INTELIGENCIA ARTIFICIAL: CARACTERIZACIÓN, SEGMENTACIÓN DE NÓDULOS PULMONARES POR TOMOGRAFÍA DE ALTA RESOLUCIÓN

• Miguel Ángel Hernández Neira División de Estudios para Graduados de la Facultad de Medicina de la Universidad del Zulia.
• Edunice Mora López División de Estudios para Graduados de la Facultad de Medicina de la Universidad del Zulia.
• Juan Pablo Monroy División de Estudios para Graduados de la Facultad de Medicina de la Universidad del Zulia.
• Edilia Elena Noguera Hernández

DOI: https://doi.org/10.5281/zenodo.17981140

Palabras clave: Nódulos pulmonares, inteligencia artificial, tomografía de tórax

Resumen

Objetivo: evaluar la inteligencia artificial (IA) mediante tomografía de alta resolución en la segmentación y caracterización de nódulos pulmonares (NP) en pacientes referidos al Hospital Clínico de Maracaibo. Metodología: investigación descriptiva, transversal, prospectiva; muestra de 19 pacientes que requerían tomografía de tórax, empleando multidetector GE. Resultados: caracterización de NP mediante IA: respecto al tamaño de nódulos el promedio fue de 1,6 cm; con mayor número de casos (6) entre 0,5 a 1cm (31,6%), múltiples 10 (52,6%). Referente al tipo: sólidos de contorno regular 14 (73,7%), predominaron en lóbulos superiores, Derecho 8 (42,1%) e Izquierdo 5 (26,3%), sin correlación entre operadores con IA. Tipo de nódulos mediante IA: fueron sólidos (n=13), coincidiendo 7 con el primer operador y 12 con el segundo; igualmente el contorno regular de 13 casos detectados por la IA, 12 coincidieron con el segundo operador, arrojando concordancia de 92,3%. Segmentación mediante IA: fueron del Lóbulo Superior D (n=8), concordando 4 con el primer observador y 5 con el segundo, no hubo concordancia con 12 casos del primer operador y con 8 casos del segundo operador. Conclusión: la IA sigue siendo operador dependiente, requiriendo estudios de nuevos algoritmos basados en el aprendizaje profundo y nuevos paradigmas de lectura.

Citas

Aberle DR, Adams AM, Berg CD, Black WC, Clapp JD, Fagerstrom RM, et al. (2011) Reduced Lung-Cancer Mortality With Low-Dose Computed Tomographic Screening. N Engl J Med; 365(5):395–409. doi: 10.1056/NEJMoa1102873

Armato SG, 3rd, McLennan G, Bidaut L. (2011) The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): a completed reference database of lung nodules on CT scans. Med Phys 2011;38:915-31.10.1118/1.3528204

Ather S, Kadir T, Gleeson F. (2020) Artificial Intelligence and Radiomics in Pulmonary Nodule Management: Current Status and Future Applications. Clin Radiol 75(1):13–9. doi: 10.1016/j.crad.2019.04.017

Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. (2018) Global Cancer Statistics 2018: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin (2018) 1168(6):394–424. doi: 10.3322/caac.21492

Chang, P. (2018) What to make of the increased hype surrounding AI in radiology. Radiology Business. Disponible en: https://www.radiologybusiness.com/sponsored/9667/topics/ imaging- informa-tics/rsna-2018-hype-ai-radiology-

Chartrand G, Cheng PM, Vorontsov E. (2017) learning: a primer for radiologists. Radiographics.37(7):2113–2131

Chung K, Mets OM, Gerke PK. (2018) Brock malignancy risk calculator for pulmonary nodules: validation outside a lung cancer screening population. Thorax; 73 (1):857-863.

Ciompi F, Chung K, van Riel S. (2017) Towards automatic pulmonary nodule management in lung cancer screening with deep learning. Sci Rep 2017;7:46479. 10.1038/srep46479.

Devaraj A, van Ginneken B, Nair A. (2017) Use of Volumetry for Lung Nodule Management: Theory and Practice. Radiology;284:630-44. 10.1148/radiol.2017151022

Gould MK, Donington J, Lynch WR. (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):e93S–e120S.

Jacobs C, Scholten E, Schreuder A. (2019) An observer study comparing radiologists with the prize-winning lung cancer detection algorithms from the 2017. Kaggle Data Science Bowl. Annual Meeting of the Radiological Society of North America, 2019.

