Convolutional neural network and transfer learning for dose volume histogram prediction for prostate cancer radiotherapy
Autor: | Jaime Pérez-Alija, Pedro Gallego, Eva M. Ambroa |
---|---|
Rok vydání: | 2021 |
Předmět: |
Male
Dose-volume histogram Artificial intelligence Computer science Convolutional neural network VMAT Dose prediction 030218 nuclear medicine & medical imaging Machine Learning 03 medical and health sciences 0302 clinical medicine Histogram Humans Radiology Nuclear Medicine and imaging Ground truth Radiological and Ultrasound Technology Artificial neural network Radiotherapy business.industry Radiotherapy Planning Computer-Assisted Deep learning Prostatic Neoplasms Confusion matrix Radiotherapy Dosage Pattern recognition Transfer learning Oncology 030220 oncology & carcinogenesis Neural Networks Computer Transfer of learning business |
Zdroj: | Medical Dosimetry r-IIB SANT PAU. Repositorio Institucional de Producción Científica del Instituto de Investigación Biomédica Sant Pau instname |
ISSN: | 0958-3947 |
Popis: | To adopt a transfer learning approach and establish a convolutional neural network (CNN) model for the prediction of rectum and bladder dose-volume histograms (DVH) in prostate patients treated with a VMAT technique. One hundred forty-four VMAT patients with intermediate or high-risk prostate cancer were included in this study. Data were split into two sets: 120 and 24 patients, respectively. The second set was used for final validation. To ensure the accuracy of the training data, we developed a ground-truth analysis for detecting and correcting for all potential outliers. We used transfer learning in combination with a pre-trained VGG-16 network. We dropped the fully connected layers from the VGG-16 and added a new fully connected neural network. The inputs for the CNN were a 2D image of the volumes contoured in the CT, but we only retained the geometrical infor-mation of every CT-slice. The outputs were the corresponding rectum and bladder DVH for every slice. We used a confusion matrix to analyze the performance of our model. Our model achieved 100% and 81% of true positive and true negative predictions, respectively. We have an overall accuracy of 87.5%, a misclassification rate of 12.5%, and a precision of 100%. We have successfully developed a model for reliable prediction of rectum and bladder DVH in prostate patients by applying a previously pre-trained CNN. To our knowledge, this is the first attempt to apply transfer learning to the prediction of DVHs that accounts for the ground truth problem. |
Databáze: | OpenAIRE |
Externí odkaz: |