EUVIMEDEUVIMED entdecken
ValidierungsstudieQuerschnittstudie

From 2D to 3D: Automated ultrasound segmentation and cross-sectional validation in murine tumor models

Weronika Smolak-Dyżewska, Jerzy Bazak, Wiktoria Brandys, Aleksandra Bienia, Aleksandra Murzyn, Bartosz Płóciennik, Gniewosz Drwięga, Julia Kozik, Agnieszka Drzał, Bartosz Leszczyński, Przemysław Spurek, Martyna Elas und 1 weitere

Computer Methods and Programs in Biomedicine · 2026-08 · 282 · 109403

DOI: 10.1016/j.cmpb.2026.109403 ↗

Kurz gefasst

Validierungsstudie in der Zeitschrift Computer Methods and Programs in Biomedicine, 2026-08. Verfasst von Weronika Smolak-Dyżewska und 12 weiteren Autor:innen. Beteiligte Einrichtung: Jagiellonian University, Faculty of Mathematics and Computer Science, Prof. S. Łojasiewicza 6 St., 30-348 Krakow, Poland; Jagiellonian University, Do…. Hauptthemen laut MEDLINE: Imaging, Three-Dimensional, Neoplasms. Die Arbeit nennt 32 Literaturangaben. Gefördert von: National Centre for Research and Development, Narodowe Centrum Nauki.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND OBJECTIVE: Ultrasound (US) is a widely used method for non-invasive tumor monitoring. Semantic segmentation of tumors in US imagery is a necessary step to reconstruct 3D geometry of a region of interest (ROI). Still, the segmentation task remains challenging due to variable signal-to-noise ratio (SNR) and modest soft-tissue contrast in US imaging. Our objective was to generate a new dataset of murine tumors and present a standardized pipeline for 3D volume reconstruction, suitable for preclinical research. METHODS: Human LN229 and murine 4T1, PanO2, B16 and LLC tumors were imaged in vivo using high-frequency US systems (Vevo F2, Vevo 2100). Our dataset comprised 3442 images, including expert-curated masks and a challenging out-of-distribution (OOD) test set. We evaluated U-Net, Res U-Net, Attention U-Net, and R2AU-Net, with and without autoencoder pretraining on unlabeled frames. For volumetry, 2D masks were converted to voxel grids using known imaging geometry; through-plane interpolation and Marching Cubes surface extraction enabled shape-agnostic 3D volume computation. We compared US-derived volumes with micro-CT, calipers, mold-based, and tumor weight. RESULTS: Across random train-test splits, all models achieved Dice > 0.90. On the subject-independent special testing dataset, performance decreased, indicating limited generalization under distribution shift; the pretrained Attention Res U-Net achieved the highest overlap (Dice 0.750, IoU 0.604), while the pretrained Attention U-Net also remained comparatively robust (Dice 0.731). The 3D reconstruction pipeline produced consistent longitudinal volumes, and cross-modal comparison showed that US-based volumes agreed with micro-CT and physical measurements. CONCLUSIONS: This study presents a standardized workflow for automated tumor segmentation and 3D ultrasound-based volumetry, enabling more objective and reproducible assessment of tumor burden in preclinical oncology studies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Imaging, Three-DimensionalNeoplasmsAnimalsMiceUltrasonographyHumansAlgorithmsReproducibility of ResultsSignal-To-Noise RatioCell Line, TumorDisease Models, AnimalAutoencoderFemaleImage Processing, Computer-AssistedAutomation

Stichwörter der Autor:innen

2D segmentation3D reconstructionMurine tumor segmentationUltrasound

Autor:innen und Institutionen

  • Jagiellonian University, Faculty of Mathematics and Computer Science, Prof. S. Łojasiewicza 6 St., 30-348 Krakow, Poland; Jagiellonian University, Doctoral School of Exact and Natural Sciences, Prof. S. Łojasiewicza 11…
  • Jagiellonian University, Faculty of Biochemistry, Biophysics and Biotechnology, Department of Computational Biophysics and Bioinformatics, Gronostajowa 7 St., 30-387 Krakow, Poland.
  • Jagiellonian University, Faculty of Biochemistry, Biophysics and Biotechnology, Department of Computational Biophysics and Bioinformatics, Gronostajowa 7 St., 30-387 Krakow, Poland; Jagiellonian University, Faculty of B…
  • Jagiellonian University, Doctoral School of Exact and Natural Sciences, Prof. S. Łojasiewicza 11 St., 30-348 Krakow, Poland; Jagiellonian University, Faculty of Biochemistry, Biophysics and Biotechnology, Department of…
  • Jagiellonian University, Faculty of Biochemistry, Biophysics and Biotechnology, Department of Biophysics and Cancer Biology, Gronostajowa 7 St., 30-387 Krakow, Poland.
  • Jagiellonian University, Department of Medical Physics, Marian Smoluchowski Institute of Physics, Łojasiewicza 11 St., 30-348 Krakow, Poland.

Förderung

  • National Centre for Research and Development
  • Narodowe Centrum Nauki

Zitieren

APA

Smolak-Dyżewska, W., Bazak, J., Brandys, W., Bienia, A., Murzyn, A., Płóciennik, B., Drwięga, G., Kozik, J., Drzał, A., Leszczyński, B., Spurek, P., Elas, M., & Krzykawska-Serda, M. (2026). From 2D to 3D: Automated ultrasound segmentation and cross-sectional validation in murine tumor models. Computer Methods and Programs in Biomedicine, 282, 109403. https://doi.org/10.1016/j.cmpb.2026.109403

Vancouver

Smolak-Dyżewska W, Bazak J, Brandys W, Bienia A, Murzyn A, Płóciennik B, et al. From 2D to 3D: Automated ultrasound segmentation and cross-sectional validation in murine tumor models. Computer Methods and Programs in Biomedicine. 2026;282:109403. doi:10.1016/j.cmpb.2026.109403

BibTeX
@article{SmolakDyewska2026,
  author = {Smolak-Dyżewska, Weronika and Bazak, Jerzy and Brandys, Wiktoria and Bienia, Aleksandra and Murzyn, Aleksandra and Płóciennik, Bartosz and Drwięga, Gniewosz and Kozik, Julia and Drzał, Agnieszka and Leszczyński, Bartosz and Spurek, Przemysław and Elas, Martyna and Krzykawska-Serda, Martyna},
  title = {From 2D to 3D: Automated ultrasound segmentation and cross-sectional validation in murine tumor models},
  journal = {Computer Methods and Programs in Biomedicine},
  year = {2026},
  volume = {282},
  pages = {109403},
  doi = {10.1016/j.cmpb.2026.109403}
}
Kostenlos starten →