Machine Learning for Detection of Correct Peripherally Inserted Central Catheter Lines and Tips Position from NICU Radiology


Abstract:


In critically ill infants, the position of a peripherally inserted central catheter (PICC) must be confirmed frequently, as the tip may move from its original position and run the risk of hyperosmolar vascular damage or extravasation into surrounding spaces. Automated detection of PICC lines and tips position holds great promise for alerting bedside clinicians to non-central PICCs. This research seeks to use natural language processing (NLP) [1], image processing [2] with machine learning (ML) techniques to predict, and segment PICC lines and tips positions based on text analysis of radiograph reports, and corresponding X-ray imaging data.

NLP: Detection of correct PICC tip positions from unstructured radiology reports [1]


Check-out our website : https://www.picclocation.com


An approximately 1 kg baby with a PICC. The yellow arrow points to the tip of the upper extremity PICC in an appropriate position in the SVC. The yellow boxed area represents the location where the tip position would be considered appropriate (SVC and the SVC/RA junction). PICC, peripherally inserted central catheter; RA, right atrium; SVC, superior vena cava.

PICCLineNet - Initial Results

Imaging: Detection of correct PICC tip positions from radiology images [2]



Segmentation of PICC lines and tip localization with deep learning - TBA...

References:


  1. M. Shah, D. Shu, V. B. S. Prasath, Y. Ni, A. Schapiro, K. Dufendach. Machine learning for detection of correct peripherally inserted central catheter tip position from radiology reports in infants. Applied Clinical Informatics, 12(04), 856-863, August 2021. doi:10.1055/s-0041-1735178

  2. Prasath et al., PICCLineNet: Detecting peripherally inserted central catheter (PICC) lines and tips from radiological images. Submitted, 2021.