Monarch

Monarch logo

A three-stage clinical decision support console for Systemic Lupus Erythematosus.

Presenter:Lezhi Lin

Authors:Manna Berry1, Lezhi Lin1, Udit Samant1, Hadi Shafat1, Jillian Zhao1 and Minh Hieu Tran6

Advisors:Dr. Andy Tran and Elyna Lin

The University of Sydney

Background:Systemic Lupus Erythematosus

SLE character illustration Cartoon human figure with lupus malar (butterfly) rash and chest rashes.

Previous Innovations: Better A three-stage machine learning pipelines

Diagnosis
Progression
Treatment

Methodology: modelling workflow

01 Data collection

  • DiagnosisGSE72509
  • ProgressionGSE65391, GSE49454
  • TreatmentGSE224705

02 Pre-processing & EDA

  • Dataset cleaning
  • Quality check
  • Standardisation

03 Gene selection

  • Ranked by importance
  • Curated to a compact panel
  • 141genes in total

04 Machine learning

Limma
LASSO
Elastic Net
GBM
Linear SVM
Random Forest

05 Performance evaluation

Imbalanced data
Stratified 5-fold cross-validation
AUROC, Balanced Accuracy, Macro-F1

A Random Forest model leading the field

01Diagnosis
02Progression
03Treatment
AUROC Higher is better
Typical machine learning products in clinical practice
0.972
0.875
0.852
0.670
0.865
0.750

Model discrimination benchmarked against literature results.

Monarch

From a blood sample to a recommendation in one click.

A Monarch Butterfly

Monarch

Decision support for SLE

Conclusions

Meet Our Team

Udit Samant

Diagnosis model

Manna Berry

Progression model

Jillian Zhao

Treatment model

Hadi Shafat

Research

Lezhi Lin

App development

Dr. Andy Tran and Elyna Lin

Our beloved advisors

Thank you.

Presenter:Lezhi Lin

Authors:Manna Berry1, Lezhi Lin1, Udit Samant1, Hadi Shafat1, Jillian Zhao1 and Minh Hieu Tran6

Advisors:Dr. Andy Tran and Elyna Lin

The University of Sydney

Appendix · 01 / 07

Full model performance.

01 Diagnosis
ModelAUROCBAL-ACCMACRO F1
RF0.9720.9390.950
LASSO0.9620.9340.934
limma0.9530.8810.849
02 Progression
ModelAUROCBAL-ACCMACRO F1
RF0.8520.7900.789
Elastic Net0.8450.8010.800
GBM0.8360.7910.789
03 Treatment
ModelAUROCBAL-ACCMACRO F1
RF0.8650.7890.791
Elastic Net0.7800.7360.723
LASSO0.7570.7220.698
Linear SVM0.7920.7090.682
limma0.6560.6210.621

Appendix · 02 / 07

Background & epidemiology.

Clinical background, epidemiology, and the global burden of the disease.

  • Aringer, M., & Bertsias, G. (2025). Early diagnosis of systemic lupus erythematosus. Rare Disease and Orphan Drugs Journal, 4, 13. https://doi.org/10.20517/rdodj.2024.59
  • Baechler, E. C., Batliwalla, F. M., Karypis, G., Gaffney, P. M., Ortmann, W. A., Espe, K. J., Shark, K. B., Grande, W. J., Hughes, K. M., Kapur, V., Gregersen, P. K., & Behrens, T. W. (2003). Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus. Proceedings of the National Academy of Sciences, 100(5), 2610–2615. https://doi.org/10.1073/pnas.0337679100
  • Kwon, Y.-C., Chun, S., Kim, K., & Mak, A. (2019). Update on the genetics of systemic lupus erythematosus: Genome-wide association studies and beyond. Cells, 8(10), 1180. https://doi.org/10.3390/cells8101180
  • Lin, D. H., Murimi-Worstell, I. B., Kan, H., Tierce, J. C., Wang, X., Nab, H., Desta, B., Hammond, E. R., & Alexander, G. C. (2022). Health care utilization and costs of systemic lupus erythematosus in the United States: A systematic review. Lupus, 31(7), 773–807. https://doi.org/10.1177/09612033221088209
  • National Health Service. (2023). Lupus. https://www.nhs.uk/conditions/lupus/ [Reviewed July 19, 2023]
  • National Institute of Arthritis and Musculoskeletal and Skin Diseases. (2022). Systemic lupus erythematosus (lupus). https://www.niams.nih.gov/health-topics/lupus [Last reviewed October 2022]
  • National Library of Medicine. (2024). Lupus. MedlinePlus. https://medlineplus.gov/lupus.html [Last updated July 1, 2024]
  • Natural Earth. (n.d.). Admin 0 – Countries (5.1.1) [Data set]. https://www.naturalearthdata.com/ Public-domain vector map dataset. Retrieved 2026.
  • Tian, J., Zhang, D., Yao, X., Huang, Y., & Lu, Q. (2023). Global epidemiology of systemic lupus erythematosus: A comprehensive systematic analysis and modelling study. Annals of the Rheumatic Diseases, 82(3), 351–356. https://doi.org/10.1136/ard-2022-223035
  • Wang, H., Li, M., Zou, K., Wang, Y., Jia, Q., Wang, L., Zhao, J., Wu, C., Wang, Q., Tian, X., Wang, Y., & Zeng, X. (2023). Annual direct cost and cost-drivers of systemic lupus erythematosus: A multi-center cross-sectional study from CSTAR registry. International Journal of Environmental Research and Public Health, 20(4), 3522. https://doi.org/10.3390/ijerph20043522

Appendix · 03 / 07

Datasets & cohorts.

