diagnóstico do câncer de ovário, particularmente na subtipagem, e pode contribuir para uma intervenção precoce e melhor atendimento ao paciente.

Downloads

Os dados de download ainda não estão disponíveis.

Referências

Schwartz, David, Travis W. Sawyer, Noah Thurston, Jennifer Barton, and Gregory Ditzler. "Ovarian cancer detection using optical coherence tomography and convolutional neural networks." Neural Computing and Applications 34, no. 11 (2022): 8977-8987.

Zhang, Zheng, and Yibo Han. "Detection of ovarian tumors in obstetric ultrasound imaging using logistic regression classifier with an advanced machine learning approach." IEEE Access 8 (2020): 44999-45008.

Sadeghi, Mohammad Hossein, Mehrosadat Alavi, and Francesco Giammarile. "The OCDA-Net: a 3D convolutional neural network-based system for classification and staging of ovarian cancer patients using FDG PET/CT examinations." (2023).

Avesani, Giacomo, Huong Elena Tran, Giulio Cammarata, Francesca Botta, Sara Raimondi, Luca Russo, Salvatore Persiani et al. "CT-based radiomics and deep learning for BRCA mutation and progression-free survival prediction in ovarian cancer using a multicentric dataset." Cancers 14, no. 11 (2022): 2739.

Butala, Anish A., Roshal R. Patel, Shwetha Manjunath, Nawar A. Latif, Ashley F. Haggerty, Joshua A. Jones, and Neil K. Taunk. "Palliative radiation therapy for metastatic, persistent, or recurrent epithelial ovarian cancer: efficacy in the era of modern technology and targeted agents." Advances in Radiation Oncology 6, no. 1 (2021): 100624.

Ziyambe, Blessed, Abid Yahya, Tawanda Mushiri, Muhammad Usman Tariq, Qaisar Abbas, Muhammad Babar, Mubarak Albathan, Muhammad Asim, Ayyaz Hussain, and Sohail Jabbar. "A Deep Learning Framework for the Prediction and Diagnosis of Ovarian Cancer in Pre-and Post-Menopausal Women." Diagnostics 13, no. 10 (2023): 1703.

Saida, Tsukasa, Kensaku Mori, Sodai Hoshiai, Masafumi Sakai, Aiko Urushibara, Toshitaka Ishiguro, Manabu Minami, Toyomi Satoh, and Takahito Nakajima. "Diagnosing ovarian cancer on MRI: a preliminary study comparing deep learning and radiologist assessments." Cancers 14, no. 4 (2022): 987.

Wang, Guangxing, Huiling Zhan, Tianyi Luo, Bingzi Kang, Xiaolu Li, Gangqin Xi, Zhiyi Liu, and Shuangmu Zhuo. "Automated ovarian cancer identification using end-to-end deep learning and second harmonic generation imaging." IEEE Journal of Selected Topics in Quantum Electronics 29, no. 4: Biophotonics (2022): 1-9.

Saba, Tanzila. "Recent advancement in cancer detection using machine learning: Systematic survey of decades, comparisons and challenges." Journal of Infection and Public Health 13, no. 9 (2020): 1274-1289.

Jenefa et al., "Real-Time Rail Safety: A Deep Convolutional Neural Network Approach for Obstacle Detection on Tracks," 2023 4th International Conference on Signal Processing and Communication (ICSPC), Coimbatore, India, 2023, pp. 101-105, doi: 10.1109/ICSPC57692.2023.10125284.

Jenefa, A. Khan, E. A. Mathew, J. S. Dani, G. Olivia and T. S. Shivani, "Enhancing Human Behaviour Analysis through Multi-Embedded Learning for Emotion Recognition in Images," 2023 7th International Conference on Intelligent Computing and Control Systems (ICICCS), Madurai, India, 2023, pp. 331-336, doi: 10.1109/ICICCS56967.2023.10142747.

E. Regi, J. A, S. V. E. Sonia, E. Naveen, L. A and V. K, "Multidisciplinary Approach to Physical Rehabilitation on Left and Right Ventricular Function in Heart Failure Patients," 2023 International Conference on Circuit Power and Computing Technologies (ICCPCT), Kollam, India, 2023, pp. 823-829, doi: 10.1109/ICCPCT58313.2023.10245112.

