The use of artificial intelligence in MRI diagnostics of rectal cancer

AIM: development of medical decision support systems for MRI diagnostics of rectal cancer: localization and segmentation of the primary tumor.

PATIENTS AND METHODS: the research included 450 MRI studies of patients with rectal cancer and 450 MRI studies of patients without a tumor lesion of the rectum. All patients with tumors of rectum had histological verification of the malignant process. Data were collected in T2Wcoronal and axial projections (MRI Philips Achieva 1.5 T). Object marking was carried out only for T2W projections, where the area of interest was segmented - rectum, sigmoid colon and tumor. The ITK-Snap program was used to label MRI images. The validated studies and labeling were used to create a machine learning model that demonstrates the capability of the dataset to build medical decision support systems. SegResNet, TransUnet, 3D Unet neural networks were used to create a basic artificial intelligence model. The data set of patients and the direct marking of MRI studies were carried out by doctors of Ryzhikh National Medical Research Center of Coloproctology. The development of the artificial intelligence model, markup validation was carried out by employees of JSC "National Center of Service Integration" and LLC "Medical Screening Systems".

RESULTS: dice similarity coefficient (DSC) of various neural networks were: TransUnet - 0.33, SegResNet - 0.50, 3D Unet - 0.42. The diagnostic efficiency of the SegResNet neural network in detecting rectal tumors with the addition of negative examples and post-processing was: accuracy 77.0%; sensitivity 98.1%; specificity 45.1%; positive predictive value 72.9%; negative predictive value of 94.1%. At this stage, AI has a fairly high sensitivity and accuracy, which indicates a high diagnostic efficiency in terms of visualizing the primary tumor and determining localization in the rectum. However, the specificity of the method is still at an unsatisfactory level (45.1%), which indicates a high percentage of false positive results in healthy patients and does not allow the model to be used as a screening method at this stage of development.

CONCLUSION: the collected dataset of MRI studies and their markup made it possible to obtain an AI model that allows solving the problem of segmenting a rectal tumor and determining its localization. The next stage in the development of AI is to improve its specificity, expand the analyzed parameters, such as the depth of tumor invasion, visualization of metastatic lymph nodes and the status of the resection margin. To further develop the model metric and improve its diagnostic capabilities, we should experiment with training parameters and increase the dataset.

1. Kaprin A.D., Starinsky V.V., Shahzadova A.O. Malignant neoplasms in Russia in 2019 (morbidity and mortality). Moscow: MNIOI. P.A. Herzen (branch FGBU “MICR” of Minzdrav of Russia). 2020; p. 252. (in Russ.).

2. Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015; 136: E359–E386. DOI:10.1002/ijc.29210

3. Kaprin A.D., Starinsky V.V., Shakhzadova A.O. The state of oncological care for the population of Russia in 2019. Moscow: MNIOI. P.A. Herzen (branch FGBU “MICR” of Minzdrav of Russia). 2020; p. 239 (in Russ.)>

4. Wang H, Fu C. Value of preoperative accurate staging of rectal cancer and the effect on the treatment strategy choice (in Chinese). Chin J Pract Surg. 2014;34:37–40. DOI:10.3969/j.issn.1674-9316.2016.02.058

5. Abraha I, Aristei C, Palumbo I, et al. Preoperative radiotherapy and curative surgery for the management of localised rectal carcinoma. Cochrane Database Syst Rev. 2018;10(10):CD002102. DOI:10.1002/14651858.CD002102.pub3

6. Jhaveri KS, Sadaf A. Role of MRI for staging of rectal cancer. Expert Rev Anticancer Ther. 2009;9(4):469-481. DOI:10.1586/era.09.13

7. Beets-Tan R, Lambregts D, Maas M, et al. Magnetic resonance imaging for clinical management of rectal cancer: Updated recommendations from the 2016 European Society of Gastrointestinal and Abdominal Radiology (ESGAR) consensus meeting [published correction appears in Eur Radiol. Eur Radiol. 2018;28(4):1465-1475. DOI:10.1007/s00330-017-5026-2

8. Xiao Y, Liu S. Artificial intelligence will change the future of imaging medicine (in Chinese). J Technol Finance. 2018;10:11–15. DOI:10.3969/j.issn.2096-4935.2018.10.006

9. Wang Y, He X, Nie H, et al. Application of artificial intelligence to the diagnosis and therapy of colorectal cancer. Am J Cancer Res. 2020;10(11):3575–3598.

10. Perone C, Cohen-Adad J. Promises and limitations of deep learning for medical image segmentation. J Med Artif Intel. 2019;2:1-2. DOI:10.21037/jmai.2019.01.01

11. Ding L, Liu G, Zhao B. et al. Artificial intelligence system of faster region-based convolutional neural network surpassing senior radiologists in evaluation of metastatic lymph nodes of rectal cancer. Chin Med J. 2019;132(4):379–87. DOI:10.1097/CM9.0000000000000095

12. Harangi B. Skin lesion classification with ensembles of deep convolutional neural networks. J Biomed Inform. 2018;86:25–32. DOI:10.1016/j.jbi.2018.08.006

13. Ehteshami Bejnordi B, Veta M, Johannes van Diest P, et al. Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women With Breast Cancer. JAMA. 2017;318(22): 2199-2210. DOI:10.1001/jama.2017.14585

14. Yushkevich PA, Piven J, Hazlett HC, et al. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage. 2006;31(3):1116-1128. DOI:10.1016/j.neuroimage.2006.01.015

15. Baid U, Chodasara S, Mohan S, et al. The rsna-asnr-miccai brats 2021 benchmark on brain tumor segmentation and radiogenomic classification. arXiv. 2021;2107.02314.

16. Magadza T, Viriri S. Deep Learning for Brain Tumor Segmentation: A Survey of State-of-the-Art. J Imaging. 2021;7(2):19. DOI:10.3390/jimaging7020019

17. Balakrishnan G, Zhao А, Sabuncu M. el al. VoxelMorph: A Learning Framework for Deformable Medical Image Registration. IEEE Transactions on Medical Imaging. 2019;38(8):1788-1800 DOI:10.1109/TMI.2019.2897538

18. Taylor F, Quirke P, Heald RJ, et al. Preoperative high-resolution magnetic resonance imaging can identify good prognosis stage I, II, and III rectal cancer best managed by surgery alone: a prospective, multicenter, European study. Ann Surg. 2011;253(4):711-719. DOI:10.1097/SLA.0b013e31820b8d52.

19. Chernyshov S.V., Khomyakov E.A., Sinitsyn R.K. et al. Latent adenocarcinoma in adenomas. Possibilities of instrumental identification. Koloproktologia. 2021; 20(2):10-17. (in Russ.). DOI:10.33878/2073-7556-2021-20-2-10-16.

20. Thrall J, Li X, Li Q. et al. Artificial Intelligence and Machine Learning in Radiology: Opportunities, Challenges, Pitfalls, and Criteria for Success. J Am Coll Radiol. 2018;15:504-508. DOI:10.1016/j.jacr.2017.12.026

21. Wu QY, Liu SL, Sun P. et al. Establishment and clinical application value of an automatic diagnosis platform for rectal cancer T-staging based on a deep neural network. Chin Med J (Engl). 2021;134(7):821-828. DOI:10.1097/CM9.0000000000001401

22. Lu Y, Yu Q, GaoY. et al. Identification of Metastatic Lymph Nodes in MR Imaging with Faster Region-Based Convolutional Neural Networks. Cancer Res. 2018;78(17):5135-5143. DOI:10.1158/0008-5472.CAN-18-0494