The public gene-expression cohorts the three models are trained and validated on.

  • Banchereau, R., Hong, S., Cantarel, B., Baldwin, N., Baisch, J., Edens, M., Cepika, A.-M., Acs, P., Turner, J., Anguiano, E., & Pascual, V. (2016). Personalized immunomonitoring uncovers molecular networks that stratify lupus patients. Cell, 165(3), 551–565. https://doi.org/10.1016/j.cell.2016.03.008 [Data set; GSE65391]
  • Chiche, L., Jourde-Chiche, N., Whalen, E., Presnell, S., Gersuk, V., Dang, K., & Chaussabel, D. (2014). Modular transcriptional repertoire analyses of adults with SLE reveal distinct type I and type II interferon signatures. Arthritis & Rheumatology, 66(6), 1583–1595. https://doi.org/10.1002/art.38628 [Data set; GSE49454]
  • Hung, T., Pratt, G. A., Sundararaman, B., Townsend, M. J., Chaivorapol, C., Bhangale, T., Graham, R. R., Ortmann, W., Behrens, T. W., Yeo, G. W., & Chaussabel, D. (2015). The Ro60 autoantigen regulates inflammatory gene expression in SLE [Data set]. NCBI Gene Expression Omnibus. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE72509 [GSE72509]
  • NCBI Gene Expression Omnibus. (2023). Whole-blood microarray expression in lupus nephritis: Treatment response by SRI-4 [Data set]. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE224705 [GSE224705]
Data access · NCBI Gene Expression Omnibus (public)

Appendix · 04 / 07

Modelling & evaluation.

The learning algorithms, feature selection, and clinical scoring behind the pipeline.

  • Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
  • Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357. https://doi.org/10.1613/jair.953
  • Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785
  • Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 273–297. https://doi.org/10.1007/BF00994018
  • Friedman, J., Hastie, T., & Tibshirani, R. (2010). Regularization paths for generalized linear models via coordinate descent. Journal of Statistical Software, 33(1), 1–22. https://doi.org/10.18637/jss.v033.i01
  • Furie, R., Petri, M. A., Wallace, D. J., Ginzler, E. M., Merrill, J. T., Stohl, W., Chatham, W. W., Strand, V., Weinstein, A., & Chevrier, M. (2009). Novel evidence-based systemic lupus erythematosus responder index. Arthritis & Rheumatism, 61(9), 1143–1151. https://doi.org/10.1002/art.24698
  • Gladman, D. D., Ibañez, D., & Urowitz, M. B. (2002). Systemic lupus erythematosus disease activity index 2000. The Journal of Rheumatology, 29(2), 288–291.
  • Hosmer, D. W., Lemeshow, S., & Sturdivant, R. X. (2013). Applied logistic regression (3rd ed.). John Wiley & Sons. https://doi.org/10.1002/9781118548387 (AUROC interpretation thresholds — 0.7–0.8 acceptable, 0.8–0.9 excellent)
  • Kursa, M. B., & Rudnicki, W. R. (2010). Feature selection with the Boruta package. Journal of Statistical Software, 36(11), 1–13. https://doi.org/10.18637/jss.v036.i11
  • Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30, 4765–4774.
  • Ritchie, M. E., Phipson, B., Wu, D., Hu, Y., Law, C. W., Shi, W., & Smyth, G. K. (2015). limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research, 43(7), e47. https://doi.org/10.1093/nar/gkv007

Appendix · 05 / 07

Competitive landscape.

Existing lupus tests, published model benchmarks, the gene-expression analogue, and market size.

