Neonatal Jaundice Assessment: A Dataset of Forehead and Sternum Images for Bilirubin Estimation

This dataset includes images and clinical data aimed at estimating bilirubin levels in newborns to enhance non-invasive neonatal jaundice detection through deep learning. Images were captured using a digital microscope connected to a smartphone via USB, saved in jpg format with a resolution of 640 x 480 pixels.
Dataset Contents:
Images: Forehead and sternum images of 300 infants, saved in jpg format (640 x 480 pixels).
Data Collection Process:
Location: Ghaem Hospital, Mashhad, Iran
Duration: January 1, 2023 - May 30, 2023
Methodology:
TSB levels were determined via blood samples.
TcB levels were measured using the MBJ20 bilirubin meter (range: 0 to 32 mg/dL).
Images were taken within 30 minutes of TSB and TcB measurements to ensure consistency.
Ethics: Approved by the ethical committee at Mashhad University of Medical Sciences (IR.MUMS.MEDICAL.REC.1400.271), with informed written and verbal consent from parents.

Usage Notes:
This dataset is ideal for researchers focusing on the development and validation of non-invasive methods for bilirubin estimation using image processing. It is particularly useful for studies related to the detection and management of neonatal jaundice.