Radiomics-Based Deep Learning Prediction of Overall Survival in Non-Small-Cell Lung Cancer Using Contrast-Enhanced Computed Tomography

Hou, Kuei-Yuan; Chen, Jyun-Ru; Wang, Yung-Chen; Chiu, Ming-Huang; Lin, Sen-Ping; Mo, Yuan-Heng; Peng, Shih-Chieh; Lu, Chia‐Feng

doi:10.3390/cancers14153798

Cited by 26 publications

(13 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the whole population, we obtained cross-validation C-indexes of 0.79, 0.63, and 0.80 for the radiomic, clinical, and clinical-radiomic models. These results are quite encouraging in comparison to previous studies, reporting sometimes lower validation C-indexes [55,56] that reached 0.75 applying a deep-learning methodology [57]. Moreover, we showed that the radiomic score was an independent predictor of OS also in homogeneous populations according to the presence or absence of specific driver gene alteration, as previously reported [15,17].…”

Section: Discussionsupporting

confidence: 88%

Association between Contrast-Enhanced Computed Tomography Radiomic Features, Genomic Alterations and Prognosis in Advanced Lung Adenocarcinoma Patients

Rinaldi,

Guerini Rocco,

Spitaleri

et al. 2023

Cancers

View full text Add to dashboard Cite

Non-invasive methods to assess mutational status, as well as novel prognostic biomarkers, are warranted to foster therapy personalization of patients with advanced non-small cell lung cancer (NSCLC). This study investigated the association of contrast-enhanced Computed Tomography (CT) radiomic features of lung adenocarcinoma lesions, alone or integrated with clinical parameters, with tumor mutational status (EGFR, KRAS, ALK alterations) and Overall Survival (OS). In total, 261 retrospective and 48 prospective patients were enrolled. A Radiomic Score (RS) was created with LASSO-Logistic regression models to predict mutational status. Radiomic, clinical and clinical-radiomic models were trained on retrospective data and tested (Area Under the Curve, AUC) on prospective data. OS prediction models were trained and tested on retrospective data with internal cross-validation (C-index). RS significantly predicted each alteration at training (radiomic and clinical-radiomic AUC 0.95–0.98); validation performance was good for EGFR (AUC 0.86), moderate for KRAS and ALK (AUC 0.61–0.65). RS was also associated with OS at univariate and multivariable analysis, in the latter with stage and type of treatment. The validation C-index was 0.63, 0.79, and 0.80 for clinical, radiomic, and clinical-radiomic models. The study supports the potential role of CT radiomics for non-invasive identification of gene alterations and prognosis prediction in patients with advanced lung adenocarcinoma, to be confirmed with independent studies.

show abstract

Section: Discussionsupporting

confidence: 88%

Association between Contrast-Enhanced Computed Tomography Radiomic Features, Genomic Alterations and Prognosis in Advanced Lung Adenocarcinoma Patients

Rinaldi,

Guerini Rocco,

Spitaleri

et al. 2023

Cancers

View full text Add to dashboard Cite

show abstract

“…Hou and colleagues [25] constructed a deep learning model with five hidden layers (20-26-32-26-20 architecture) to predict overall survival in patients with pathologically proven NSCLC at all stages. The input layer consisted of eight radiomic features (based on pre-treatment contrast-enhanced chest CT) and five clinical features.…”

Section: Survival Assessmentmentioning

confidence: 99%

Artificial Neural Networks in Lung Cancer Research: A Narrative Review

Prisciandaro

Sedda

Cara

et al. 2023

JCM

View full text Add to dashboard Cite

Background: Artificial neural networks are statistical methods that mimic complex neural connections, simulating the learning dynamics of the human brain. They play a fundamental role in clinical decision-making, although their success depends on good integration with clinical protocols. When applied to lung cancer research, artificial neural networks do not aim to be biologically realistic, but rather to provide efficient models for nonlinear regression or classification. Methods: We conducted a comprehensive search of EMBASE (via Ovid), MEDLINE (via PubMed), Cochrane CENTRAL, and Google Scholar from April 2018 to December 2022, using a combination of keywords and related terms for “artificial neural network”, “lung cancer”, “non-small cell lung cancer”, “diagnosis”, and “treatment”. Results: Artificial neural networks have shown excellent aptitude in learning the relationships between the input/output mapping from a given dataset, without any prior information or assumptions about the statistical distribution of the data. They can simultaneously process numerous variables, managing complexity; hence, they have found broad application in tasks requiring attention. Conclusions: Lung cancer is the most common and lethal form of tumor, with limited diagnostic and treatment methods. The advances in tailored medicine have led to the development of novel tools for diagnosis and treatment. Artificial neural networks can provide valuable support for both basic research and clinical decision-making. Therefore, tight cooperation among surgeons, oncologists, and biostatisticians appears mandatory.

show abstract

“…Radiologists and nuclear medicine doctors see radiomics as having the potential to provide a quantitative signature of tumors, that are often impossible to be detected by human experts [23]. Therefore, radiomics features have been used in the diagnosis and prognosis of multiple cancer types, including those of the breast [24,25], prostate [26], lung [27], head and neck [28], rectal [29], and others. Numerous studies have validated the predictive power of IBSI radiomics features for generalization to multiple cancer types.…”

Section: Related Workmentioning

confidence: 99%

Integration of Radiomics and Tumor Biomarkers in Interpretable Machine Learning Models

Brocki

Chung

2023

Cancers

View full text Add to dashboard Cite

Despite the unprecedented performance of deep neural networks (DNNs) in computer vision, their clinical application in the diagnosis and prognosis of cancer using medical imaging has been limited. One of the critical challenges for integrating diagnostic DNNs into radiological and oncological applications is their lack of interpretability, preventing clinicians from understanding the model predictions. Therefore, we studied and propose the integration of expert-derived radiomics and DNN-predicted biomarkers in interpretable classifiers, which we refer to as ConRad, for computerized tomography (CT) scans of lung cancer. Importantly, the tumor biomarkers can be predicted from a concept bottleneck model (CBM) such that once trained, our ConRad models do not require labor-intensive and time-consuming biomarkers. In our evaluation and practical application, the only input to ConRad is a segmented CT scan. The proposed model was compared to convolutional neural networks (CNNs) which act as a black box classifier. We further investigated and evaluated all combinations of radiomics, predicted biomarkers and CNN features in five different classifiers. We found the ConRad models using nonlinear SVM and the logistic regression with the Lasso outperformed the others in five-fold cross-validation, with the interpretability of ConRad being its primary advantage. The Lasso is used for feature selection, which substantially reduces the number of nonzero weights while increasing the accuracy. Overall, the proposed ConRad model combines CBM-derived biomarkers and radiomics features in an interpretable ML model which demonstrates excellent performance for lung nodule malignancy classification.

show abstract

Radiomics-Based Deep Learning Prediction of Overall Survival in Non-Small-Cell Lung Cancer Using Contrast-Enhanced Computed Tomography

Cited by 26 publications

References 34 publications

Association between Contrast-Enhanced Computed Tomography Radiomic Features, Genomic Alterations and Prognosis in Advanced Lung Adenocarcinoma Patients

Association between Contrast-Enhanced Computed Tomography Radiomic Features, Genomic Alterations and Prognosis in Advanced Lung Adenocarcinoma Patients

Artificial Neural Networks in Lung Cancer Research: A Narrative Review

Integration of Radiomics and Tumor Biomarkers in Interpretable Machine Learning Models

Contact Info

Product

Resources

About