Concurrent Patient Evidence for Benchmarking in Radiomics
-
By
-
Wenze Kan
-
Haoliang Pei
-
Mingxin Zhang
-
October 6, 2026
Clinical Report: Concurrent Patient Evidence for Benchmarking in Radiomics
Background
The use of radiomics-based AI in oncology is gaining attention. Understanding the limitations of these models is crucial, particularly in the context of neoadjuvant therapies for NSCLC, where timely and accurate predictions can influence surgical decisions and patient outcomes.
Data Highlights
| Parameter | AI Sensitivity | AI Specificity | Conventional Criteria Sensitivity | Conventional Criteria Specificity |
|---|---|---|---|---|
| Pathological Response | 0.77 | 0.79 | 0.42 | 0.97 |
Key Findings
- The meta-analysis included 16 AI validation datasets with 1031 patients.
- AI sensitivity was reported at 0.77, while specificity was 0.79.
- Conventional criteria showed lower sensitivity (0.42) but higher specificity (0.97).
- Patients not undergoing surgery due to disease progression were excluded from the analysis.
- Misinterpretation of imaging results can lead to false negatives in assessing pathological responses.
Clinical Implications
Clinicians should be cautious when interpreting AI predictions for pathological responses, particularly in the context of surgical decision-making.
Conclusion
The findings highlight the need for improved study designs that incorporate concurrent patient data.
Related Resources & Content
- Jiang Z, Xu Y, Jia S, Liu H, J Med Internet Res, 2026 -- Radiomics-based AI for predicting neoadjuvant immunochemotherapy pathological response in non-small cell lung cancer: systematic review and meta-analysis
- Deng Y, Zhang X, Hu F, Lan X, Eur J Nucl Med Mol Imaging, 2025 -- Quantitative 18F-FDG PET/CT model for predicting pathological complete response to neoadjuvant immunochemotherapy in NSCLC: comparison with RECIST 1.1 and PERCIST
- Forde PM, Spicer J, Lu S, et al, N Engl J Med, 2022 -- Neoadjuvant nivolumab plus chemotherapy in resectable lung cancer
- ASCO Publications — A Framework for Harmonization of Radiomics Data for Multicenter Studies and Clinical Trials
- ASCO Publications — Cancer Radiomic and Perfusion Imaging Automated Framework: Validation on Musculoskeletal Tumors
- ASCO Publications — Cancer Radiomic and Perfusion Imaging Automated Framework: Validation on Musculoskeletal Tumors
- European Radiology — Key Recommendations for Radiomics Practice from the European Society of Medical Imaging Informatics
- Management of Stage III Non–Small Cell Lung Cancer: ASCO Guideline Rapid Recommendation Update
- A Framework for Harmonization of Radiomics Data for Multicenter Studies and Clinical Trials
- Radiomics-Based AI for Predicting Neoadjuvant Immunochemotherapy Pathological Response in Non–Small Cell Lung Cancer: Systematic Review and Meta-Analysis - PMC
- The Image Biomarker Standardization Initiative: Standardized Convolutional Filters for Reproducible Radiomics and Enhanced Clinical Insights | Radiology
Based on findings from:
Same-Patient, Same–Time Point Evidence for Radiomics Benchmarking
Wenze Kan, Haoliang Pei, Mingxin Zhang. Journal Of Medical Internet Research, 2026.
https://www.jmir.org/2026/1/e110058
This content is an AI-generated, fully rewritten summary based on a published scholarly article. It does not reproduce the original text and is not a substitute for the original publication. Readers are encouraged to consult the source for full context, data, and methodology.
