THE IMPORTANCE OF ENDOSCOPY AND IMAGING METHODS FOR THE DIAGNOSIS OF COELIAC DISEASE: A SYSTEMATIC REVIEW/A IMPORTANCIA DA ENDOSCOPIA E DOS METODOS DE IMAGEM PARA DIAGNOSTICO DA DOENCA CELIACA: UMA REVISAO SISTEMATICA/LA IMPORTANCIA DE LA ENDOSCOPIA Y LOS METODOS DE IMAGEN PARA EL DIAGNOSTICO DE LA ENFERMEDAD CELIACA: UNA REVISION SISTEMATICA

INTRODUCTION: Celiac disease is an autoimmune component enteropathy that has gluten as a precipitating factor. Diagnosis is established by the association of laboratory and clinical findings, because there is no specific test to firm it, so imaging can help in this aspect. Insistent mucosal inflammation alters the conformation of intestinal villi causing lesions, which can be seen in upper endoscopy (ADS) and evaluated by histopathology. In this sense, it is necessary to evaluate its real importance in the diagnostic context, since ADS, as well as other imaging tests, can contribute to advances in the management of Celiac Disease. OBJECTIVES: To analyze the literature about the use of endoscopy and imaging methods for the diagnosis of celiac disease. METHODOLOGY: This is a systematic review of the specialized literature, in the PubMed database, with the descriptors: "Celiac Disease" AND "Endoscopy" AND "Diagnostic Imaging", in the last 5 years. Seven scientific articles were selected. Only articles with free full text and performed on humans were included and those that did not fit the proposed objectives were excluded. RESULTS: Celiac Disease (CD) is an increasingly diagnosed condition, requiring diagnostic accuracy for adequate treatment and favorable prognosis. In this paper, we evaluated seven studies that explored the role of endoscopy and imaging methods in the diagnosis of CD. Two studies investigated the FICE (Flexible Spectral Imaging Color Enhancement) technique and its ability to detect classical changes of the celiac mucosa. The results showed that FICE significantly increased the detection of villus changes compared to the common endoscope, although it remains subject to subjective and operator-dependent analysis. Three studies addressed the use of neural networks, deep learning and machine learning to improve the identification and diagnosis of diseases of the gastrointestinal tract, including CD. Promising results have shown that these computational models provide in-place second opinions for endoscopists, improving diagnostic accuracy. Another effective methodology was the endoscopic analysis of crypts and villi, which allowed a more histological evaluation of CD, influencing prognosis and short and long-term treatment behaviors. The three groups of methodologies studied showed positive results, representing promising practices for a more accurate diagnosis of CD. These advances have the potential to reduce false positives and improve the a