Classifying Dry Eye Disease Patients from Healthy Controls Using Machine Learning and Metabolomics Data.
| Title: | Classifying Dry Eye Disease Patients from Healthy Controls Using Machine Learning and Metabolomics Data. |
|---|---|
| Authors: | Amouei Sheshkal S; Department of Computer Science, Oslo Metropolitan University, 0166 Oslo, Norway.; Department of Holistic Systems, SimulaMet, 0167 Oslo, Norway.; Ifocus Eye Clinic, 5527 Haugesund, Norway.; Gundersen M; Ifocus Eye Clinic, 5527 Haugesund, Norway.; Department of Life Sciences and Health, Oslo Metropolitan University, 0166 Oslo, Norway.; Alexander Riegler M; Department of Computer Science, Oslo Metropolitan University, 0166 Oslo, Norway.; Department of Holistic Systems, SimulaMet, 0167 Oslo, Norway.; Aass Utheim Ø; Department of Ophthalmology, Oslo University Hospital, 0450 Oslo, Norway.; Gunnar Gundersen K; Ifocus Eye Clinic, 5527 Haugesund, Norway.; Rootwelt H; Department of Medical Biochemistry, Oslo University Hospital, 0450 Oslo, Norway.; Prestø Elgstøen KB; Department of Medical Biochemistry, Oslo University Hospital, 0450 Oslo, Norway.; Lewi Hammer H; Department of Computer Science, Oslo Metropolitan University, 0166 Oslo, Norway.; Department of Holistic Systems, SimulaMet, 0167 Oslo, Norway. |
| Source: | Diagnostics (Basel, Switzerland) [Diagnostics (Basel)] 2024 Nov 29; Vol. 14 (23). Date of Electronic Publication: 2024 Nov 29. |
| Publication Type: | Journal Article |
| Language: | English |
| Journal Info: | Publisher: MDPI AG Country of Publication: Switzerland NLM ID: 101658402 Publication Model: Electronic Cited Medium: Print ISSN: 2075-4418 (Print) Linking ISSN: 20754418 NLM ISO Abbreviation: Diagnostics (Basel) Subsets: PubMed not MEDLINE |
| Imprint Name(s): | Original Publication: Basel, Switzerland : MDPI AG, [2011]- |
| Abstract: | Background: Dry eye disease is a common disorder of the ocular surface, leading patients to seek eye care. Clinical signs and symptoms are currently used to diagnose dry eye disease. Metabolomics, a method for analyzing biological systems, has been found helpful in identifying distinct metabolites in patients and in detecting metabolic profiles that may indicate dry eye disease at early stages. In this study, we explored the use of machine learning and metabolomics data to identify cataract patients who suffer from dry eye disease, a topic that, to our knowledge, has not been previously explored. As there is no one-size-fits-all machine learning model for metabolomics data, choosing the most suitable model can significantly affect the quality of predictions and subsequent metabolomics analyses. Methods: To address this challenge, we conducted a comparative analysis of eight machine learning models on two metabolomics data sets from cataract patients with and without dry eye disease. The models were evaluated and optimized using nested k-fold cross-validation. To assess the performance of these models, we selected a set of suitable evaluation metrics tailored to the data set's challenges. Results: The logistic regression model overall performed the best, achieving the highest area under the curve score of 0.8378, balanced accuracy of 0.735, Matthew's correlation coefficient of 0.5147, an F1-score of 0.8513, and a specificity of 0.5667. Additionally, following the logistic regression, the XGBoost and Random Forest models also demonstrated good performance. Conclusions: The results show that the logistic regression model with L2 regularization can outperform more complex models on an imbalanced data set with a small sample size and a high number of features, while also avoiding overfitting and delivering consistent performance across cross-validation folds. Additionally, the results demonstrate that it is possible to identify dry eye in cataract patients from tear film metabolomics data using machine learning models. |
