Celina Dubin MD-Using random forest models to predict drug reaction with eosinophilia and systemic…
This article was co-authored by Celina Dubin, MD, a dermatology resident at the Icahn School of Medicine at Mount Sinai, and was published…
Celina Dubin MD-Using random forest models to predict drug reaction with eosinophilia and systemic symptoms development from CBC parameters.
Read the abstract here: https://www.jidonline.org/article/S0022-202X(25)00645-1/fulltext
This article was co-authored by Celina Dubin, MD, a dermatology resident at the Icahn School of Medicine at Mount Sinai, and was published in the Journal of Investigative Dermatology. Drug reaction with eosinophilia and systemic symptoms (DRESS) is a severe adverse reaction that is often clinically indistinguishable from simple morbilliform drug eruptions at rash onset. The authors implemented random forest models to classify patients as DRESS-positive or DRESS-negative using complete blood count (CBC) with differential metrics, monocyte-to-lymphocyte ratio (MLR), neutrophil-to-monocyte ratio (NMR), neutrophil-to-lymphocyte ratio (NLR), and the pan-immune inflammation value (PIV), defined as (neutrophil × platelet × monocyte) / lymphocyte, collected within 10 days of rash onset. Models with varying tree depths were trained on a dataset of 261 patients using a 75:25 train–test split. Model performance was compared to identify the optimal depth for accurate DRESS prediction. The no-depth model achieved the highest accuracy (90.5%) but showed signs of overfitting. The 3-depth model underfit, achieving an accuracy of 71.4%. The 5-depth model demonstrated more balanced performance, achieving 81% accuracy. The 7-depth model showed the best overall balance of metrics, with an accuracy of 85.7%, a precision of 91%, and a recall of 83% for DRESS-positive predictions. These results highlight the potential of random forest models for early DRESS diagnosis. In particular, the 7-depth model demonstrated strong diagnostic capability, suggesting that routine CBC parameters may support accurate DRESS prediction at rash onset, prior to the development of additional clinical indicators.
READ THE ABSTRACT IN THE JOURNAL OF INVESTIGATIVE DERMATOLOGY HERE:
More Sources:
[embed]
[embed]
[embed]
메타데이터
- post_id
- ebeeb6d6875c
- slug
- celina-dubin-md-using-random-forest-models-to-predict-drug-reaction-with-eosinophilia-and-systemic-ebeeb6d6875c
- url
- https://medium.com/@celinadubin/celina-dubin-md-using-random-forest-models-to-predict-drug-reaction-with-eosinophilia-and-systemic-ebeeb6d6875c
- canonical_url
- https://medium.com/@celinadubin/celina-dubin-md-using-random-forest-models-to-predict-drug-reaction-with-eosinophilia-and-systemic-ebeeb6d6875c
- author_url
- https://medium.com/@celinadubin
- status
- ok
- fetched_at
- 2026-06-09 15:37:30