Krizhevsky A, Sutskever I, Hinton G E. (2012) “Imagenet classification with deep convolutional neural networks. Disponible en: https://dl.acm.org/doi/10. 5555/2999134.2999257.

Lachance, Chantelle C., Walter, Melissa. (2020) Artificial Intelligence for Classification of Lung Nodules: A Review of Clinical Utility, Diagnostic Accuracy, Cost-Effectiveness, and Guidelines. Canadian Agency for Drugs and Technologies in Health. Disponible en: https://pubmed.ncbi.nlm.nih.gov/33074628/

Li D, Mikela Vilmun B, Frederik Carlsen J, et al. (2019) The Performance of Deep Learning Algorithms on Automatic Pulmonary Nodule Detection and Classification Tested on Different Datasets That Are Not Derived from LIDC-IDRI: A Systematic Review. Diagnostics (Basel); 9:207.10.3390/diagnostics9040207

MacMahon H, Naidich DP, Goo JM. (2017) Guidelines for management of incidental pulmonary nodules detected on CT images: from the Fleischner Society Radiol.; 284(1):228–243.

Moyer VA, Moyer VA, LeFevre ML, Siu AL, Peters JJ, Baumann LC. (2014) Screening for Lung Cancer: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med; 160(5):330–8. Disponible en: doi: 10.7326/M13-2771

McWilliams A, Tammemagi MC, Mayo JR, et al. (2013) Probability of cancer in pulmonary nodules detected on first screening CT. N Engl J Med;369:910-9. 10.1056/NEJMoa1214726

Patel, V.K. (2013) A Practical Algorithmic Approach to the Diagnosis and Management of Solitary Pulmonary Nodules Diagnosis of Pulmonary Nodules Part 2: Pretest Probability and Algorithm. Chest, 143, pp. 840 http://dx.doi.org/10.1378/chest.12-1487

Sociedad Venezolana Anticancerosa (2019). Boletín de los Pronósticos de la Mortalidad e Incidencia del Cáncer en Venezuela 2019. 3a Edición - Centro de Estadística y Matemática Aplicada de la Universidad Simón Bolívar. Disponible en: www.cancervenezuela.org

Pinsky PF, Gierada DS, Nath PH. ( 2013) National lung screening trial: variability in nodule detection rates in chest CT studies. Radiology; 268:865-73.

Rubin GD, Lyo JK, Paik DS. (2005) Pulmonary nodules on multi-detector row CT scans: performance comparison of radiologists and computer-aided detection. Radiology 2005; 234:274-83. 10.1148/radiol.2341040589

Setio AAA, Traverso A, de Bel T. (2017) Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: The LUNA16 challenge. Med Image Anal 2017;42:1-13.

Tammemagi M, Ritchie AJ, Atkar-Khattra S. (2019) Predicting Malignancy Risk of Screen-Detected Lung Nodules-Mean Diameter or Volume. J Thorac Oncol; 14: 203-211.

Van Riel SJ, Ciompi F, Winkler Wille M. (2017) Malignancy risk estimation of pulmonary nodules in screening CTs: Comparison between a computer model and human observers. PLoS One 2017;12:e0185032.

Wu W, Hu H, Gong J, Li X, Huang G, Nie S. (2019) Malignant-benign classification of pulmonary nodules based on random forest aided by clustering analysis. Phys Med Biol; 64(3):035017.

Yasaka K, Katsura M, Hanaoka S, Sato J, Ohtomo K. (2016) High-Resolution CT With New Model-Based Iterative Reconstruction With Resolution Preference Algorithm in Evaluations of Lung Nodules: Comparison With Conventional Model-Based Iterative Reconstruction and Adaptive Statistical Iterative Reconstruction. Eur J Radiol (2016)85(3):599–606. doi: 10.1016/j.ejrad.2016.01.001

Zaharchuk G, Gong E, M. Wintermark, M. (2018) Deep Learning in Neuroradiology. Am. J. Neuro-radioL, October 2018, 39 (10) 1776-1784. https://doi.org/10.3174/ajnr.A5543

Zhang C, Sun X, Dang K, Li K, Guo XW, Chang J, et al. (2019) Toward an Expert Level of Lung Cancer Detection and Classification Using a Deep Convolutional Neural Network. Oncologist (2019) 24(9):1159–65. doi: 10.1634/theoncologist.2018-0908.