Purpose: This study aimed to establish and evaluate the value of integrated models involving F-FDG PET/CT-based radiomics and clinicopathological information in the prediction of pathological complete response (pCR) to neoadjuvant therapy (NAT) for non-small cell lung cancer (NSCLC).

Methods: A total of 106 eligible NSCLC patients were included in the study. After volume of interest (VOI) segmentation, 2,016 PET-based and 2,016 CT-based radiomic features were extracted. To select an optimal machine learning model, a total of 25 models were constructed based on five sets of machine learning classifiers combined with five sets of predictive feature resources, including PET-based alone radiomics, CT-based alone radiomics, PET/CT-based radiomics, clinicopathological features, and PET/CT-based radiomics integrated with clinicopathological features. Area under the curves (AUCs) of receiver operator characteristic (ROC) curves were used as the main outcome to assess the model performance.

Results: The hybrid PET/CT-derived radiomic model outperformed PET-alone and CT-alone radiomic models in the prediction of pCR to NAT. Moreover, addition of clinicopathological information further enhanced the predictive performance of PET/CT-derived radiomic model. Ultimately, the support vector machine (SVM)-based PET/CT radiomics combined clinicopathological information presented an optimal predictive efficacy with an AUC of 0.925 (95% CI 0.869-0.981) in the training cohort and an AUC of 0.863 (95% CI 0.740-0.985) in the test cohort. The developed nomogram involving radiomics and pathological type was suggested as a convenient tool to enable clinical application.

Conclusions: The F-FDG PET/CT-based SVM radiomics integrated with clinicopathological information was an optimal model to non-invasively predict pCR to NAC for NSCLC.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11310012PMC
http://dx.doi.org/10.3389/fonc.2024.1425837DOI Listing

Publication Analysis

Top Keywords

pet/ct-based radiomics
12
radiomics
9
prediction pathological
8
pathological complete
8
complete response
8
neoadjuvant therapy
8
non-small cell
8
cell lung
8
lung cancer
8
f-fdg pet/ct-based
8

Similar Publications

Purpose: This study aimed to develop and evaluate a machine learning model combining clinical, radiomics, and deep learning features derived from PET/CT imaging to predict lymph node metastasis (LNM) in patients with non-small cell lung cancer (NSCLC). The model's interpretability was enhanced using Shapley additive explanations (SHAP).

Methods: A total of 248 NSCLC patients who underwent preoperative PET/CT scans were included and divided into training, test, and external validation sets.

View Article and Find Full Text PDF

AI potential in PET/CT cancer imaging.

Hell J Nucl Med

December 2024

Laboratory of Computing, Medical Informatics, and Biomedical-Imaging Technologies, School of Medicine, Aristotle University of Thessaloniki, Box 323, 54124, Thessaloniki, Greece.

Positron emission tomography/computed tomography (PET/CT) is a hybrid medical imaging technique that combines PET and CT to provide detailed images of the body's anatomical structures and metabolic activity. It is frequently used for oncology and other medical diagnoses. This overview aims to examine how artificial intelligence (AI) has been used in PET/CT, based on recent state-of-art.

View Article and Find Full Text PDF
Article Synopsis
  • - The study developed habitat radiomic models to predict overall survival in patients with hepatocellular carcinoma (HCC) by analyzing intratumoral heterogeneity using F-FDG PET/CT images.
  • - Researchers analyzed data from 137 HCC patients, employing k-means clustering to identify intratumoral habitats and extracting 4032 radiomic features, ultimately creating multiple predictive models using machine learning techniques.
  • - The best model, combining radiomic and clinical data, showed improved prediction accuracy for overall survival, achieving a C-index of 0.747 and outperforming other models in long-term survival predictions.
View Article and Find Full Text PDF
Article Synopsis
  • An AI-based model was developed to predict cervical lymph node metastasis (CLNM) in patients with esophageal squamous cell carcinoma (ESCC) using Radiomics and deep learning features from PET/CT images, combined with clinical data.
  • The study involved 300 ESCC patients for training and testing, plus an additional 111 patients for external validation; various machine learning algorithms were tested, with logistic regression being the final model chosen.
  • The final DL-Radiomics-Clinical (DRC) model showed the best performance metrics, including an AUC of 0.955 and high accuracy rates, indicating its potential as a non-invasive tool for predicting CLNM in ESCC patients.
View Article and Find Full Text PDF
Article Synopsis
  • - Head and neck cancers rank as the seventh most common worldwide, and lymph node metastasis (LNM) is a key factor that negatively impacts survival; traditional imaging methods struggle with accurate detection of LNM.
  • - This study systematically searched various databases for research on AI models diagnosing LNM in head and neck cancers, finding 23 relevant articles that primarily focused on internal validation due to a lack of external validation.
  • - The analysis showed high diagnostic accuracy for AI models, with pooled AUC scores of 91% for CT-based, 84% for MRI-based, and 92% for PET/CT-based radiomics; further, deep learning models performed similarly well, indicating a need for more extensive multicenter research for
View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!