Adverse pregnancy outcomes in women with systemic lupus erythematosus: can we improve predictions with machine learning?
Objectives Nearly 20% of pregnancies in patients with SLE result in an adverse pregnancy outcome (APO). We previously developed an APO prediction model using logistic regression and data from Predictors of pRegnancy Outcome: bioMarkers In Antiphospholipid Antibody Syndrome and Systemic Lupus Erythem...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMJ Publishing Group
2022-09-01
|
| Series: | Lupus Science and Medicine |
| Online Access: | https://lupus.bmj.com/content/9/1/e000769.full |
| _version_ | 1811208065103429632 |
|---|---|
| author | Jane Salmon Mimi Y Kim Marta M Guerra Melissa J Fazzari |
| author_facet | Jane Salmon Mimi Y Kim Marta M Guerra Melissa J Fazzari |
| author_sort | Jane Salmon |
| collection | DOAJ |
| description | Objectives Nearly 20% of pregnancies in patients with SLE result in an adverse pregnancy outcome (APO). We previously developed an APO prediction model using logistic regression and data from Predictors of pRegnancy Outcome: bioMarkers In Antiphospholipid Antibody Syndrome and Systemic Lupus Erythematosus (PROMISSE), a large multicentre study of pregnant women with mild/moderate SLE and/or antiphospholipid antibodies. Our goal was to determine whether machine learning (ML) approaches improve APO prediction and identify other risk factors.Methods The PROMISSE data included 41 predictors from 385 subjects; 18.4% had APO (preterm delivery due to placental insufficiency/pre-eclampsia, fetal/neonatal death, fetal growth restriction). Logistic regression with stepwise selection (LR-S), least absolute shrinkage and selection operator (LASSO), random forest (RF), neural network (NN), support vector machines (SVM-RBF), gradient boosting (GB) and SuperLearner (SL) were compared by cross-validated area under the ROC curve (AUC) and calibration.Results Previously identified APO risk factors, antihypertensive medication use, low platelets, SLE disease activity and lupus anticoagulant (LAC), were confirmed as important with each algorithm. LASSO additionally revealed potential interactions between LAC and anticardiolipin IgG, among others. SL performed the best (AUC=0.78), but was statistically indistinguishable from LASSO, SVM-RBF and RF (AUC=0.77 for all). LR-S, NN and GB had worse AUC (0.71–0.74) and calibration scores.Conclusions We predicted APO with reasonable accuracy using variables routinely assessed prior to the 12th week of pregnancy. LASSO and some ML methods performed better than a standard logistic regression approach. Substantial improvement in APO prediction will likely be realised, not with increasingly complex algorithms but by the discovery of new biomarkers and APO risk factors. |
| first_indexed | 2024-04-12T04:16:02Z |
| format | Article |
| id | doaj.art-b968b6e6a5c04bda92bee57bd96ff78f |
| institution | Directory Open Access Journal |
| issn | 2053-8790 |
| language | English |
| last_indexed | 2024-04-12T04:16:02Z |
| publishDate | 2022-09-01 |
| publisher | BMJ Publishing Group |
| record_format | Article |
| series | Lupus Science and Medicine |
| spelling | doaj.art-b968b6e6a5c04bda92bee57bd96ff78f2022-12-22T03:48:24ZengBMJ Publishing GroupLupus Science and Medicine2053-87902022-09-019110.1136/lupus-2022-000769Adverse pregnancy outcomes in women with systemic lupus erythematosus: can we improve predictions with machine learning?Jane Salmon0Mimi Y Kim1Marta M Guerra2Melissa J Fazzari3Division of Rheumatology, Department of Medicine, Hospital for Special Surgery, Weill Cornell Medical College, New York City, New York, USA3 Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, New York, USA2 Medicine, Hospital for Special Surgery, New York, New York, USAEpidemiology and Population Health, Albert Einstein College of Medicine, Bronx, New York, USAObjectives Nearly 20% of pregnancies in patients with SLE result in an adverse pregnancy outcome (APO). We previously developed an APO prediction model using logistic regression and data from Predictors of pRegnancy Outcome: bioMarkers In Antiphospholipid Antibody Syndrome and Systemic Lupus Erythematosus (PROMISSE), a large multicentre study of pregnant women with mild/moderate SLE and/or antiphospholipid antibodies. Our goal was to determine whether machine learning (ML) approaches improve APO prediction and identify other risk factors.Methods The PROMISSE data included 41 predictors from 385 subjects; 18.4% had APO (preterm delivery due to placental insufficiency/pre-eclampsia, fetal/neonatal death, fetal growth restriction). Logistic regression with stepwise selection (LR-S), least absolute shrinkage and selection operator (LASSO), random forest (RF), neural network (NN), support vector machines (SVM-RBF), gradient boosting (GB) and SuperLearner (SL) were compared by cross-validated area under the ROC curve (AUC) and calibration.Results Previously identified APO risk factors, antihypertensive medication use, low platelets, SLE disease activity and lupus anticoagulant (LAC), were confirmed as important with each algorithm. LASSO additionally revealed potential interactions between LAC and anticardiolipin IgG, among others. SL performed the best (AUC=0.78), but was statistically indistinguishable from LASSO, SVM-RBF and RF (AUC=0.77 for all). LR-S, NN and GB had worse AUC (0.71–0.74) and calibration scores.Conclusions We predicted APO with reasonable accuracy using variables routinely assessed prior to the 12th week of pregnancy. LASSO and some ML methods performed better than a standard logistic regression approach. Substantial improvement in APO prediction will likely be realised, not with increasingly complex algorithms but by the discovery of new biomarkers and APO risk factors.https://lupus.bmj.com/content/9/1/e000769.full |
| spellingShingle | Jane Salmon Mimi Y Kim Marta M Guerra Melissa J Fazzari Adverse pregnancy outcomes in women with systemic lupus erythematosus: can we improve predictions with machine learning? Lupus Science and Medicine |
| title | Adverse pregnancy outcomes in women with systemic lupus erythematosus: can we improve predictions with machine learning? |
| title_full | Adverse pregnancy outcomes in women with systemic lupus erythematosus: can we improve predictions with machine learning? |
| title_fullStr | Adverse pregnancy outcomes in women with systemic lupus erythematosus: can we improve predictions with machine learning? |
| title_full_unstemmed | Adverse pregnancy outcomes in women with systemic lupus erythematosus: can we improve predictions with machine learning? |
| title_short | Adverse pregnancy outcomes in women with systemic lupus erythematosus: can we improve predictions with machine learning? |
| title_sort | adverse pregnancy outcomes in women with systemic lupus erythematosus can we improve predictions with machine learning |
| url | https://lupus.bmj.com/content/9/1/e000769.full |
| work_keys_str_mv | AT janesalmon adversepregnancyoutcomesinwomenwithsystemiclupuserythematosuscanweimprovepredictionswithmachinelearning AT mimiykim adversepregnancyoutcomesinwomenwithsystemiclupuserythematosuscanweimprovepredictionswithmachinelearning AT martamguerra adversepregnancyoutcomesinwomenwithsystemiclupuserythematosuscanweimprovepredictionswithmachinelearning AT melissajfazzari adversepregnancyoutcomesinwomenwithsystemiclupuserythematosuscanweimprovepredictionswithmachinelearning |