ABSTRACT
RESUMEN: La clasificación de los Tumores Primarios del Sistema Nervioso Central (SNC) tiene su origen en la descripción morfológica, cuyo análisis histopatológico ha permitido identificar la línea celular involucrada en estos tumores y obtener el reconocimiento de ciertas características de estas lesiones y su evolución clínica. El estudio molecular ha venido a complementar el diagnóstico inicial permitiendo reconocer entidades que no son distinguibles de otra manera y que han variado los conceptos y definiciones de varias entidades patológicas que modifican el horizonte visible de estas enfermedades. El papel de las imágenes de Resonancia Magnética (RM) en el manejo de los tumores intraaxiales se puede dividir ampliamente en el diagnóstico y la clasificación de los tumores, la planificación del tratamiento y el tratamiento posterior. El presente artículo resume la evidencia epidemiológica relacionada en la clasificación de los tumores primarios del SNC con marcadores moleculares y biomarcadores de imágenes de RM, apuntando a la importancia del uso de la investigación clínica con el manejo terapéutico.
SUMMARY: The classification of primary tumors of the Central Nervous System (CNS) has its origin in the morphological description whose histopathological analysis has allowed to identify the cell line involved in these tumors and obtain the recognition of certain characteristics of these lesions and their clinical evolution. The molecular study has come to complement the initial diagnosis allowing to recognize entities that are not distinguishable in another way and that have varied the concepts and definitions of various pathological entities modifying the visible horizon of these diseases. The role of Magnetic Resonance (MR) images in the management of intraaxial tumors can be broadly divided into the diagnosis and classification of tumors, treatment planning and subsequent treatment. The present article summarizes the epidemiologic evidence related to the classification of primary tumors of the CNS with molecular markers and MR imaging biomarkers.
Subject(s)
Humans , Magnetic Resonance Imaging , Central Nervous System Neoplasms/classification , Central Nervous System Neoplasms/diagnostic imaging , Brain Neoplasms/diagnostic imaging , BiomarkersABSTRACT
PURPOSE: Medical image analysis using deep neural networks has been actively studied. For accurate training of deep neural networks, the learning data should be sufficient and have good quality and generalized characteristics. However, in medical images, it is difficult to acquire sufficient patient data because of the difficulty of patient recruitment, the burden of annotation of lesions by experts, and the invasion of patients' privacy. In comparison, the medical images of healthy volunteers can be easily acquired. To resolve this data bias problem, the proposed method synthesizes brain tumor images from normal brain images. METHODS: Our method can synthesize a huge number of brain tumor multicontrast MR images from numerous healthy brain multicontrast MR images and various concentric circles. Because tumors have complex characteristics, the proposed method simplifies them into concentric circles that are easily controllable. Then, it converts the concentric circles into various realistic tumor masks through deep neural networks. The tumor masks are used to synthesize realistic brain tumor images from normal brain images. RESULTS: We performed a qualitative and quantitative analysis to assess the usefulness of augmented data from the proposed method. Data augmentation by the proposed method provided significant improvements to tumor segmentation compared with other GAN-based methods. Intuitive experimental results are available online at https://github.com/KSH0660/BrainTumor. CONCLUSIONS: The proposed method can control the grade tumor masks by the concentric circles, and synthesize realistic brain tumor multicontrast MR images. In terms of data augmentation, the proposed method can successfully synthesize brain tumor images that can be used to train tumor segmentation networks or other deep neural networks.
