An intelligent fusion model for Ki-67 prediction in non-small cell lung cancer: A cloud-based prediction system integrating radiomics.
2/5 보강
OpenAlex 토픽 ·
Radiomics and Machine Learning in Medical Imaging
Lung Cancer Diagnosis and Treatment
Effects of Radiation Exposure
[BACKGROUND] The expression level of Ki-67 affects the prognosis of NSCLC patients.
- 표본수 (n) 525
- 95% CI 0.97-0.99
APA
zhenyu cao, Xiaoling Xu, et al. (2026). An intelligent fusion model for Ki-67 prediction in non-small cell lung cancer: A cloud-based prediction system integrating radiomics.. European journal of radiology, 200, 112866. https://doi.org/10.1016/j.ejrad.2026.112866
MLA
zhenyu cao, et al.. "An intelligent fusion model for Ki-67 prediction in non-small cell lung cancer: A cloud-based prediction system integrating radiomics.." European journal of radiology, vol. 200, 2026, pp. 112866.
PMID
42000471 ↗
Abstract 한글 요약
[BACKGROUND] The expression level of Ki-67 affects the prognosis of NSCLC patients.
[OBJECTIVE] Accurate preoperative prediction of Ki-67 expression in non-small cell lung cancer (NSCLC) is crucial for prognostic stratification.
[METHODS] This multicenter retrospective study enrolled 876 NSCLC patients (January 2015-December 2024) from four institutions, randomly divided into training (n = 525), testing (n = 175), and external validation (n = 176) sets. Radiomic features were extracted from intratumoral and peritumoral (0-12 mm) regions on CT images to construct intra-, peri-, and combined (intra + peri) radiomic scores (Rad-score). Deep learning scores (DL-score) were generated using ResNet101 for whole-lung and tumor-specific analyses. A random forest model integrating Rad-scores, DL-scores, and clinical parameters (lobulation, emphysema, etc.) was developed and validated across all datasets.
[RESULTS] The combined model (intra + peri Rad-score, intra-tumor DL-score, and clinical features) achieved AUCs of 0.98 (95% CI: 0.97-0.99), 0.92 (0.88-0.96), and 0.92 (0.87-0.96) in training, testing, and external validation sets, with corresponding F1-scores of 0.90, 0.75, and 0.70. SHAP interpretation identified intra-tumor DL-score as the most significant predictor (feature contribution: 46.8%).
[CONCLUSION] The multimodal random forest model enables noninvasive and accurate Ki-67 prediction in NSCLC, demonstrating superior generalizability and interpretability to guide personalized therapeutic strategies.
[CLINICAL RELEVANCE] Integrating deep learning with intratumoral and peritumoral radiomics enhances the preoperative prediction of Ki-67 expression in patients with non-small cell lung cancer.
[OBJECTIVE] Accurate preoperative prediction of Ki-67 expression in non-small cell lung cancer (NSCLC) is crucial for prognostic stratification.
[METHODS] This multicenter retrospective study enrolled 876 NSCLC patients (January 2015-December 2024) from four institutions, randomly divided into training (n = 525), testing (n = 175), and external validation (n = 176) sets. Radiomic features were extracted from intratumoral and peritumoral (0-12 mm) regions on CT images to construct intra-, peri-, and combined (intra + peri) radiomic scores (Rad-score). Deep learning scores (DL-score) were generated using ResNet101 for whole-lung and tumor-specific analyses. A random forest model integrating Rad-scores, DL-scores, and clinical parameters (lobulation, emphysema, etc.) was developed and validated across all datasets.
[RESULTS] The combined model (intra + peri Rad-score, intra-tumor DL-score, and clinical features) achieved AUCs of 0.98 (95% CI: 0.97-0.99), 0.92 (0.88-0.96), and 0.92 (0.87-0.96) in training, testing, and external validation sets, with corresponding F1-scores of 0.90, 0.75, and 0.70. SHAP interpretation identified intra-tumor DL-score as the most significant predictor (feature contribution: 46.8%).
[CONCLUSION] The multimodal random forest model enables noninvasive and accurate Ki-67 prediction in NSCLC, demonstrating superior generalizability and interpretability to guide personalized therapeutic strategies.
[CLINICAL RELEVANCE] Integrating deep learning with intratumoral and peritumoral radiomics enhances the preoperative prediction of Ki-67 expression in patients with non-small cell lung cancer.
🏷️ 키워드 / MeSH 📖 같은 키워드 OA만
같은 제1저자의 인용 많은 논문 (5)
- CRISPR-based functional genomics for dissecting therapeutic dependency in primary acute myeloid leukemia samples.
- Study on the mechanism of PSENEN in modulating cisplatin resistance in NSCLC through stromal cell interactions.
- Isoquinoline alkaloids with bioactive constituents from the roots and rhizomes of .
- Predictive Factors and Nomogram for Malignant Pulmonary Nodules (≤ 1 cm).
- Targeting metabolic dysregulation in HCC: Development and validation of a dual-biomarker diagnostic model integrating ADH1C and BHMT.
🏷️ 같은 키워드 · 무료전문 — 이 논문 MeSH/keyword 기반
- Nanotechnology-Assisted Molecular Profiling: Emerging Advances in Circulating Tumor DNA Detection.
- Building Hybrid Pharmacometric-Machine Learning Models in Oncology Drug Development: Current State and Recommendations.
- Role of Cytokines in Oligometastatic Non-Small-Cell Lung Cancer Treated with Stereotactic Radiation Therapy: An Observational Pilot Study.
- Development and validation of a novel nomogram model for predicting postoperative survival of T4N0M0 NSCLC: a population-based survival analysis.
- Machine learning integrating MRI and clinical features predicts early recurrence of hepatocellular carcinoma after resection.
- Cost comparison of osimertinib plus platinum-pemetrexed versus amivantamab plus lazertinib for the first-line treatment of patients with locally advanced or metastatic epidermal growth factor receptor-mutated non-small cell lung cancer.