• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.17.2021.tde-08092021-115434
Document
Author
Full name
Jéssica Deponti Gobbi
Institute/School/College
Knowledge Area
Date of Defense
Published
Ribeirão Preto, 2021
Supervisor
Committee
Jorge, Rodrigo (President)
Abud, Daniel Giansante
Vianna, Raul Nunes Galvarro
Title in Portuguese
Eficácia da fotografia de retina baseada em smartphone no rastreamento e diagnóstico precoce de retinopatia diabética
Keywords in Portuguese
Diagnóstico precoce
Retina
Retinopatia diabética
Técnicas de diagnóstico oftalmológico
Tecnologia de baixo custo
Telemedicina
Abstract in Portuguese
Objetivo: Avaliar a eficácia da imagem e fotografia de retina obtida por um dispositivo baseado em Smartphone no rastreamento e diagnóstico precoce de Retinopatia diabética (RD) em pacientes diabéticos. Métodos: Foi realizado um estudo prospectivo aberto com noventa e nove pacientes diabéticos (194 olhos). Os pacientes inclusos foram recrutados e submetidos a exame oftalmológico para rastreio de RD. Ao exame foi capturado um vídeo do fundo de olho com o uso do dispositivo baseado em Smartphone e uma fotografia com a tecnologia padrão ouro através de retinógrafo que será utilizada como referência na avaliação dos resultados. Ambas formas de documentação foram avaliadas por dois especialistas mascarados, segundo a acurácia com relação a presença de retinopatia diabética ou não e também com relação a sua gravidade. Resultados: Em relação à presença ou ausência da RD, a concordância entre as imagens do dispositivo baseado em Smartphone e da retinografia, foi de: 84,07%, com Kappa de: 0.67; Sensibilidade: 71,0% (IC: 65,05% - 78,16%); Especificidade: 94,06% (IC: 90,63% - 97,49%); Acurácia: 84,07%; Valor preditivo positivo: 90,62%; Valor preditivo negativo: 80,51%. Quanto à classificação entre retinopatia diabética proliferativa e retinopatia diabética não proliferativa a concordância entre as imagens do dispositivo baseado em Smarphone e as imagens da retinografia foi de: 90,00%; com Kappa de: 0.78; Sensibilidade: 86,96%; (IC: 79,07% - 94,85%); Especificidade: 91,49% (IC: 84,95% - 98,03%); Acurácia: 90,00%; Valor preditivo positivo: 83,33%; Valor preditivo negativo: 93,48%. Em relação ao grau de classificação da RD a concordância entre as imagens do dispositivo baseado em Smarphone e as imagens da retinografia foi de: 69,23% com Kappa de: 0.52. Considerando a presença ou ausência de exsudatos duros maculares, a concordância entre as imagens do dispositivo baseado em Smarphone e as imagens da retinografia foi de: 84,07%, com Kappa de: 0.67; Sensibilidade: 71,60% (intervalo de confiança - IC: 65,05% - 78,16%); Especificidade: 94,06% (intervalo de confiança - IC: 90,63% - 97,49%); Acurácia: 84,07%; Valor preditivo positivo: 90,62%; Valor preditivo negativo: 80,51%. Conclusão: O dispositivo baseado em Smartphone mostrou uma concordância considerável na detecção da RD, tornando-o uma ferramenta promissora no rastreamento e diagnóstico precoce de RD em pacientes diabéticos com a vantagem da portabilidade, fácil manuseio, baixo custo e também com a facilidade de carregar as imagens de fundo de olho obtidas pelo Smartphone diretamente do aparelho para especialistas de maneira remota.
Title in English
Efficacy of smartphone-based retinal photography in screening and early diagnosis of diabetic retinopathy
Keywords in English
Diabetic retinopathy
Early diagnosis
Low cost technology
Ophthalmological diagnosis techniques
Retina
Telemedicine
Abstract in English
Purpose: To evaluate the efficacy of retinal image and photography obtained by a Smartphone-based device in screening and early diagnosing Diabetic Retinopathy (DR) in diabetic patients. Methods: An open prospective study was carried out in ninety-nine diabetic patients (194 eyes). The included patients were recruited and submitted to an ophthalmological examination for screening for DR. During the examination, images of the fundus were captured by using both a smartphone-based device, which produced a video to be analyzed, and a gold-standard tabletop retinography camera system, which produced photographs that were later used in the evaluation of results as reference for comparison. Both forms of documentation were evaluated by two distinct blinded specialists, according to the accuracy of detecting the presence or absence of diabetic retinopathy and also of detecting its severity. Results: Concerning the presence or absence of DR, the agreement rate between images of the smartphone-based device and the regular (tabletop) retinography system was: 84.07%, with a Kappa of: 0.67; Sensitivity: 71.0% (CI: 65.05% - 78.16%); Specificity: 94.06% (CI: 90.63% - 97.49%); Accuracy: 84.07%; Positive predictive value: 90.62%; Negative predictive value: 80.51%. As for the classification between proliferative diabetic retinopathy and non-proliferative diabetic retinopathy, agreement between the images of the Smartphone-based device and those of the regular retinography system was: 90.00%; with Kappa of: 0.78; Sensitivity: 86.96%; (CI: 79.07% - 94.85%); Specificity: 91.49% (CI: 84.95% - 98.03%); Accuracy: 90.00%; Positive predictive value: 83.33%; Negative predictive value: 93.48%. Regarding the degree of classification of DR, the agreement rate between the images of theSmartphone-based device and those of the regular retinography system was: 69.23% with Kappa of: 0.52. As relating to the presence or absence of hard macular exudates, the agreement between the images of the Smartphone-based device and those of the regular retinography system was: 84.07%, with Kappa of: 0.67; Sensitivity: 71.60% (CI: 65.05% - 78.16%); Specificity: 94.06% (CI: 90.63% - 97.49%); Accuracy: 84.07%; Positive predictive value: 90.62%; Negative predictive value: 80.51%. Conclusion: The Smartphone-based device showed an enhanced accuracy in the detection of DR, making it a promising tool in the screening and early diagnosis of DR in diabetic patients with the advantage of portability, easy handling, low cost and also the additional convenience represented by the possibility of directly sharing the images obtained through the Smarphone-based device with remotely located specialists.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
JESSICADEPONTIGOBBI.pdf (807.89 Kbytes)
Publishing Date
2021-10-04
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.