Optimizing the Dutch newborn screening for congenital hypothyroidism by incorporating amino acids and acylcarnitines in a machine learning-based model
Authors' conclusion
The PPV improved significantly (26-48%) by adding several AAs and ACs to our machine learning-based model, suggesting that adding these parameters benefits the current algorithm.
Study context
- Population
- Dutch newborn screening (NBS) data 2007–2017: 1079 false-positive referrals, 515 newborns with primary (431) and central congenital hypothyroidism (CH) (84), and 1842 healthy controls
- Comparison
- Random forest model incorporating amino acids (AAs) and acylcarnitines (ACs) vs current Dutch T4-based CH NBS
- Outcome
- Screening performance for CH
Methodological detailsmachine learning-based model, random forest model, retrospective newborn screening data 2007–2017
- Machine learning-based model
- Random forest model
- Retrospective newborn screening data 2007–2017
Performance
As reported in the abstract
| Test or model | Sensitivity (artificial) | PPV | AUC |
|---|---|---|---|
| Random forest model | 100% | 48% | 0.99 |
Random forest model
Sensitivity (artificial)100%
PPV48%
AUC0.99
Parts of this Brief are withheld: the automatic checks could not confirm them against the abstract.
Original publication
Optimizing the Dutch newborn screening for congenital hypothyroidism by incorporating amino acids and acylcarnitines in a machine learning-based model. Eur Thyroid J. Published 3 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →