AI in Paediatrics and Neonatology
In children and newborns, most AI work is still research: sepsis prediction in the NICU, retinopathy of prematurity screening, and brain injury imaging. Reviews say standard reporting and external validation are lacking. The nearest to routine use is chest X-ray AI and autism screening tools in some settings.
What AI does in Paediatrics and Neonatology
AI flags abnormal chest films, helpful where radiologists are scarce.
Examples: Qure.ai qXR
A device combines parent questionnaire, video and clinician input.
Examples: Cognoa Canvas Dx
Scribes and LLMs draft notes and plain-language instructions for parents.
Examples: Heidi, Abridge
Models read monitor data and labs to warn early.
Examples: Academic NICU models
Deep learning grades fundus photos for ROP severity.
Examples: Academic models
Models read cranial ultrasound, MRI or EEG for injury and seizures.
Examples: Academic models
routine = in everyday use in many centres · early adoption = in use at some sites, evidence still building · research = studies only
Key ideas for this specialty
New to the basics? Start with the theory lessons.
Tools
AI-assisted autism diagnosis for ages 1.5 to 6.
Regulatory status: Site describes it as FDA authorized
Chest X-ray AI used for TB and lung screening programmes in India.
Regulatory status: Site states FDA cleared and CE marked
AI scribe with a free tier.
Regulatory status: Site lists privacy compliance; no device clearance stated
Point-of-care EEG with AI seizure detection; check indications for neonates and children before use.
Regulatory status: Site does not state clearance
A listing here is not an endorsement. Ask any vendor for its CDSCO licence and class before clinical use.
Where to learn
More than 15 short webinars on AI for paediatricians, from basics to talking with families.
Hands-on diagnosis, prognosis and treatment models; needs Python and some deep learning background.
Interviews with people building and testing medical AI; hosts are Raj Manrai and Andrew Beam.
India's national ethics framework; useful before you put any AI tool into a thesis or ward workflow.
India's paediatric journal policy on LLM use in writing and research.
Five short courses that teach clinicians how to judge AI tools; CME available for physicians.
Indian course built for clinicians, from IIT Madras faculty.
Do it yourself
Download the Kermany paediatric chest X-ray set from Kaggle and fine-tune a small pretrained CNN. Check results on the 624-image test set, then discuss why a single-centre dataset may not carry over to Indian children.
Open the dataset or tool →Use the CHB-MIT paediatric EEG set on PhysioNet to build a simple seizure detector and compute false alarms per hour.
Open the dataset or tool →Use public or properly de-identified data only, and get ethics approval before any project on your own patients.
Papers worth reading
- The past, current, and future of neonatal intensive care units with artificial intelligence: a systematic review npj Digital Medicine, 2023
- AI models in clinical neonatology: a review of modeling approaches and a consensus proposal for standardized reporting of model performance Pediatric Research, 2024
- Reimagining Pediatrics in a World of Artificial Intelligence: Will We Be Empowered or Imperiled? Mayo Clinic Proceedings: Digital Health, 2025
- Indian Pediatrics' Policy Regarding Artificial Intelligence (AI)-Enabled Large Language Models Indian Pediatrics, 2023
Societies and guidance
AI hub with webinars and links to AMA and CHAI resources.
Society profile; IAP's journal Indian Pediatrics has published an AI policy and a President's Page on AI.
Links checked 3 October 2026. “Not listed” means we could not confirm the detail, not that it is absent. Educational material, not clinical advice.