Development and internal validation of a multimodal deep learning model integrating imaging, clinical and pathological features to predict false-negatives in non-specific benign lung biopsies: a dual-center study.
4 affiliations
- 1Department of Radiology, Daping Hospital, Army Medical University, Chongqing, China.
- 2Department of Nuclear Medicine, Southwest Hospital, Army Medical University, Chongqing, China.
- 3Department of Digital Medicine, School of Biomedical Engineering and Imaging Medicine, Third Military Medical University (Army Medical University), Chongqing, China.
- 4Department of Medical Engineering, The 920th Hospital of Joint Logistics Support Force, Kunming, China.
Abstract
Computed tomography (CT)-guided percutaneous transthoracic needle biopsy (PTNB) is widely used for lung lesion evaluation. Non-specific benign PTNB results create important clinical challenges. These outcomes carry a considerable risk of false-negative malignancy, with reported rates as high as 49%. Missed cancer diagnoses may delay curative-intent treatment. This study developed and internally validated a multimodal deep learning model. The model combines CT imaging, clinical variables, and pathological features. Its goal was to distinguish false-negative malignancy from non-specific benign PTNB results. In this dual-center retrospective study, consecutive patients with non-specific benign PTNB results from two tertiary hospitals (January 2020-December 2022) were screened, and 403 patients (mean age, 57.19±10.2 years; 289 male patients) with complete clinical, imaging, and follow-up data were included. The reference standard for the final diagnosis combined at least 2 years of imaging stability or resolution, surgical pathology, repeat biopsy, or documented clinical progression. Candidate predictors included CT imaging features, clinical variables (demographics, smoking status, tumor markers), and four pathological descriptors. Three models were developed: an imaging model (CT only), an imaging-clinical model (CT and clinical data), and a combined model (CT, clinical data, and pathological features). Patients were randomly split into training (80%) and test (20%) sets. Model performance was evaluated using receiver operating characteristic (ROC) analysis and compared using the DeLong test. Diagnostic accuracy was also compared with that of the four thoracic radiologists (4-15 years of experience). Overall, 90 patients (22.3%) had false-negative malignancy, with 19 events in the test set (n=78). The combined model achieved an area under the curve (AUC) of 0.825 [95% confidence interval (CI): 0.710, 0.923], significantly outperforming the imaging model (AUC 0.655, 95% CI: 0.507, 0.794; P=0.04). It was numerically superior to the imaging-clinical model (AUC, 0.789; 95% CI: 0.667, 0.892; P=0.26), with no statistically significant difference observed between the two models. At the optimal threshold of 0.225, the combined model showed an accuracy of 0.782 (95% CI: 0.678, 0.859), a sensitivity of 0.842 (95% CI: 0.624, 0.945), and a specificity of 0.763 (95% CI: 0.640, 0.853), noting that 75.6% of test cases were true negatives. The combined model's accuracy exceeded that of the four radiologists (range, 58.97-73.08%), although this comparison was descriptive without formal statistical testing. This multimodal model shows promising discriminative performance for predicting false-negative PTNB results. Since performance metrics and the 0.225 threshold were obtained from internal random hold out validation, external independent-cohort validation and threshold optimization are mandatory prior to clinical decision-support application.
Related
- Development and validation of a deep learning model integrating ¹⁸F-FDG PET/CT radiomics and clinical features to predict major pathological response in non-small cell lung cancer patients receiving chemoimmunotherapy.
- A multimodal deep learning radiomics nomogram integrating clinicopathological features for predicting pathological complete response after neoadjuvant therapy in triple-negative breast cancer: a multicenter retrospective study
- Development and validation of a fall risk prediction model integrating objective gait metrics for patients with cerebral small vessel disease: protocol for a two-phase study
- A multimodal optimisation framework for cardiac electrophysiology model personalisation using computed tomography and 12-lead electrocardiogram: Algorithm development and validation
- MRI-Based Radiomics Model for Pretreatment Prediction of Pathological Lymph Node Metastasis in Neuroendocrine Carcinoma of the Cervix: Dual-Center Development and External Validation.
