<sup>18</sup>F-FDG PET-based ensemble deep learning model for the prediction of lymphovascular invasion in colorectal cancer patients.
Authors
Affiliations (5)
Affiliations (5)
- First Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang, China.
- Harbin Medical University Cancer Hospital, Harbin, Heilongjiang, China.
- Shanghai Public Health Clinical Center, Fudan University, Shanghai, China.
- Shanghai Tenth People's Hospital, Shanghai, China.
- First Affiliated Hospital of Harbin Medical University, Harbin, Heilongjiang, China. [email protected].
Abstract
The aim of this study was to investigate the diagnostic performance of the 2.5-dimensional (2.5D) ensemble deep learning (DL) model based on <sup>18</sup>F-fuorodeoxyglucose (FDG) positron emission tomography (PET) images in predicting lymphovascular invasion (LVI) in colorectal cancer (CRC) patients. In this retrospective study, 177 CRC patients who underwent preoperative <sup>18</sup>F-FDG PET/computed tomography were enrolled and assigned to the training cohort or the internal test cohort. Three inputs were determined according to the manually-delineated tumor volume of interest (VOI): the PrimaryLesion-2.5D input containing only the tumor VOI, the ProximalPeritumoral-2.5D input extending 10 mm outward from the VOI boundary, and the DistalPeritumoral-2.5D input extending 20 mm outward from the VOI boundary. Five common DL algorithm models, including VGG16, Googlenet, ResNet50, DenseNet201, and Vision Transformer were evaluated. Support vector machine was used to integrate the model outputs with good prediction performance to establish the Fusion model. The Radiomics and Clinical models were constructed for comparative analysis. The performance of the model was statistically analyzed by the area under the curve (AUC), accuracy, F1-score, parameter amount and inference time. The ProximalPeritumoral-DenseNet201 model (training cohort: AUC = 0.840, accuracy = 0.772, F1-score = 0.714; internal test cohort: AUC = 0.738, accuracy = 0.796, F1-score = 0.645; parameter amount = 18.097 M, inference time = 47.500 ms) and PrimaryLesion-ResNet50 model (training cohort: AUC = 0.746, accuracy = 0.740, F1-score = 0.628; internal test cohort: AUC = 0.733, accuracy = 0.704, F1-score = 0.619; parameter amount = 23.512 M, inference time = 38.200 ms) achieved an optimal balance between performance and computational efficiency. The performance of the Fusion model combined with the ProximalPeritumoral-DenseNet201 model and PrimaryLesion-ResNet50 model was further improved, with an AUC of 0.874, an accuracy of 0.821, and an F1-score of 0.766 in the training cohort. In the internal test cohort, the AUC was 0.824, the accuracy was 0.815, and the F1-score was 0.722. The Fusion model outperformed the Radiomics and Clinical models. Moreover, it showed good clinical utility and calibration. The 2.5D ensemble DL model based on <sup>18</sup>F-FDG PET images performed well for the prediction of LVI in CRC, proving its potential as a precision medical support tool for CRC patients.