| References: | Eur J Ophthalmol. 2020 Sep;30(5):840-855. (PMID: 32515220); Am J Ophthalmol. 2021 Jun;226:252-261. (PMID: 33529589); Sci Rep. 2022 Mar 14;12(1):4361. (PMID: 35288652); J Proteome Res. 2018 Jan 5;17(1):337-347. (PMID: 29110491); Clin Ophthalmol. 2024 Jan 16;18:151-161. (PMID: 38259819); Ocul Surf. 2022 Jan;23:74-86. (PMID: 34843999); iScience. 2022 Jan 22;25(2):103798. (PMID: 35169688); Sci Rep. 2017 Apr 12;7:46453. (PMID: 28401923); Ecology. 2007 Nov;88(11):2783-92. (PMID: 18051647); Environ Sci Technol. 2021 Aug 3;55(15):10542-10557. (PMID: 34260856); Genes Dis. 2023 Jul 07;11(3):100979. (PMID: 38299197); Front Genet. 2022 Nov 24;13:1017340. (PMID: 36506316); Clin Ophthalmol. 2023 Apr 27;17:1233-1243. (PMID: 37138823); Anal Bioanal Chem. 2016 Apr;408(10):2497-504. (PMID: 26869345); Comput Biol Med. 2022 Jul;146:105505. (PMID: 35477047); J Proteome Res. 2021 Aug 6;20(8):4010-4021. (PMID: 34296888); Curr Eye Res. 2024 Jul;49(7):708-716. (PMID: 38567868); Environ Sci Technol. 2012 Jul 17;46(14):7454-62. (PMID: 22702400); Sci Rep. 2016 Dec 13;6:38881. (PMID: 27958387); BMC Bioinformatics. 2018 Dec 21;19(Suppl 20):506. (PMID: 30577740); Clin Ophthalmol. 2014 Mar 20;8:581-90. (PMID: 24672224); Sci Rep. 2024 Jan 2;14(1):152. (PMID: 38168582); JMIR Med Inform. 2020 Feb 20;8(2):e16153. (PMID: 32130150); Am J Ophthalmol. 2019 Jun;202:47-54. (PMID: 30721689); BMC Genomics. 2006 Jun 08;7:142. (PMID: 16762068); J Mass Spectrom. 2004 Apr;39(4):378-83. (PMID: 15103651); Eur J Mass Spectrom (Chichester). 2020 Jun;26(3):165-174. (PMID: 32276547); Clin Ophthalmol. 2023 Mar 11;17:829-835. (PMID: 36937165); Br J Ophthalmol. 2015 Nov;99(11):1488-92. (PMID: 25883086); Clin Ophthalmol. 2024 Feb 26;18:591-604. (PMID: 38435373); Cornea. 2020 Feb;39(2):161-167. (PMID: 31517720); Int J Mol Sci. 2019 Aug 01;20(15):. (PMID: 31374809); J Proteome Res. 2023 Aug 4;22(8):2734-2742. (PMID: 37395192); Ocul Surf. 2021 Jan;19:183-189. (PMID: 32504855); IEEE Trans Syst Man Cybern B Cybern. 2009 Apr;39(2):539-50. (PMID: 19095540); Ocul Surf. 2019 Jan;17(1):9-19. (PMID: 30419303); Reumatologia. 2015;53(3):152-6. (PMID: 27407242); Front Mol Biosci. 2021 Jun 17;8:634141. (PMID: 34222327); Metabolomics. 2019 Nov 15;15(12):150. (PMID: 31728648); Front Immunol. 2020 Jul 17;11:1527. (PMID: 32765529); Bioinformatics. 2014 Aug 1;30(15):2155-61. (PMID: 24711654); Anal Chem. 2021 Feb 2;93(4):2471-2479. (PMID: 33471512); Anal Bioanal Chem. 2018 Mar;410(7):1931-1941. (PMID: 29380019); Ann Neurol. 2020 Oct;88(4):736-746. (PMID: 32748431); Diagnostics (Basel). 2023 Nov 21;13(23):. (PMID: 38066735); Eye Vis (Lond). 2020 Apr 16;7:22. (PMID: 32322599); Ocul Surf. 2017 Jul;15(3):539-574. (PMID: 28736342); J Clin Epidemiol. 2020 Jun;122:56-69. (PMID: 32169597); Psychol Methods. 2009 Dec;14(4):323-48. (PMID: 19968396); Infect Dis Ther. 2022 Aug;11(4):1695-1713. (PMID: 35835943); Clin Chim Acta. 2017 Jan;464:223-227. (PMID: 27931880) |
| Contributed Indexing: | Keywords: classification; dry eye disease; hyper-parameters tuning; machine learning; metabolomics |
| Entry Date(s): | Date Created: 20241217 Latest Revision: 20250104 |
| Update Code: | 20260130 |
| PubMed Central ID: | PMC11640104 |
| DOI: | 10.3390/diagnostics14232696 |
| PMID: | 39682603 |
| Database: | MEDLINE |
Journal Article