BibTeX
@article{yuan2026development,
title = {{Development and internal validation of a multimodal deep learning model integrating imaging, clinical and pathological features to predict false-negatives in non-specific benign lung biopsies: a dual-center study.}},
author = {Jing Yuan and Wei Li and Chunhua Liu and Luxing He and Qian Yang and Ruili Mao and Xuequan Huang and Weiguo Zhang and Jingjing Liu},
journal = {Translational lung cancer research},
year = {2026},
month = sep,
doi = {10.21037/tlcr-2026-0726},
url = {https://europepmc.org/article/MED/42999642},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Development and internal validation of a multimodal deep learning model integrating imaging, clinical and pathological features to predict false-negatives in non-specific benign lung biopsies: a dual-center study.
AU - Jing Yuan
AU - Wei Li
AU - Chunhua Liu
AU - Luxing He
AU - Qian Yang
AU - Ruili Mao
AU - Xuequan Huang
AU - Weiguo Zhang
AU - Jingjing Liu
PY - 2026
DA - 2026/09/29
JO - Translational lung cancer research
DO - 10.21037/tlcr-2026-0726
UR - https://europepmc.org/article/MED/42999642
AB - Computed tomography (CT)-guided percutaneous transthoracic needle biopsy (PTNB) is widely used for lung lesion evaluation. Non-specific benign PTNB results create important clinical challenges. These outcomes carry a considerable risk of false-negative malignancy, with reported rates as high as 49%. Missed cancer diagnoses may delay curative-intent treatment. This study developed and internally validated a multimodal deep learning model. The model combines CT imaging, clinical variables, and pathological features. Its goal was to distinguish false-negative malignancy from non-specific benign PTNB results. In this dual-center retrospective study, consecutive patients with non-specific benign PTNB results from two tertiary hospitals (January 2020-December 2022) were screened, and 403 patients (mean age, 57.19±10.2 years; 289 male patients) with complete clinical, imaging, and follow-up data were included. The reference standard for the final diagnosis combined at least 2 years of imaging stability or resolution, surgical pathology, repeat biopsy, or documented clinical progression. Candidate predictors included CT imaging features, clinical variables (demographics, smoking status, tumor markers), and four pathological descriptors. Three models were developed: an imaging model (CT only), an imaging-clinical model (CT and clinical data), and a combined model (CT, clinical data, and pathological features). Patients were randomly split into training (80%) and test (20%) sets. Model performance was evaluated using receiver operating characteristic (ROC) analysis and compared using the DeLong test. Diagnostic accuracy was also compared with that of the four thoracic radiologists (4-15 years of experience). Overall, 90 patients (22.3%) had false-negative malignancy, with 19 events in the test set (n=78). The combined model achieved an area under the curve (AUC) of 0.825 [95% confidence interval (CI): 0.710, 0.923], significantly outperforming the imaging model (AUC 0.655, 95% CI: 0.507, 0.794; P=0.04). It was numerically superior to the imaging-clinical model (AUC, 0.789; 95% CI: 0.667, 0.892; P=0.26), with no statistically significant difference observed between the two models. At the optimal threshold of 0.225, the combined model showed an accuracy of 0.782 (95% CI: 0.678, 0.859), a sensitivity of 0.842 (95% CI: 0.624, 0.945), and a specificity of 0.763 (95% CI: 0.640, 0.853), noting that 75.6% of test cases were true negatives. The combined model's accuracy exceeded that of the four radiologists (range, 58.97-73.08%), although this comparison was descriptive without formal statistical testing. This multimodal model shows promising discriminative performance for predicting false-negative PTNB results. Since performance metrics and the 0.225 threshold were obtained from internal random hold out validation, external independent-cohort validation and threshold optimization are mandatory prior to clinical decision-support application.
SN - 2218-6751
SN - 2226-4477
ER -