  • Exagen. (n.d.). AVISE Lupus [Diagnostic test]. Exagen. https://exagen.com/tests/lupus/
  • Jiang, Z., Shao, M., Dai, X., Pan, Z., & Liu, D. (2022). Identification of diagnostic biomarkers in systemic lupus erythematosus based on bioinformatics analysis and machine learning. Frontiers in Genetics, 13, 865559. https://doi.org/10.3389/fgene.2022.865559
  • Kegerreis, B., Catalina, M. D., Bachali, P., Geraci, N. S., Labonte, A. C., Zeng, C., Stocks, N., Hubbard, E. L., Grammer, A. C., & Lipsky, P. E. (2019). Machine learning approaches to predict lupus disease activity from gene expression data. Scientific Reports, 9, 9617. https://doi.org/10.1038/s41598-019-45989-0
  • Lee, D.-J., Tsai, P.-H., Chen, C.-C., & Dai, Y.-H. (2023). Incorporating knowledge of disease-defining hub genes and regulatory network into a machine learning-based model for predicting treatment response in lupus nephritis after the first renal flare. Journal of Translational Medicine, 21, 76. https://doi.org/10.1186/s12967-023-03931-z
  • Leventhal, E. L., Daamen, A. R., Grammer, A. C., & Lipsky, P. E. (2023). An interpretable machine learning pipeline based on transcriptomics predicts phenotypes of lupus patients. iScience, 26(10), 108042. https://doi.org/10.1016/j.isci.2023.108042
  • Li, Y., Yao, L., Lee, Y. A., Huang, Y., Merkel, P. A., Vina, E., Yeh, Y.-Y., Li, Y., Allen, J. M., Bian, J., & Guo, J. (2025). A fair machine learning model to predict flares of systemic lupus erythematosus. JAMIA Open, 8(4), ooaf072. https://doi.org/10.1093/jamiaopen/ooaf072
  • Munguía-Realpozo, P., Etchegaray-Morales, I., Mendoza-Pinto, C., Méndez-Martínez, S., Osorio-Peña, Á. D., Ayón-Aguilar, J., & García-Carrasco, M. (2023). Current state and completeness of reporting clinical prediction models using machine learning in systemic lupus erythematosus: A systematic review. Autoimmunity Reviews, 22(5), 103294. https://doi.org/10.1016/j.autrev.2023.103294
  • Paik, S., Shak, S., Tang, G., Kim, C., Baker, J., Cronin, M., Baehner, F. L., Walker, M. G., Watson, D., Park, T., Hiller, W., Fisher, E. R., Wickerham, D. L., Bryant, J., & Wolmark, N. (2004). A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. New England Journal of Medicine, 351(27), 2817–2826. https://doi.org/10.1056/NEJMoa041588
  • Progentec Diagnostics. (n.d.-a). aiSLE DX Flare Risk Index [Diagnostic test]. Progentec. https://www.progentec.com/aisle-dx-fri
  • Progentec Diagnostics. (n.d.-b). aiSLE MGMT: Lupus care-management platform [Software]. Progentec. https://www.progentec.com/aisle-mgmt
  • Research and Markets. (2025, January 16). Systemic lupus erythematosus (SLE) market forecast at $6.19 billion by 2034 [Market report]. GlobeNewswire. https://www.globenewswire.com/…
  • Virtue Market Research. (2025, December 16). Point-of-care testing for systemic lupus erythematosus (SLE) market [Market report]. OpenPR. https://www.openpr.com/news/4316397/the-global-point-of-care-testing-for-systemic-lupus

Appendix · 06 / 07

Technology & craft.

The tools the system is built with, and the design language it follows.

Appendix · 07 / 07

Team & acknowledgements.

DATA3888 · 2026
The University of Sydney

Monarch - Developers
Manna Berry1Development of Progression Model & Assistance with Backend
[email protected] Faculty of Engineering J12, The University of Sydney, NSW 2006
Lezhi Lin1Development of App Frontend & Presentation Slides
[email protected] School of Mathematics and Statistics F07, The University of Sydney, NSW 2006 Australia
Udit Samant1Development of Diagnosis Model & General App Backend
[email protected] School of Computer Science J12, The University of Sydney, NSW 2006 Australia
Hadi Shafat1Interdisciplinary Aspects Research & Assistance with Backend
[email protected] School of Computer Science J12, The University of Sydney, NSW 2006 Australia
Jillian Zhao1Development of Treatment Model & Assistance with Backend & Background Research
[email protected] School of Computer Science J12, The University of Sydney, NSW 2006 Australia
Minh Hieu Tran6Assistance with Exploratory Data Analysis
[email protected] School of Computer Science J12, The University of Sydney, NSW 2006 Australia
Acknowledgements
  • We acknowledge the Gadigal of the Eora Nation, the Traditional Custodians of the land on which the University of Sydney stands, and pay our respects to Elders past and present.
  • This slide is submitted in partial fulfillment of the assessment requirements for DATA3888 Data Science Capstone at The University of Sydney. Our work rests on the work of open-source maintainers across R, Bioconductor, and the modelling libraries used here, as well as the DATA3888 teaching team for project structure, feedback, and course support.
  • We are extremely grateful to our advisors, Dr. Andy Tran and Elyna Lin, for all the guidance, thoughtful feedback, and steady support during both the workshops and consultations, throughout the project.
  • We thank fellow students Anina Xinyu Shu and Sicheng Chen for their valuable insights on the development of this project and the construction of the presentation.
  • We acknowledge the original data contributors and study participants behind the public GEO cohorts. Their shared expression and clinical metadata made the modelling, validation, and patient-level demonstrations possible.
  • We acknowledge the use of AI-assisted tools to support drafting, code iteration, interface refinement, and debugging. All AI-assisted outputs were reviewed, edited, and validated by the team, who remain responsible for the final analysis, design decisions, and implementation.