Articles producció científica> Medicina i Cirurgia

A New Deep Learning Algorithm with Activation Mapping for Diabetic Retinopathy: Backtesting after 10 Years of Tele-Ophthalmology

  • Identification data

    Identifier: imarina:9281706
    Authors:
    Pareja-Rios, AliciaCeruso, SabatoRomero-Aroca, PedroBonaque-Gonzalez, Sergio
    Abstract:
    We report the development of a deep learning algorithm (AI) to detect signs of diabetic retinopathy (DR) from fundus images. For this, we use a ResNet-50 neural network with a double resolution, the addition of Squeeze-Excitation blocks, pre-trained in ImageNet, and trained for 50 epochs using the Adam optimizer. The AI-based algorithm not only classifies an image as pathological or not but also detects and highlights those signs that allow DR to be identified. For development, we have used a database of about half a million images classified in a real clinical environment by family doctors (FDs), ophthalmologists, or both. The AI was able to detect more than 95% of cases worse than mild DR and had 70% fewer misclassifications of healthy cases than FDs. In addition, the AI was able to detect DR signs in 1258 patients before they were detected by FDs, representing 7.9% of the total number of DR patients detected by the FDs. These results suggest that AI is at least comparable to the evaluation of FDs. We suggest that it may be useful to use signaling tools such as an aid to diagnosis rather than an AI as a stand-alone tool.
  • Others:

    Author, as appears in the article.: Pareja-Rios, Alicia; Ceruso, Sabato; Romero-Aroca, Pedro; Bonaque-Gonzalez, Sergio;
    Department: Medicina i Cirurgia
    URV's Author/s: Romero Aroca, Pedro
    Keywords: Tele-ophthalmology Diabetic retinopathy Deep learning Artificial-intelligence Artificial intelligence
    Abstract: We report the development of a deep learning algorithm (AI) to detect signs of diabetic retinopathy (DR) from fundus images. For this, we use a ResNet-50 neural network with a double resolution, the addition of Squeeze-Excitation blocks, pre-trained in ImageNet, and trained for 50 epochs using the Adam optimizer. The AI-based algorithm not only classifies an image as pathological or not but also detects and highlights those signs that allow DR to be identified. For development, we have used a database of about half a million images classified in a real clinical environment by family doctors (FDs), ophthalmologists, or both. The AI was able to detect more than 95% of cases worse than mild DR and had 70% fewer misclassifications of healthy cases than FDs. In addition, the AI was able to detect DR signs in 1258 patients before they were detected by FDs, representing 7.9% of the total number of DR patients detected by the FDs. These results suggest that AI is at least comparable to the evaluation of FDs. We suggest that it may be useful to use signaling tools such as an aid to diagnosis rather than an AI as a stand-alone tool.
    Thematic Areas: Medicine, general & internal Medicine (miscellaneous) Medicine (all)
    licence for use: https://creativecommons.org/licenses/by/3.0/es/
    Author's mail: pedro.romero@urv.cat
    Author identifier: 0000-0002-7061-8987
    Record's date: 2024-09-07
    Papper version: info:eu-repo/semantics/publishedVersion
    Link to the original source: https://www.mdpi.com/2077-0383/11/17/4945
    Licence document URL: https://repositori.urv.cat/ca/proteccio-de-dades/
    Papper original source: Journal Of Clinical Medicine. 11 (17):
    APA: Pareja-Rios, Alicia; Ceruso, Sabato; Romero-Aroca, Pedro; Bonaque-Gonzalez, Sergio; (2022). A New Deep Learning Algorithm with Activation Mapping for Diabetic Retinopathy: Backtesting after 10 Years of Tele-Ophthalmology. Journal Of Clinical Medicine, 11(17), -. DOI: 10.3390/jcm11174945
    Article's DOI: 10.3390/jcm11174945
    Entity: Universitat Rovira i Virgili
    Journal publication year: 2022
    Publication Type: Journal Publications
  • Keywords:

    Medicine (Miscellaneous),Medicine, General & Internal
    Tele-ophthalmology
    Diabetic retinopathy
    Deep learning
    Artificial-intelligence
    Artificial intelligence
    Medicine, general & internal
    Medicine (miscellaneous)
    Medicine (all)
  • Documents:

  • Cerca a google

    Search to google scholar