A Multi-Model, Pixel-Native Framework for Automated Computed Tomography Series Labeling and Characterization: Proof-of-Concept Study - Scorecard - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

A Comprehensive Pixel-Based Approach for the Automated Labeling and Analysis of Computed Tomography Series: A Proof-of-Concept Investigation

  • By

  • Yutong Wen

  • Anton Sheahan Quinsten

  • Cynthia Sabrina Schmidt

  • Christian Bojahr

  • Judith Kohnke

  • Kamyar Arzideh

  • Sina Warmer

  • Sebastian Blex

  • Ann-Christin Jacoby

  • Max Eberts

  • Hanna Lehmann

  • Olivia Barbara Pollok

  • Mathias Holtkamp

  • Luca Salhöfer

  • Lale Umutlu

  • Michael Forsting

  • Johannes Haubold

  • Felix Nensa

  • Katarzyna Borys

  • René Hosch

  • September 29, 2026

Share

Clinical Scorecard: A Comprehensive Pixel-Based Approach for the Automated Labeling and Analysis of Computed Tomography Series: A Proof-of-Concept Investigation

At a Glance

CategoryDetail
ConditionComputed Tomography Imaging
Key MechanismsAutomated labeling and analysis using AI and deep learning techniques.
Target PopulationPatients undergoing CT imaging.
Care SettingRadiology departments utilizing CT imaging.

Key Highlights

  • AI advancements in radiology include segmentation, body composition assessment, and virtual biopsy.
  • Current workflows rely on DICOM metadata, which is often incomplete or ambiguous.
  • Inconsistent metadata entries hinder automated routing of imaging series.
  • Manual curation is often required to select appropriate input series for AI processing.
  • Variability in metadata can complicate research reproducibility and clinical reliability.

Guideline-Based Recommendations

Diagnosis

    Management

      Monitoring & Follow-up

        Risks

        • Inconsistent DICOM metadata can lead to incorrect series being used for AI processing.

        Patient & Prescribing Data

        Patients undergoing various CT examinations.

        Automated methods for image series selection are needed to improve workflow efficiency.

        Clinical Best Practices

        • Standardization of DICOM metadata entries across institutions is recommended.
        • Visual inspection may be necessary for ambiguous series descriptions.

        Related Resources & Content

        Original Source(s)

        A Multi-Model, Pixel-Native Framework for Automated Computed Tomography Series Labeling and Characterization: Proof-of-Concept Study

        • by Yutong Wen, Anton Sheahan Quinsten, Cynthia Sabrina Schmidt, Christian Bojahr, Judith Kohnke, Kamyar Arzideh, Sina Warmer, Sebastian Blex, Ann-Christin Jacoby, Max Eberts, Hanna Lehmann, Olivia Barbara Pollok, Mathias Holtkamp, Luca Salhöfer, Lale Umutlu, Michael Forsting, Johannes Haubold, Felix Nensa, Katarzyna Borys, René Hosch

        • September 29, 2026

        Related Content

        Tele-guided lung ultrasound may be acceptable to paramedics

        Trainees rated remote-guided imaging favorably, while technical reliability and communication emerged as barriers to implementation.

        Reducing time from presentation to diagnosis of scaphoid fractures with cone beam CT: a before-and-after study

        FDA guidance spurs adoption of change control plans for radiology AI devices

        Investigators examined adoption and public documentation of change control plans for FDA-cleared radiology AI devices.