Semantic CT features and differentiation model: new primary lung cancer versus metastasis after previous malignancy - Summary - MDSpire

Semantic CT features and differentiation model: new primary lung cancer versus metastasis after previous malignancy

  • By

  • Hardeep Singh Kalsi

  • Kristofer Linton-Reid

  • Changhyun Kim

  • Mitchell Chen

  • Victoria Crowe

  • Esubalew Alemu

  • Samir Mahboobani

  • David Gibeon

  • Alexander Procter

  • Mohsen Hajhosseiny

  • Cara Owens

  • Emily C. Bartlett

  • Nuria Porta

  • Thesha Thavaraja

  • Simon Doran

  • Anand Devaraj

  • Bhupinder Sharma

  • Arjun Nair

  • Eric O. Aboagye

  • Richard W. Lee

  • May 6, 2026

  • 0 min

Share

Objective:

To evaluate the role of semantic CT imaging features in differentiating between second primary lung cancer and lung metastasis in patients with a history of prior malignancies, emphasizing the clinical significance of this differentiation.

Key Findings:
  • Morphological characteristics such as spiculation and lobulation were significant in differentiating SPLC from LM, with statistical significance noted.
  • Irregular contours were more common in primary lung cancer, while smooth borders were associated with metastasis, indicating distinct imaging profiles.
  • The presence of emphysema and feeding vessels showed varying correlations with malignancy, warranting further investigation.
Interpretation:

Structured reporting of semantic CT features enhances the accuracy of distinguishing between new primary lung cancer and metastatic disease, addressing a critical diagnostic challenge in cancer survivors and potentially improving patient outcomes.

Limitations:
  • The study is retrospective and relies on existing imaging data, which may introduce bias, particularly in case selection.
  • Variability in radiologist interpretation could affect consistency in lesion classification, impacting diagnostic reliability.
Conclusion:

Utilizing semantic CT features can improve diagnostic stratification of lung lesions in patients with a history of cancer, potentially guiding treatment decisions more effectively by providing clearer imaging criteria.

Original Source(s)

Semantic CT features and differentiation model: new primary lung cancer versus metastasis after previous malignancy

  • by Hardeep Singh Kalsi, Kristofer Linton-Reid, Changhyun Kim, Mitchell Chen, Victoria Crowe, Esubalew Alemu, Samir Mahboobani, David Gibeon, Alexander Procter, Mohsen Hajhosseiny, Cara Owens, Emily C. Bartlett, Nuria Porta, Thesha Thavaraja, Simon Doran, Anand Devaraj, Bhupinder Sharma, Arjun Nair, Eric O. Aboagye, Richard W. Lee

  • May 6, 2026

Related Content

Diffuse large B-cell lymphoma presenting as a scalp tumor with skull destruction and neurological symptoms: a case report

Data requirements of an informative, directive, and preventive mobile health application for gastric cancer

Percutaneous nephrostomy as a marker of clinical vulnerability in non-metastatic muscle-invasive bladder cancer: prognostic and infectious implications