NOVEL APPROACHES TO DEVELOPMENT OF ARTIFICIAL INTELLIGENCE ALGORITHMS IN THE LUNG CANCER DIAGNOSTICS

The relevance of developing an intelligent automated diagnostic system (IADS) for lung cancer (LC) detection stems from the social significance of this disease and its leading position among all cancer diseases. Theoretically, the use of IADS is possible at a stage of screening as well as at a stage of adjusted diagnosis of LC. The recent approaches to training the IADS do not take into account the clinical and radiological classification as well as peculiarities of the LC clinical forms, which are used by the medical community. This defines difficulties and obstacles of using the available IADS. The authors are of the opinion that the closeness of a developed IADS to the «doctor’s logic» contributes to a better reproducibility and interpretability of the IADS usage results. Most IADS described in the literature have been developed on the basis of neural networks, which have several disadvantages that affect reproducibility when using the system. This paper proposes a composite algorithm using machine learning methods such as Deep Forest and Siamese neural network, which can be regarded as a more efficient approach for dealing with a small amount of training data and optimal from the reproducibility point of view. The open datasets used for training IADS include annotated objects which in some cases are not confirmed morphologically. The paper provides a description of the LIRA dataset developed by using the diagnostic results of St. Petersburg Clinical Research Center of Specialized Types of Medical Care (Oncology), which includes only computed tomograms of patients with the verified diagnosis. The paper considers stages of the machine learning process on the basis of the shape features, of the internal structure features as well as a new developed system of differential diagnosis of LC based on the Siamese neural networks. A new approach to the feature dimension reduction is also presented in the paper, which aims more efficient and faster learning of the system.

1. Forman D., Bray F., Brewster D.H., Gombe Mbalawa C., Kohler B., Piñeros M., Steliarova-Foucher E., Swaminathan R., Ferlay J. Cancer Incidence in Five Continents // IARC Scientific Publications. 2014. Vol. 10, Nо. 164. Р. 10.

2. Lyon, International Agency for Research on Cancer. Available from: http://ci5.iarc.fr. (accessed: 26.02.2017)

3. Карамова Д.А., Кудинова А.И., Толстова А.В., Савельев В.Н. Исследование заболеваемости злокачественными заболеваниями легких // Авиценна. 2018. № 17. С. 44–47. [Karamova D.A., Kudinova A.I., Tolstova A.V., Savel’ev V.N. Issledovanie zabolevaemosti zlokachestvennymi zabolevaniyami legkih. Avicenna, 2018, Nо. 17, рр. 44–47 (In Russ.)].

4. Гомболевский В.А., Барчук А.А., Лайпан А.Ш., Ветшева Н.Н., Владзимирский А.В., Морозов С.П. Организация и эффективность скрининга злокачественных новообразований легких методом низкодозной компьютерной томографии // РадиологияПрактика. 2018. № 1 (67). С. 28–36. [Gombolevskij V.A., Barchuk A.A., Lajpan A.Sh., Vetsheva N.N., Vladzimirskij A.V., Morozov S.P. Organizaciya i ehffektivnost’ skrininga zlokachestvennyh novoobrazovanij legkih metodom nizkodoznoj komp’yuternoj tomografii. Radiologiya-Praktika, 2018, Nо. 1 (67), рр. 28–36 (In Russ.)].

5. Hastie T., Tibshirani R., Friedman J. The Elements of Statistical Learning. Springer, 2014. 739 р.

6. Kostis W.J., Reeves A.P., Yankelevitz D.F., Henschke C.I. Threedimensional segmentation and growth-rate estimation of small pulmonary nodules in helical CT images // IEEE Transactions on Medical Imaging. 2003. Nо. 22 (10). P. 1259–1274.

7. Doi K. Current status and future potential of computer-aided diagnosis in medical imaging // The British Journal of Radiology. 2005. Nо. 78. P. 3–19.

8. Firmino M., Morais A.H., Mendoca R.M., Dantas M.R., Hekis H.R., Valentim R. Computer-aided detection system for lung cancer in computed tomography scans: review and future prospects // Biomedical engineering online. 2014. Nо. 13 (1). P. 41.

9. Rehman M.Z., Javaid M., Shah S.I.A., Gilani S.O., Jamil M., Butt S.I. An appraisal of nodules detection techniques for lung cancer in CT images // Biomedical Signal Processing and Control. 2018. Nо. 41. P. 140–151.

10. Кобцова Т.В., Мелдо А.А., Хейнштейн В.А. Нетипичные находки в диагностике периферического рака легкого // Материалы V юбилейного международного конгресса «Кардиоторакальная радиология». СПб.: Человек и его здоровье, 2018. С. 52–53. [Kobcova T.V., Meldo A.A., Hejnshtejn V.A. Netipichnye nahodki v diagnostike perifericheskogo raka lyogkogo. Materialy V yubilejnogo mezhdunarodnogo kongressa «Kardiotorakal’naya radiologiya». Saint Petersburg: Izdatel’stvo «Chelovek i ego zdorov’e», 2018, рр. 52–53 (In Russ.)].

