uFlowAM: an unsupervised framework for detection and visualization of abnormal intracardiac microflow on early-pregnancy fetal cardiac microflow imaging - Takeaways - MDSpire

uFlowAM: an unsupervised framework for detection and visualization of abnormal intracardiac microflow on early-pregnancy fetal cardiac microflow imaging

  • By

  • Yan Xia

  • Yarui Wei

  • Zhanru Lan

  • Ning Wu

  • Yongxin Li

  • Xueqin Ji

  • July 9, 2026

  • 0 min

Share

  • 1

    uFlowAM is an unsupervised approach designed to detect and visualize abnormal intracardiac microflow in early-pregnancy fetal cardiac imaging.

  • 2

    The study analyzed 650 early-pregnancy fetal cardiac microflow imaging examinations from fetuses referred for suspected congenital heart disease.

  • 3

    uFlowAM achieved an AUC of 0.94 in the internal cohort and 0.92 in the external cohort, demonstrating high diagnostic accuracy.

  • 4

    Reader-level AUC improved from 0.85 to 0.92 with uFlowAM assistance, indicating enhanced consistency in identifying microflow abnormalities.

  • 5

    uFlowAM is intended as an assistive tool for fetal congenital heart disease assessment, not as a replacement for complete echocardiographic examination.

Original Source(s)

uFlowAM: an unsupervised framework for detection and visualization of abnormal intracardiac microflow on early-pregnancy fetal cardiac microflow imaging

  • by Yan Xia, Yarui Wei, Zhanru Lan, Ning Wu, Yongxin Li, Xueqin Ji

  • July 9, 2026

Related Content

Clinical value of first-morning urine exfoliated cell HPV detection in the diagnosis of high-grade cervical lesions

The hidden dangers of cerebral venous sinus thrombosis: insights into maternal morbidity and mortality during pregnancy

Phthalates Linked to Shorter Gestation

Researchers linked several pregnancy urinary biomarkers—especially plasticizer and combustion-related chemical metabolites—to small shifts in gestational age and fetal growth measures in the ECHO Cohort.