The advances in the development of Facilitative Playbacks extracted from High-Speed videoendo- scopic sequences of the vocal folds are hindered by a notable lack of publicly available datasets annotated with the semantic segmentations corresponding to the area of the glottal gap. This fact also limits the reproducibility and further exploration of existing research in this field. To address this gap, GIRAFE (Glottal Imaging Repository for Advanced Segmentation, Analysis, and Facilitative Playbacks Evaluation) is a data repository designed to facilitate the devel- opment of advanced techniques for the semantic segmentation, analysis, and fast evaluation of High-Speed videoendoscopic sequences of the vocal folds. The repository includes 65 high-speed videoendoscopic recordings from a cohort of 50 patients (30 female, 20 male). The dataset com- prises 15 recordings from healthy controls, 26 from patients with diagnosed voice disorders, and 24 with an unknown health condition. All of them were manually annotated by an expert, including the masks corresponding to the semantic segmentation of the glottal gap. The repository is also complemented with the automatic segmentation of the glottal area using different state-of-the-art approaches. This data set has already supported several studies, which demonstrates its usefulness for the development of new glottal gap segmentation algorithms from High-Speed-Videoendoscopic sequences to improve or create new Facilitative Playbacks. Despite these advances and others in the field, the broader challenge of performing an accurate and completely automatic semantic segmentation method of the glottal area remains open.

Purpose: This study aims to develop a comprehensive simulation framework to connect radiation effects from the microscopic to the nanoscopic scale. Method: The process begins with a Geant4-DNA simulation based on the example ”molecularDNA”, producing a dataset of twelve different types of early DNA damages within an Escherichia coli (E. coli) bacterium, generated by proton irradiation at different kinetic energies, giving a nano-scale view of the particle–matter interaction. Then we pass to the micro-scale with a Geant4 simulation, based on the example ”radiobiology”, providing a microscopic view of proton interactions with matter through the Linear Energy Transfer (LET). Then GANDALF (Generative ANsatz for DNA damage evALuation and Forecast) Machine Learning (ML) toolkit, a Neural Network (NN)-based regression system, is employed to correlate the micro-scale LET data with the nano-scale occurrences of DNA damages in the E. coli bacterium. Results: The trained ML algorithm provides a practical tool to convert LET curves versus depth in a water phantom into DNA damage curves for twelve distinct types of DNA damage. To assess the performance, we evaluated the choice and optimization of the regression system based on its interpolation and extrapolation capabilities, ensuring the model could reliably predict DNA damage under various conditions. Conclusions: Through the synergistic integration of Geant4, Geant4-DNA and ML, the study provides a tool to easily convert the results at the micro-scale of Geant4 to those at the nano-scale of Geant4-DNA without having to deal with the high CPU time requirements of the latter.

Purpose: Based on considerable interest to enlarge the experimental database of radioresistant cells after their irradiation with helium ions, HTB140, MCF-7 and HTB177 human malignant cells are exposed to helium ion beams having different linear energy transfer (LET). Materials and methods: The cells are irradiated along the widened 62 MeV/u helium ion Bragg peak, providing LET of 4.9, 9.8, 23.4 and 36.8 keV/µm. Numerical simulations with the Geant4 toolkit are used for the experimental design. Cell survival is evaluated and compared with reference γ-rays. DNA double strand breaks are assessed via γ-H2AX foci. Results: With the increase of LET, surviving fractions at 2 Gy decrease, while RBE (2 Gy, γ) gradually increase. For HTB140 cells, above the dose of 4 Gy, a slight saturation of survival is observed while the increase of RBE (2 Gy, γ) remains unaffected. With the increase of LET the increase of γ-H2AX foci is revealed at 0.5 h after irradiation. There is no significant difference in the number of foci between the cell lines for the same LET. From 0.5 to 24 h, the number of foci drops reaching its residual level. For each time point, there are small differences in DNA DSB among the three cell lines. Conclusion: Analyses of data acquired for the three cell lines irradiated by helium ions, having different LET, reveal high elimination capacity and creation of a large number of DNA DSB with respect to γ-rays, and are between those reported for protons and carbon ions.

