EUVIMEDEUVIMED entdecken
Crossref

Longitudinal deep learning models for tracking disease progression in ovarian cancer using PET/CT imaging and clinical reports

Mohammad Hossein Sadeghi, Sedigheh Sina, Mehrosadat Alavi, Francesco Giammarile, Zahra Nasiri Feshani, Amir Hossein Farshchitabrizi, Zahra Rakeb, Seyed Alireza Mirhosseini

Physical and Engineering Sciences in Medicine · 2026

DOI: 10.1007/s13246-025-01669-0 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physical and Engineering Sciences in Medicine, 2026. Verfasst von Mohammad Hossein Sadeghi und 7 weiteren Autor:innen. Beteiligte Einrichtung: Nuclear Engineering Department, School of Mechanical Engineering, Shiraz University, Shiraz, Iran. Hauptthemen laut MEDLINE: Ovarian Neoplasms, Deep Learning, Positron Emission Tomography Computed Tomography, Disease Progression. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 55 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Ovarian cancer is often diagnosed at advanced stages, with high-grade serous ovarian cancer (HGSOC) accounting for 70-80% of fatalities. Current predictive tools, limited by single-time-point data, fail to capture subtle temporal changes indicative of relapse. To evaluate the performance of OvarXNet, a novel deep learning framework integrating longitudinal PET/CT imaging and clinical data for early prediction of ovarian cancer relapse. This retrospective study included 58 advanced-stage HGSOC patients (mean age, 56 ± 10.4 years) who underwent [18F]FDG PET/CT scans from April 2019 to January 2025. Patients with uncontrolled diabetes or recent cancers were excluded. Each patient had a median of three PET/CT scans and associated clinical data. The OvarXNet framework combines 3D convolutional neural networks (CNNs) for volumetric feature extraction and bidirectional gated recurrent units for temporal analysis. Statistical analyses included area under the receiver operating characteristic curve (AUC), precision-recall (PR) metrics, and calibration plots. Fifty-eight patients (mean age 56 ± 10.4 years) contributed 1914 image sets post-augmentation. OvarXNet achieved an AUC of 0.92, outperforming single-time-point CNN (AUC: 0.84) and LSTM-based models (AUC: 0.89). PR analysis confirmed superior model performance (PR-AUC: OvarXNet > 0.90 vs. single-time-point CNN: 0.82). Calibration plots demonstrated robust probability estimates. Attention mechanisms highlighted time points with elevated CA-125 or progression-related clinical notes, enhancing interpretability. OvarXNet significantly improves early relapse prediction in advanced-stage HGSOC by leveraging longitudinal imaging and clinical data. The framework's accuracy and interpretability support its potential for guiding personalized treatment strategies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Ovarian NeoplasmsDeep LearningPositron Emission Tomography Computed TomographyDisease ProgressionHumansFemaleMiddle AgedRetrospective StudiesAgedLongitudinal StudiesROC CurveImage Processing, Computer-Assisted

Stichwörter der Autor:innen

Bi-GRUDeep learningLongitudinal analysisOvarian cancerPET/CT

Autor:innen und Institutionen

  • Nuclear Engineering Department, School of Mechanical Engineering, Shiraz University, Shiraz, Iran.
  • Nuclear Engineering Department, School of Mechanical Engineering, Shiraz University, Shiraz, Iran. samirasina@shirazu.ac.ir.
  • Radiation Research Center, School of Mechanical Engineering, Shiraz University, Shiraz, Iran. samirasina@shirazu.ac.ir.
  • Ionizing and Non-Ionizing Radiation Protection Research Center, School of Paramedical Sciences, Shiraz University of Medical Sciences, Shiraz, Iran.
  • Nuclear Medicine and Diagnostic Imaging Section, Division of Human Health, International Atomic Energy Agency (IAEA), Vienna, Austria.
  • Nuclear Medicine Department, PET/CT Center, Kowsar Hospital, Shiraz, Iran.

Zitieren

APA

Sadeghi, M. H., Sina, S., Alavi, M., Giammarile, F., Feshani, Z. N., Farshchitabrizi, A. H., Rakeb, Z., & Mirhosseini, S. A. (2026). Longitudinal deep learning models for tracking disease progression in ovarian cancer using PET/CT imaging and clinical reports. Physical and Engineering Sciences in Medicine. https://doi.org/10.1007/s13246-025-01669-0

Vancouver

Sadeghi MH, Sina S, Alavi M, Giammarile F, Feshani ZN, Farshchitabrizi AH, et al. Longitudinal deep learning models for tracking disease progression in ovarian cancer using PET/CT imaging and clinical reports. Physical and Engineering Sciences in Medicine. 2026. doi:10.1007/s13246-025-01669-0

BibTeX
@article{Sadeghi2026,
  author = {Sadeghi, Mohammad Hossein and Sina, Sedigheh and Alavi, Mehrosadat and Giammarile, Francesco and Feshani, Zahra Nasiri and Farshchitabrizi, Amir Hossein and Rakeb, Zahra and Mirhosseini, Seyed Alireza},
  title = {Longitudinal deep learning models for tracking disease progression in ovarian cancer using PET/CT imaging and clinical reports},
  journal = {Physical and Engineering Sciences in Medicine},
  year = {2026},
  doi = {10.1007/s13246-025-01669-0}
}
Kostenlos starten →