Table 2.
Selected papers identified through a structured MEDLINE search.
| Data type | Authors | Title | Year | Journal | Main topic of focus | Type of statistical technique |
|---|---|---|---|---|---|---|
| Clinical | Mossotto et al.2 | Classification of paediatric inflammatory bowel disease using machine learning | 2017 | Sci Rep | Diagnostic classification and cluster discovery | Support vector machine, hierarchical clustering |
| Clinical | Ashton et al.17 | Analysis and hierarchical clustering of blood results before diagnosis in paediatric inflammatory bowel disease | 2020 | IBD | Cluster discovery using clinical data | Hierarchical clustering |
| Exome and transcriptome | Ashton et al.19 | Deleterious genetic variation across the NOD signalling pathway is associated with reduced NFKB signalling transcription and upregulation of alternative inflammatory transcripts in paediatric inflammatory bowel disease | 2022 | IBD | Cluster discovery using molecular data | Hierarchical clustering |
| Clinical | Dhaliwal et al.18 | Accurate classification of paediatric colonic inflammatory bowel disease subtype using a random forest machine learning classifier | 2021 | JPGN | Diagnostic classification | Random forest classifier |
| Clinical, molecular + biochemical | Kugathasan et al.20 | Prediction of complicated disease course for children newly diagnosed with Crohn’s disease: a multicentre inception cohort study | 2017 | The Lancet | Prediction of complex disease | Competing risk model |
| Clinical, molecular + biochemical | Hyams et al.21 | Clinical and biological predictors of response to standardised paediatric colitis therapy (PROTECT): a multicentre inception cohort study | 2019 | The Lancet | Prediction of corticosteroid-free remission | Logistic regression |
| Biochemical | Ungaro et al.22 | Machine learning identifies novel blood protein predictors of penetrating and stricturing complications in newly diagnosed paediatric Crohn’s disease | 2021 | APT | Prediction of complex disease | Random forest Survival |
| Molecular | Douglas et al.24 | Multi-omics differentially classify disease state and treatment outcome in paediatric Crohn’s disease | 2018 | Microbiome | Prediction of response to induction treatment | Random forest classifier |
| Molecular | Jones et al.25 | Bacterial taxa and functions are predictive of sustained remission following exclusive enteral nutrition in paediatric Crohn’s disease | 2020 | IBD | Prediction of response to exclusive enteral nutrition | Random forest classifier |
All studies use machine learning techniques to investigate the classification, group discovery, outcome and therapy prediction in paediatric inflammatory bowel disease.