Publications by authors named "Hazem A A Nomer"

Article Synopsis
  • The study focuses on using deep learning to improve the speed and quality of radiation treatment planning by predicting patient-specific 3D dose distributions in real time.
  • It analyzed the impact of training dataset size and model complexity on the accuracy of dose predictions for 1250 prostate patients, using various sizes of neural network models.
  • Results showed that more training data increases prediction accuracy, with the most effective model achieving low prediction errors in dose distribution despite not reaching a plateau in accuracy at 1000 training patients, indicating potential for further optimization.
View Article and Find Full Text PDF
Article Synopsis
  • Intensity modulated proton therapy (IMPT) faces challenges from patient setup and proton range uncertainties, leading to longer planning times with robust optimization methods.
  • A deep learning (DL) model was developed to predict dose distributions in various error scenarios for head and neck cancer patients, effectively evaluating plan robustness.
  • The model demonstrated impressive accuracy, aligning closely with ground truth plans for most patients, and was able to predict full 3D dose distributions quickly, making it a viable option for enhancing IMPT treatment planning.
View Article and Find Full Text PDF