Szymon Płotka, Karol Pustelnik, Paula Szenejko, Kinga Żebrowska, Iga Rzucidło-Szymańska, Natalia Szymecka-Samaha, Tomasz Łęgowik, Katarzyna Kosińska-Kaczyńska, Przemysław Korzeniowski, Piotr Biliński, Asma Khalil, Robert Brawura-Biskupski-Samaha, Ivana Išgum, Clara I Sánchez, Arkadiusz Sitek 

In this collaborative study, the authors propose a deep learning-based method aimed at improving the precision and efficiency of fetal biometry in prenatal ultrasound.

The study, titled “Direct estimation of fetal biometry measurements from ultrasound video scans through deep learning,” tackles a long-standing challenge in obstetric imaging: the reliance on skilled professionals to manually identify specific standard anatomical planes—such as those of the fetal head, abdomen, and femur—for measurement. This traditional approach demands expertise, is time-intensive, and often results in inconsistencies due to intra- and interobserver variation.

To address these limitations, the authors developed an end-to-end deep learning framework capable of automatically identifying the required standard planes and directly extracting key biometric measurements from entire ultrasound video sequences. Unlike previous methods that require manual frame selection or partial automation, this approach performs both detection and measurement in a unified pipeline.

Importantly, it is—according to the authors—the first method to estimate the full set of fetal biometry parameters directly from ultrasound video data without human intervention. This advancement has the potential to improve diagnostic consistency and reduce operator dependency, particularly in resource-constrained settings, marking a significant step forward for AI-assisted prenatal diagnostics.

READ HERE

Autors: Szymon Płotka, Karol Pustelnik, Paula Szenejko, Kinga Żebrowska, Iga Rzucidło-Szymańska, Natalia Szymecka-Samaha, Tomasz Łęgowik, Katarzyna Kosińska-Kaczyńska, Przemysław Korzeniowski, Piotr Biliński, Asma Khalil, Robert Brawura-Biskupski-Samaha, Ivana Išgum, Clara I Sánchez, Arkadiusz Sitek 

DOI: 10.1016/j.ajogmf.2025.101623 

Keywords: deep learning in medicine, fetal biometry, ultrasound video analysis, prenatal ultrasound