E. Regi, J. A and E. Naveen, "Liquid Biopsy for Non-Invasive Monitoring of Tumour Evolution and Response to Therapy," 2023 International Conference on Circuit Power and Computing Technologies (ICCPCT), Kollam, India, 2023, pp. 815-822, doi: 10.1109/ICCPCT58313.2023.10246001.

Xiao, Yinan, Meiyu Bi, Hongyan Guo, and Mo Li. "Multi-omics approaches for biomarker discovery in early ovarian cancer diagnosis." EBioMedicine 79 (2022).

Zhang, Ruiqian, Michelle KY Siu, Hextan YS Ngan, and Karen KL Chan. "Molecular biomarkers for the early detection of ovarian cancer." International Journal of Molecular Sciences 23, no. 19 (2022): 12041.

Yang, Yuqi, Qiong Huang, Zuoxiu Xiao, Min Liu, Yan Zhu, Qiaohui Chen, Yumei Li, and Kelong Ai. "Nanomaterial-based biosensor developing as a route toward in vitro diagnosis of early ovarian cancer." Materials Today Bio 13 (2022): 100218.

Gahlawat, Aoife Ward, Tania Witte, Lisa Haarhuis, and Sarah Schott. "A novel circulating miRNA panel for non-invasive ovarian cancer diagnosis and prognosis." British Journal of Cancer 127, no. 8 (2022): 1550-1556.

Brewer, Hannah R., Yasemin Hirst, Marc Chadeau-Hyam, Eric Johnson, Sudha Sundar, and James M. Flanagan. "Association between purchase of over-the-counter medications and ovarian cancer diagnosis in the cancer loyalty card study (CLOCS): observational case-control study." JMIR public health and surveillance 9, no. 1 (2023): e41762.

Gao, Yue, Shaoqing Zeng, Xiaoyan Xu, Huayi Li, Shuzhong Yao, Kun Song, Xiao Li et al. "Deep learning-enabled pelvic ultrasound images for accurate diagnosis of ovarian cancer in China: a retrospective, multicentre, diagnostic study." The Lancet Digital Health 4, no. 3 (2022): e179-e187.

Chen, Dongfei, Yanfang Wu, Richard D. Tilley, and J. Justin Gooding. "Rapid and ultrasensitive electrochemical detection of DNA methylation for ovarian cancer diagnosis." Biosensors and Bioelectronics 206 (2022): 114126.

Yesilkaya, Bartu, Matjaž Perc, and Yalcin Isler. "Manifold learning methods for the diagnosis of ovarian cancer." Journal of Computational Science 63 (2022): 101775.

Sengupta, Duhita, Sk Nishan Ali, Aditya Bhattacharya, Joy Mustafi, Asima Mukhopadhyay, and Kaushik Sengupta. "A deep hybrid learning pipeline for accurate diagnosis of ovarian cancer based on nuclear morphology." PloS one 17, no. 1 (2022): e0261181.

Zhu, Jie Wei, Parsa Charkhchi, and Mohammad R. Akbari. "Potential clinical utility of liquid biopsies in ovarian cancer." Molecular Cancer 21, no. 1 (2022): 114.

Chudecka-Głaz, Anita, Aleksandra Strojna, Kaja Michalczyk, Sylwia Wieder-Huszla, Krzysztof Safranow, Edyta Skwirczyńska, and Anna Jurczak. "Evaluation of He4 Use in the Diagnosis of Ovarian Cancer: First and Second Recurrence, and an Analysis of HE4 Concentration during Second-and Third-Line Chemotherapy." Diagnostics 13, no. 3 (2023): 452.

Huang, Weitong, Hanna Suominen, Tommy Liu, Gregory Rice, Carlos Salomon, and Amanda S. Barnard. "Explainable discovery of disease biomarkers: The case of ovarian cancer to illustrate the best practice in machine learning and Shapley analysis." Journal of Biomedical Informatics 141 (2023): 104365.