Improved outcome models with denoising diffusion.

PURPOSE: Radiotherapy outcome modelling often suffers from class imbalance in the modelled endpoints. One of the main options to address this issue is by introducing new synthetically generated datapoints, using generative models, such as Denoising Diffusion Probabilistic Models (DDPM). In this study, we implemented DDPM to improve performance of a tumor local control model, trained on imbalanced dataset, and compare this approach with other common techniques. METHODS: A dataset of 535 NSCLC patients treated with SBRT (50 Gy/5 fractions) was used to train a deep learning outcome model for tumor local control prediction. The dataset included complete treatment planning data (planning CT images, 3D planning dose distribution and patient demographics) with sparsely distributed endpoints (6-7 % experiencing local failure). Consequently, we trained a novel conditional 3D DDPM model to generate synthetic treatment planning data. Synthetically generated treatment planning datapoints were used to supplement the real training dataset and the improvement in the model's performance was studied. Obtained results were also compared to other common techniques for class imbalanced training, such as Oversampling, Undersampling, Augmentation, Class Weights, SMOTE and ADASYN. RESULTS: Synthetic DDPM-generated data were visually trustworthy, with Fréchet inception distance (FID) below 50. Extending the training dataset with the synthetic data improved the model's performance by more than 10%, while other techniques exhibited only about 4% improvement. CONCLUSIONS: DDPM introduces a novel approach to class-imbalanced outcome modelling problems. The model generates realistic synthetic radiotherapy planning data, with a strong potential to increase performance and robustness of outcome models.

A high-capacity alkaline phosphatase reporter system for the rapid analysis of specificity and relative affinity of peptides from phage-display libraries.

We describe a novel reporter enzyme cassette system which enables the analysis of large numbers of linear and cyclic peptides in terms of their binding to a specific target molecule. In this system, peptides selected for target binding from random peptide phage-display libraries are expressed as cloned fusion proteins with bacterial alkaline phosphatase. The binding specificity and relative affinity of each peptide-enzyme fusion protein is then evaluated in a target-specific ELISA. This strategy enables direct identification of the highest affinity peptides, specific for a given target, which can then be sequenced at the DNA level to derive their peptide sequences. This eliminates the need to sequence large numbers of clones to establish consensus sequences for binding peptides. This approach also eliminates the need for peptide synthesis or phage ELISA to determine relative binding affinities, which can be technically difficult. Identification of binding peptides based on specificity and relative affinity, rather than conforming to an amino acid consensus sequence, enables the rapid evaluation of hundreds of candidate peptides and identification of rare (non-consensus) binding peptides which may otherwise be missed.