11. Bramer M. Principles of Data Mining. Springer, 2007. 354 р.

12. Battaglia P.W., Hamrick J.B., Bapst V., Sances-Gonzales A. et al. Relational inductive biases, deep learning, and graph networks // arXive1806.01261. Jul. 2018.

13. Zhou Z.-H., Feng J. Deep forest: Towards an alternative to deep neural networks // Proceedings of the 26th International Joint Conference on Artificial Intelligence (IJCAI’17). Melbourne: Australia, 2017. P. 3553–3559.

14. Zhi-Hua Zhou Ji Feng. Deep Forest // National Science Review. 2018. 8 Oct. https://doi.org/10.1093/nsr/nwy108.

15. Utkin L.V., Ryabinin M.A. A Siamese deep forest // KnowledgeBased Systems. 2018. Vol. 139. P. 13–22.

16. Тегированные результаты компьютерных томографий легких: а.с. 2018620500 Рос. Федерация / Морозов С.П., Кульберг Н.С., Гомболевский В.А. и соавт.; заявитель и правообладатель: ГБУЗ «НПЦМР ДЗМ». № 2018620148; заявл. 06.02.2018; опубл. 28.03.2018, Бюл. № 4. 1 с. [Tegirovannye rezul’taty komp’yuternyh tomografij legkih: a.s. 2018620500 Ros. Federaciya / Morozov S.P., Kul’berg N.S., Gombolevskij V.A. et al.; zayavitel’ i pravoobladatel’: GBUZ «NPCMR DZM». No. 2018620148; zayavl. 06.02.2018; opubl. 28.03.2018, Byul. No. 4. 1 р. (In Russ.)].

17. Имянитов Е.Н. Рак легкого в начале XXl века // Русский медицинский журнал. 2007. № 5. С. 400. [Imyanitov E.N. Rak legkogo v nachale XXI veka. Russkij medicinskij zhurnal, 2007, No. 5, р. 400 (In Russ.)].

18. Имянитов Е.Н. Современные представления о молекулярных мишенях в опухолях легкого // Практическая онкология. 2018. Т. 19, № 2. С. 93–104. [Imyanitov E.N. Sovremennye predstavleniya o molekulyarnyh mishenyah v opuholyah legkogo. Prakticheskaya onkologiya, 2018, Vol. 19, No. 2, рр. 93–104 (In Russ.)].

19. http://www.cancerimagingarchive.net/

20. Armato III S.G., McLennan G. et al. The lung image databas econsortium (LIDC) and image database resource initiative (IDRI): a completed reference database of lung nodules on CT scans // Medical. Physics. 2011. Vol. 38, No. 2b. P. 915–931.

21. Causey J., Zhang J., Ma S., Jiang B., Qualls J., Politte D.G., Prior F., Zhang S., Huang X. Highly accurate model for prediction of lung nodule malignancy with CT scans // arXiv: 1802.01756. Feb. 2018.

22. Моисеенко В.М., Мелдо А.А., Уткин Л.В., Прохоров И.Ю., Рябинин М.А., Богданов А.А. Автоматизированная система обнаружения объемных образований в легких как этап развития искусственного интеллекта в диагностике рака легкого // Лучевая диагностика и терапия. 2018. № 3. С. 62–68. [Moiseenko V.M., Meldo A.A., Utkin L.V., Prohorov I.Yu., Ryabinin M.A., Bogdanov A.A. Avtomatizirovannaya sistema obnaruzheniya ob»emnyh obrazovanij v legkih kak ehtap razvitiya iskusstvennogo intellekta v diagnostike raka legkogo. Luchevaya diagnostika i terapiya, 2018, No. 3, рр. 62–68 (In Russ.)].

23. Прохоров И.Ю., Рябинин М.А., Мелдо А.А., Уткин Л.В. Формирование баз данных с целью машинного обучения в диагностике рака легкого // Конгресс Российского общества рентгенологов и радиологов. Сборник тезисов. СПб., 2018. С. 124–125. [Prohorov I.Yu., Ryabinin M.A., Meldo A.A., Utkin L.V. Formirovanie baz dannyh s cel’yu mashinnogo obucheniya v diagnostike raka legkogo. Kongress Rossijskogo obshchestva rentgenologov i radiologov. Sbornik tezisov. Saint Petersburg, 2018, рр. 124–125 (In Russ.)].

24. Koch G., Zemel R., Salakhutdinov R. Siamese neural networks for one-shot image recognition // Proceedings of the 32nd International Conference on Machine Learning. Lille, France. 2015. Vol. 37. P. 1–8.