EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

3D deep convolution neural network for radiation pneumonitis prediction following stereotactic body radiotherapy

Rishabh Kapoor iD, William Sleeman iD, Jatinder Palta iD, Elisabeth Weiss

Journal of Applied Clinical Medical Physics · 2023-03 · 24 (3)

DOI: 10.1002/acm2.13875 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Applied Clinical Medical Physics, online erschienen am 22.12.2022. Verfasst von Rishabh Kapoor und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiation Oncology, Virginia Commonwealth University, Richmond, Virginia, USA. Hauptthemen laut MEDLINE: Radiosurgery, Radiation Pneumonitis, Carcinoma, Non-Small-Cell Lung, Lung Neoplasms. Die Arbeit wurde bisher 26-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this study, we investigated 3D convolutional neural networks (CNNs) with input from radiographic and dosimetric datasets of primary lung tumors and surrounding lung volumes to predict the likelihood of radiation pneumonitis (RP). Pre‐treatment, 3‐ and 6‐month follow‐up computed tomography (CT) and 3D dose datasets from one hundred and ninety‐three NSCLC patients treated with stereotactic body radiotherapy (SBRT) were retrospectively collected and analyzed for this study. DenseNet‐121 and ResNet‐50 models were selected for this study as they are deep neural networks and have been proven to have high accuracy for complex image classification tasks. Both were modified with 3D convolution and max pooling layers to accept 3D datasets. We used a minority class oversampling approach and data augmentation to address the challenges of data imbalance and data scarcity. We built two sets of models for classification of three (No RP, Grade 1 RP, Grade 2 RP) and two (No RP, Yes RP) classes as outputs. The 3D DenseNet‐121 models performed better (F1 score [0.81], AUC [0.91] [three class]; F1 score [0.77], AUC [0.84] [two class]) than the 3D ResNet‐50 models (F1 score [0.54], AUC [0.72] [three‐class]; F1 score [0.68], AUC [0.71] [two‐class]) ( p = 0.017 for three class predictions). We also attempted to identify salient regions within the input 3D image dataset via integrated gradient (IG) techniques to assess the relevance of the tumor surrounding volume for RP stratification. These techniques appeared to indicate the significance of the tumor and surrounding regions in the prediction of RP. Overall, 3D CNNs performed well to predict clinical RP in our cohort based on the provided image sets and radiotherapy dose information.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

RadiosurgeryRadiation PneumonitisCarcinoma, Non-Small-Cell LungLung NeoplasmsHumansRetrospective StudiesNeural Networks, Computer

Stichwörter der Autor:innen

DenseNet 121ResNet-50artificial intelligencedeep learningoutcome predictionradiation pneumonitis

Autor:innen und Institutionen

  • Department of Radiation Oncology, Virginia Commonwealth University, Richmond, Virginia, USA.

ORCID-Profile: Rishabh Kapoor, William Sleeman, Jatinder Palta

Zitieren

APA

Kapoor, R., Sleeman, W., Palta, J., & Weiss, E. (2023). 3D deep convolution neural network for radiation pneumonitis prediction following stereotactic body radiotherapy. Journal of Applied Clinical Medical Physics, 24(3). https://doi.org/10.1002/acm2.13875

Vancouver

Kapoor R, Sleeman W, Palta J, Weiss E. 3D deep convolution neural network for radiation pneumonitis prediction following stereotactic body radiotherapy. Journal of Applied Clinical Medical Physics. 2023;24(3). doi:10.1002/acm2.13875

BibTeX
@article{Kapoor2023,
  author = {Kapoor, Rishabh and Sleeman, William and Palta, Jatinder and Weiss, Elisabeth},
  title = {3D deep convolution neural network for radiation pneumonitis prediction following stereotactic body radiotherapy},
  journal = {Journal of Applied Clinical Medical Physics},
  year = {2023},
  volume = {24},
  number = {3},
  doi = {10.1002/acm2.13875}
}
Kostenlos starten →