No abstract

Background and Purpose:To examine the feasibility of improving breast-conserving radiotherapy with simultaneous integrated boost (SIB) and analyzing the efficiency of forward versus inverse intensity-modulated radiotherapy (IMRT) techniques in providing the same.Materials and Methods:Three-dimensional conformal radiotherapy (3DCRT) field-in-field (FIF) plans with simultaneous and sequential boost and IMRT SIB plans were generated for the datasets of 20 patients who had undergone breast-conserving surgery. The 3 plans were compared dosimetrically for efficiency in terms of planning target volume (PTV) coverage (PTV 95%), homogeneity and conformity, dose delivered to ipsilateral/contralateral lungs (I/L: V10, V20, C/L: Vmean, V5), heart and contralateral breast (Vmean, V30 for heart and Vmean, V1, V5 for C/L breast).Results:The FIF 3DCRT plan with SIB (PLAN B) was more homogeneous than the classical technique with sequential boost (PLAN A). There were less hot spots in terms of Dmax (63.7 ± 1.3) versus Dmax (68.9 ± 1), P < 0.001 and boost V107%, B (0.3 ± 0.7) versus A (3.5 ± 5.99), P = 0.001. The IMRT SIB (PLAN C) did not provide any significant dosimetric advantage over the 3DCRT SIB technique. IMRT SIB plan C was associated with increased dose to contralateral lung in-terms of V5 (10.35 +/- 18.23) vs. (1.13 +/- 4.24), P = 0.04 and Vmean (2.12 ± 2.18) versus Vmean (0.595 ± 0.89), P = 0.008. There was 3-fold greater exposure in terms of Monitor Unit (MU) (1024.9 ± 298.32 versus 281.05 ± 20.23, P < 0.001) and treatment delivery time.Conclusions:FIF 3DCRT SIB provides a dosimetrically acceptable and technically feasible alternative to the classical 3DCRT plan with sequential boost for breast-conserving radiotherapy. It reduces treatment time by 2 weeks. IMRT SIB does not appear to have any dosimetric advantage; it is associated with significantly higher doses to contralateral lung and heart and radiation exposure in terms of MU.

Designing solvent systems is the key to achieving the facile synthesis and separation of desired products from chemical processes. In this regard, many machine-learning models have been developed to predict the solubilities of given solute-solvent pairs. However, breakthroughs in developing predictive models for solubility are needed, which can be accomplished through a remarkable expansion and integration of experimental and computational solubility databases. To maximize predictive accuracy, these two databases should not be separately trained when developing ML models. In addition, they should not be simply combined without reconciling the discrepancies between different magnitudes of errors and uncertainties. Here, we introduce self-evolving solubility databases and graph neural networks developed through semi-supervised self-training approaches. Solubilities from quantum-mechanical calculations are referred to during semi-supervised learning, but they are not directly added to the database. Such methodologies enable the augmentation of databases while correcting the discrepancy between experiments and computation and improving the predictive accuracy against experimental solubilities. The resulting model was successfully applied to two practical examples relevant to solvent selection in organic reactions and separation processes: (i) linear relationship between reaction rates and solvation free energy for three organic reactions, and (ii) partition coefficients for lignin-derived monomers and drug-like molecules.

Designing solvent systems is key to achieving the facile synthesis and separation of desired products from chemical processes, so many machine learning models have been developed to predict solubilities. However, breakthroughs are needed to address deficiencies in the model’s predictive accuracy and generalizability; this can be addressed by expanding and integrating experimental and computational solubility databases. To maximize predictive accuracy, these two databases should not be trained separately, and they should not be simply combined without reconciling the discrepancies from different magnitudes of errors and uncertainties. Here, we introduce self-evolving solubility databases and graph neural networks developed through semi-supervised self-training approaches. Solubilities from quantum-mechanical calculations are referred to during semi-supervised learning, but they are not directly added to the experimental database. Dataset augmentation is performed from 11,637 experimental solubilities to >900,000 data points in the integrated database, while correcting for the discrepancies between experiment and computation. Our model was successfully applied to study solvent selection in organic reactions and separation processes. The accuracy (mean absolute error around 0.2 kcal/mol for the test set) is quantitatively useful in exploring Linear Free Energy Relationships between reaction rates and solvation free energies for 11 organic reactions. Our model also accurately predicted the partition coefficients of lignin-derived monomers and drug-like molecules. We anticipate this approach will be attractive to other areas of predictive chemistry where experimental, computational, and any other heterogeneous data sources should be combined.

Designing solvent systems is the key to achieving the facile synthesis and separation of desired products from chemical processes. In this regard, many machine-learning models have been developed to predict the solubilities of given solute-solvent pairs. However, breakthroughs in developing predictive models for solubility are needed, which can be accomplished through a remarkable expansion and integration of experimental and computational solubility databases. To maximize predictive accuracy, these two databases should not be separately trained when developing ML models. In addition, they should not be simply combined without reconciling the discrepancies between different magnitudes of errors and uncertainties. Here, we introduce self-evolving solubility databases and graph neural networks developed through semi-supervised self-training approaches. Solubilities from quantum-mechanical calculations are referred to during semi-supervised learning, but they are not directly added to the database. Such methodologies enable the augmentation of databases while correcting the discrepancy between experiments and computation and improving the predictive accuracy against experimental solubilities. The resulting model was successfully applied to two practical examples relevant to solvent selection in organic reactions and separation processes: (i) linear relationship between reaction rates and solvation free energy for three organic reactions, and (ii) partition coefficients for lignin-derived monomers and drug-like molecules.