focalinterest.net

AI Summary of Scholarly Research

This page presents an AI-generated summary of a published research paper. The original authors did not write or review this article. [See full disclosure ↓]

MRI-based model assessed sarcoma grade and Ki-67 expression

Written by

editor@focalinterest.com

in

Research area:medicine-clinicaloncology

What the study found

The study found that an automated MRI-based clinical-radiomics model may assess soft tissue sarcoma grade and Ki-67 expression, which are pathological measures used to describe tumor aggressiveness and cell proliferation. It also found that the model may improve the diagnostic performance of less-experienced radiologists.

Why the authors say this matters

The authors say this matters because histological grade and Ki-67 expression are prognostic risk factors in soft tissue sarcoma, and these assessments usually require biopsy, which is invasive and may be affected by tumor heterogeneity. The study suggests that an MRI-based approach could provide a noninvasive alternative for assessment.

What the researchers tested

The researchers conducted a retrospective study of 186 patients with pathologically confirmed soft tissue sarcoma from three hospitals. They developed an automatic segmentation model and compared it with manual segmentations, then built clinical-imaging signature models using structural MRI radiomics, structural MRI plus apparent diffusion coefficient (ADC) radiomics, and those features combined with clinical information and MRI semantic features.

What worked and what didn't

The segmentation model showed good performance, with Dice coefficients of 0.80 for extremity cases and 0.73 for trunk cases. In validation, the best model for grade was the logistic regression clinical-imaging signature model, and the best model for Ki-67 expression was the support vector machine model, with AUCs of 0.846 and 0.742, respectively. Using the model improved diagnostic performance for the two less-experienced radiologists, but the abstract does not report a comparable improvement for the most experienced radiologist.

What to keep in mind

The study was retrospective and included 186 patients, so the findings are based on a specific multicenter sample. The abstract does not describe limitations beyond the reported validation design, and it does not state how the model would perform outside the studied hospitals or in other patient groups.

Key points

  • The study evaluated an automated MRI-based clinical-radiomics pipeline for soft tissue sarcoma grade and Ki-67 expression.
  • Soft tissue sarcoma grade and Ki-67 expression are described as prognostic risk factors and usually require biopsy.
  • The segmentation model achieved Dice coefficients of 0.80 in extremity cases and 0.73 in trunk cases.
  • The best validation AUCs were 0.846 for grade and 0.742 for Ki-67 expression.
  • Model use improved diagnostic performance for the two less-experienced radiologists.

Disclosure

Research title:
MRI-based model assessed sarcoma grade and Ki-67 expression
Authors:
Jinge Li, Yifeng Zhu, Honghai Chen, Lina Zhang, Xiangwen Li, Jie Zhou, Kai Zhang, Jie Huang, X L Yang, Jiaye Zhang, Y I F E I Li, Wenjia Wang, Juan Tao, Shaowu Wang
Institutions:
Affiliated Hospital of Hangzhou Normal University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Medical University, Dalian Municipal Central Hospital, First Affiliated Hospital of Dalian Medical University, Fudan University, General Electric (Spain), Huashan Hospital, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University, Second Affiliated Hospital of Dalian Medical University
Publication date:
2026-06-30
DOI:
10.1002/jmri.70424
OpenAlex record:
View
AI provenance: This post was generated by gpt-5.4-mini (OpenAI). The original authors did not write or review this post.

More posts