White matter injury detection based on preterm infant cranial ultrasound images

Abstract

IntroductionWhite matter injury (WMI) is now the major disease that seriously affects the quality of life of preterm infants and causes cerebral palsy of children, which also causes periventricular leuko-malacia (PVL) in severe cases. The study aimed to develop a method based on cranial ultrasound images to evaluate the risk of WMI.MethodsThis study proposed an ultrasound radiomics diagnostic system to predict the WMI risk. A multi-task deep learning model was used to segment white matter and predict the WMI risk simultaneously. In total, 158 preterm infants with 807 cranial ultrasound images were enrolled. WMI occurred in 32preterm infants (20.3%, 32/158).ResultsUltrasound radiomics diagnostic system implemented a great result with AUC of 0.845 in the testing set. Meanwhile, multi-task deep learning model preformed a promising result both in segmentation of white matter with a Dice coefficient of 0.78 and prediction of WMI risk with AUC of 0.863 in the testing cohort.DiscussionIn this study, we presented a data-driven diagnostic system for white matter injury in preterm infants. The system combined multi-task deep learning and traditional radiomics features to achieve automatic detection of white matter regions on the one hand, and design a fusion strategy of deep learning features and manual radiomics features on the other hand to obtain stable and efficient diagnostic performance.

A Multiplanar Radiomics Model Based on Cranial Ultrasound to Predict the White Matter Injury in Premature Infants and an Analysis of its Correlation With Neurodevelopment

“…However, the advantages of ultrasound, such as convenience, inexpensiveness, bedside examination, and easy dynamic monitoring, make it more suitable for early diagnosis and follow-up of WMI in preterm infants. 4,5 At present, ultrasonography relies on a visual perspective for the assessment of white matter lesions in premature babies, which lacks objectivity and consistency in diagnostic results, and loses much of the detailed information in the images that cannot be identified by unaided eyes. Some studies have indicated that the diagnostic performance of ultrasound for WMI is inferior to that of MRI.…”

“…Although MRI is clear and accurate, it has the disadvantages of high noise, long time, high price, difficult popularization, and inability to carry out real‐time bedside examination. However, the advantages of ultrasound, such as convenience, inexpensiveness, bedside examination, and easy dynamic monitoring, make it more suitable for early diagnosis and follow‐up of WMI in preterm infants 4,5 . At present, ultrasonography relies on a visual perspective for the assessment of white matter lesions in premature babies, which lacks objectivity and consistency in diagnostic results, and loses much of the detailed information in the images that cannot be identified by unaided eyes.…”

Neurosonography: Shaping the future of neuroprotection strategies in extremely preterm infants

“…The study segmented images of 158 preterm infants (32 of whom had WMI), and the platform performed well in both diagnostic performance and white matter segmentation. This could have been achieved by MR imaging in the past, but now it can be achieved by AI analysis of ultrasound images, providing a new plan for the diagnosis of WMI in preterm infants [ 25 ].…”

“…For detecting white matter injury in preterm infants, Zhu et al [ 25 ] employed a multi-task deep learning model to simultaneously perform white matter segmentation and predict white matter injury risk. They included 807 cranial ultrasound images from 158 preterm infants.…”