Purpose: Interdisciplinary scientific communities have shown large interest to achieve a mechanistic description of radiation-induced biological damage, aiming to predict biological results produced by different radiation quality exposures. Monte Carlo track-structure simulations are suitable and reliable for the study of early DNA damage induction used as input for assessing DNA damage. This study presents the most recent improvements of a Geant4-DNA simulation tool named “dsbandrepair”. Methods: “dsbandrepair” is a Monte Carlo simulation tool based on a previous code (FullSim) that estimates the induction of early DNA single-strand breaks (SSBs) and double-strand breaks (DSBs). It uses DNA geometries generated by the DNAFabric computational tool for simulating the induction of early single-strand breaks (SSBs) and double-strand breaks (DSBs). Moreover, the new tool includes some published radiobiological models for survival fraction and un-rejoined DSB. Its application for a human fibroblast cell and human umbilical vein endothelial cell containing both heterochromatin and euchromatin was conducted. In addition, this new version offers the possibility of using the new IRT-syn method for computing the chemical stage. Results: The direct and indirect strand breaks, SSBs, DSBs, and damage complexity obtained in this work are equivalent to those obtained with the previously published simulation tool when using the same configuration in the physical and chemical stages. Simulation results on survival fraction and un-rejoined DSB are in reasonable agreement with experimental data. Conclusions: “dsbandrepair” is a tool for simulating DNA damage and repair, benchmarked against experimental data. It has been released as an advanced example in Geant4.11.2.

Background: This study investigated alternative, non-invasive methods for human papillomavirus (HPV) detection in head and neck cancers (HNCs). We compared two approaches: analyzing computed tomography (CT) scans with a Deep Learning (DL) model and using radiomic features extracted from CT images with machine learning (ML) models. Methods: Fifty patients with histologically confirmed HNC were included. We first trained a modified ResNet-18 DL model on CT data to predict HPV status. Next, radiomic features were extracted from manually segmented regions of interest near the oropharynx and used to train four ML models (K-Nearest Neighbors, logistic regression, decision tree, random forest) for the same purpose. Results: The CT-based model achieved the highest accuracy (90%) in classifying HPV status. Among the ML models, K-Nearest Neighbors performed best (80% accuracy). Weighted Ensemble methods combining the CT-based model with each ML model resulted in moderate accuracy improvements (70–90%). Conclusions: Our findings suggest that CT scans analyzed by DL models hold promise for non-invasive HPV detection in HNC. Radiomic features, while less accurate in this study, offer a complementary approach. Future research should explore larger datasets and investigate the potential of combining DL and radiomic techniques.

Background: This study aimed to develop a novel human cell geometry for the Geant4-DNA simulation toolkit that explicitly incorporates all 23 chromosome pairs of the human cell. This approach contrasts with the existing, default human cell, geometrical model, which utilizes a continuous Hilbert curve. Methods: A Python-based tool named “complexDNA” was developed to facilitate the design of both simple and complex DNA geometries. This tool was employed to construct a human cell geometry with individual pairs of chromosomes. Subsequently, the performance of this chromosomal model was compared to the standard human cell model provided in the “molecularDNA” Geant4-DNA example. Results: Simulations using the new chromosomal model revealed minimal discrepancies in DNA damage yield and fragment size distribution compared to the default human cell model. Notably, the chromosomal model demonstrated significant computational efficiency, requiring approximately three times less simulation time to achieve equivalent results. Conclusions: This work highlights the importance of incorporating chromosomal structure into human cell models for radiation biology research. The “complexDNA” tool offers a valuable resource for creating intricate DNA structures for future studies. Further refinements, such as implementing smaller voxels for euchromatin regions, are proposed to enhance the model's accuracy.