PEDICTOR: Framework for Predicting Mortality in Critically Ill Children Using Time Series PICU Data

Tuesday, March 24, 2026

11:15 AM - 12:15 PM Central Time

Location: Connections Central - RST 03

Naveed Siddiqui, n/a (he/him/his)

Aga Khan University Hospital, Sindh, PAK

Disclosure(s): No relevant financial relationship(s) to disclose.
MG

Muhammad Amin Ghias, NA

Mr
Institute of Business Administration (IBA)

Disclosure information not submitted.
HN

Hassan Naqvi, NA

MR
Aga Khan University Hospital

Disclosure information not submitted.
HS

Humaira Shafaq, NA

Ms
Aga Khan University Hospital

Disclosure information not submitted.
RO

Rabiya Owais, NA

Ms
Aga Khan University Hospital, Sindh, Pakistan

Disclosure information not submitted.
Qalab Abbas, Physician (he/him/his)

Aga Khan University, PAK

Disclosure(s): No relevant financial relationship(s) to disclose.
FJ

Fyezah Jehan, MBBS, FCPS, MPH

Professor, Paediatrics and Child Health
Aga Khan University

Disclosure information not submitted.
SH

Saadia Humayun, NA

Ms
Institute of Business Administration (IBA)

Disclosure information not submitted.
TM

Tariq Mahmood, NA

prof
Institute of Business Administration (IBA)

Disclosure information not submitted.

Introduction: Pediatric mortality prediction remains a critical task in intensive care. Traditional statistical approaches struggle to fully utilize rich patient data such as vitals and lab results. This study proposes a novel framework, PEDICTOR, that integrates expert domain knowledge with machine learning (ML) and deep learning (DL) techniques to improve mortality prediction in pediatric intensive care unit (PICU) patients.

Methods: Pediatric mortality prediction remains a critical task in intensive care. Traditional statistical approaches struggle to fully utilize rich patient data such as vitals and lab results. This study proposes a novel framework, PEDICTOR, that integrates expert domain knowledge with machine learning (ML) and deep learning (DL) techniques to improve mortality prediction in pediatric intensive care unit (PICU) patients.

Results: The PEDICTOR stacker achieved an average per-minute F1-score of 0.441 (95% CI: 0.439–0.442) and an AUCROC of 0.932 (95% CI: 0.9314–0.9325) for predicting expired cases. The framework also attained an average 30-minute prediction F1-score of 0.591 (95% CI: 0.589–0.592) and an average hourly prediction F1-score of 0.649 (95% CI: 0.648–0.651). Results were consistent across all five pediatric age groups.

Conclusions: PEDICTOR offers a robust, age-aware mortality prediction system leveraging ensemble ML-DL stacking and cluster-based oversampling. Its ability to accurately predict per-minute, 30 minute, and hourly outcomes can support timely clinical decision-making. This framework can help clinicians intervene early, allocate resources efficiently, and potentially save lives in pediatric critical care settings.