مرکز اطلاعات علمی Scientific Information Database (SID) - Trusted Source for Research and Academic Resources

video

مرکز اطلاعات علمی Scientific Information Database (SID) - Trusted Source for Research and Academic Resources

sound

مرکز اطلاعات علمی Scientific Information Database (SID) - Trusted Source for Research and Academic Resources

Persian Version

مرکز اطلاعات علمی Scientific Information Database (SID) - Trusted Source for Research and Academic Resources

View:

27
مرکز اطلاعات علمی Scientific Information Database (SID) - Trusted Source for Research and Academic Resources

Download:

15
مرکز اطلاعات علمی Scientific Information Database (SID) - Trusted Source for Research and Academic Resources

Cites:

Information Journal Paper

Title

Prediction of T2DM Using Conjunctival Sac Microbiota, a Machine Learning Approach

Pages

  31-46

Abstract

 Background: Association of T2DM and OS disorders addresses a human eye metagenome drift. Despite the clarity of diabetic retinopathy, process of involvement of conjunctival sac microbiota is still ambiguous. We seek predictive value of OS microbiota using ML-based methods, since the early diagnosis of diabetic retinopathy and conjunctival sac microbiome diabetic dysbiosis could be beneficial both for clinical and research purposes. Material and Methods: 16S rRNA characterization of human eye metagenome for samples of 192 patients (with mean age of 66 years and 56 % females) with different onsets of T2DM is analyzed using various metrics including abundance and diversity indices and LDA at phyla, families, and genera levels. We took advantage of variance threshold, Chi-squared significance, and LDA Effect Size (LEfSe) feature selection strategies for inclusion of predictive families and genera in the T2DM prediction model. ML models with different algorithms including RF, GB, SVM, and ANN are implemented. Generalizability and robust performance of the models are also ensured using a 5-fold cross-validation process. DeLong’, s test is also used to investigate different performance of the methods. Results: Microbiome analyses revealed that eye metagenome profiles of the patients with <15 years of T2DM history show significantly higher richness and diversity. ML model performance shows ROC-AUC of ~0. 8. ML model with the superior performance exhibit sensitivity and accuracy of 0. 86 and 0. 68, respectively, in the prediction of T2DM occurrence. Conclusion: significant correlation and co-occurrence of T2DM and eye microbiome dysbiosis is trackable and well-optimized ML-strategies can predict T2DM onsets based on the microbiome of conjunctival sac.

Multimedia

  • No record.

    • Cites

  • No record.

    • References

  • No record.

    • Cite

    APA: Copy

    Setareh, Soroush, Kavianfar, Azadeh, & Taherkhani, Hamidreza. (2021). Prediction of T2DM Using Conjunctival Sac Microbiota, a Machine Learning Approach. JOURNAL OF OPHTHALMIC AND OPTOMETRIC SCIENCES, 5(2), 31-46. SID. https://sid.ir/paper/1094786/en

    Vancouver: Copy

    Setareh Soroush, Kavianfar Azadeh, Taherkhani Hamidreza. Prediction of T2DM Using Conjunctival Sac Microbiota, a Machine Learning Approach. JOURNAL OF OPHTHALMIC AND OPTOMETRIC SCIENCES[Internet]. 2021;5(2):31-46. Available from: https://sid.ir/paper/1094786/en

    IEEE: Copy

    Soroush Setareh, Azadeh Kavianfar, and Hamidreza Taherkhani, “Prediction of T2DM Using Conjunctival Sac Microbiota, a Machine Learning Approach,” JOURNAL OF OPHTHALMIC AND OPTOMETRIC SCIENCES, vol. 5, no. 2, pp. 31–46, 2021, [Online]. Available: https://sid.ir/paper/1094786/en

    Related Journal Papers

  • No record.

    • Related Seminar Papers

  • No record.

    • Related Plans

  • No record.

    • Recommended Workshops






    Move to top
    telegram sharing button
    whatsapp sharing button
    linkedin sharing button
    twitter sharing button
    email sharing button
    email sharing button
    email sharing button
    sharethis sharing button