Explore the vast range of titles available.

This article presents a biochemical kinetic model for the non-homologous end joining (NHEJ) of DNA double-strand break (DSB) repair pathway. The model is part of a theoretical framework to encompass all cellular DSB repair pathways. The NHEJ model was developed by taking into consideration the biological characteristics of the repair processes in the absence of homologous recombination (HR), the major alternative pathway for DSB repair. The model considers fast and slow components of the repair kinetics resulting in a set of differential equations that were solved numerically. In the absence of available published data for reaction rate constants for the repair proteins involved in NHEJ, we propose reaction rate constants for the solution of the equations. We assume as a first approximation that the reaction rate constants are applicable to mammalian cells under same conditions. The model was tested by comparing measured and simulated DSB repair kinetics obtained with HR-deficient cell lines irradiated by X rays in the dose range of 20–80 Gy. Measured data for initial protein recruitment to a DSB were used to independently estimate rate constants for Ku70/Ku80 and DNA-dependent protein kinase catalytic subunit (DNA-PKcs). We show here based on the model of DSB repair described in this article, application of the model in the accompanying article (Taleei et al., Radiat. Res. 179, 540–548, 2013) and by simulation of repair times for each individual DSB produced by individual tracks of electrons, that the complexity of damage may explain the slow kinetics of DNA DSB repair.

The physical properties of particles used in radiation therapy, such as protons, have been well characterized and their dose distributions are superior to photon-based treatments. However, proton therapy may also have inherent biologic advantages that have not been capitalized on. Unlike photon beams, the linear energy transfer (LET) and hence biologic effectiveness of particle beams varies along the beam path. Selective placement of areas of high effectiveness could enhance tumor cell kill and simultaneously spare normal tissues. However, previous methods for mapping spatial variations in biologic effectiveness are time-consuming and often yield inconsistent results with large uncertainties. Thus the data needed to accurately model relative biological effectiveness to guide novel treatment planning approaches are limited. We used Monte Carlo modeling and high-content automated clonogenic survival assays to spatially map the biologic effectiveness of scanned proton beams with high accuracy and throughput while minimizing biological uncertainties. We found that the relationship between cell kill, dose and LET, is complex and non-unique. Measured biologic effects were substantially greater than in most previous reports, and non-linear surviving fraction response was observed even for the highest LET values. Extension of this approach could generate data needed to optimize proton therapy plans incorporating variable RBE.

Background The integration of mobile cone‐beam computed tomography (CBCT) into brachytherapy workflows offers clinical advantages such as immediate verification of applicator placement and adaptive treatment planning. These benefits require sufficient image quality to delineate applicators, target volumes, and organs at risk. A systematic evaluation of automatic exposure control (AEC) settings, radiation dose, and image quality is essential to ensure clinically acceptable imaging while minimizing patient exposure. Purpose This study evaluates the characteristics of AEC and its impact on image quality and radiation dose in a mobile CBCT system used for brachytherapy. Methods The Elekta ImagingRing CBCT system was used to scan a CatPhan phantom under two imaging protocols: Medium Dose Limit (MDL) and Ultra‐High Dose Limit (UHDL). This system employs a two‐layer mAs modulation process, consisting of preset mA values based on body mass index (BMI) and adjusted mA based on real‐time AEC. A bolus was used to simulate larger patient sizes. Real‐time x‐ray tube current at 10 degrees intervals was recorded. Image quality was evaluated using image noise, noise power spectrum (NPS), modulation transfer function (MTF), Hounsfield Unit (HU) linearity, uniformity index (UI), and contrast‐to‐noise ratio (CNR) across different protocols. Results AEC effectively modulated x‐ray tube current in the MDL protocol after x‐ray attenuation through the scanned phantom was measured. The UHDL protocol demonstrated greater noise reduction than the MDL. MTF values were comparable between protocols, indicating preserved spatial resolution in the MDL protocol. HU linearity was consistent across all protocols, with R2 > 0.993. Conclusion AEC in mobile CBCT optimized radiation dose and image quality by adjusting tube current based on attenuation. The MDL protocol reduced radiation exposure while maintaining image quality, making it a viable option for verifying applicator placement and treatment planning in brachytherapy. The UHDL protocol achieved noise reduction with the maximum available tube current.

Purpose The COVID‐19 pandemic has led to disorder in work and livelihood of a majority of the modern world. In this work, we review its major impacts on procedures and workflow of clinical physics tasks, and suggest alternate pathways to avoid major disruption or discontinuity of physics tasks in the context of small, medium, and large radiation oncology clinics. We also evaluate scalability of medical physics under the stress of “social distancing”. Methods Three models of facilities characterized by the number of clinical physicists, daily patient throughput, and equipment were identified for this purpose. For identical objectives of continuity of clinical operations, with constraints such as social distancing and unavailability of staff due to system strain, however with the possibility of remote operations, the performance of these models was investigated. General clinical tasks requiring on‐site personnel presence or otherwise were evaluated to determine the scalability of the three models at this point in the course of disease spread within their surroundings. Results The clinical physics tasks within three models could be divided into two categories. The former, which requires individual presence, include safety‐sensitive radiation delivery, high dose per fraction treatments, brachytherapy procedures, fulfilling state and nuclear regulatory commission's requirements, etc. The latter, which can be handled through remote means, include dose planning, physics plan review and supervision of quality assurance, general troubleshooting, etc. Conclusion At the current level of disease in the United States, all three models have sustained major system stress in continuing reduced operation. However, the small clinic model may not perform if either the current level of infections is maintained for long or staff becomes unavailable due to health issues. With abundance, and diversity of innovative resources, medium and large clinic models can sustain further for physics‐related radiotherapy services.

We present a microscopic formalism that accounts for the formation of nano-scale bubbles owing to a burst of water molecules after the passage of high energy charged particles that lead to the formation of “hot” non-ionizing excitations or thermal spikes (TS). We construct amorphous track structures to account for the formation of TS by ionizing radiation in liquid water. Subsequently, we simulate sudden expansion and collective motion of water molecules by employing a molecular dynamics (MD) simulation that allows computation of O ( 10 6 ) particle trajectories and breaking/forming of chemical bonds on the fly using a reactive force field, ReaxFF. We calculate the fluctuations of thermodynamic variables before and after TS formation to model the macroscopic abrupt changes in the system, possibly the occurrence of a first-order phase transition, and go beyond the accessible simulation times by engaging fluid dynamic equations with appropriate underlying symmetries and boundary conditions. We demonstrate the coexistence of a rapidly growing condensed state of water and a hot spot that forms a stable state of diluted water at high temperatures and pressures, possibly at a supercritical phase. Depending on the temperature of TS, the thin shell of a highly dense state of water grows by three to five times the speed of sound in water, forming a thin layer of shock wave (SW) buffer, wrapping around the nano-scale cylindrical symmetric bubble. The stability of the bubble, as a result of the incompressibility of water at ambient conditions and the surface tension, allows the transition of supersonic SW to a subsonic contact discontinuity and dissipation to thermo-acoustic sound waves. Thus, TS gradually decays to acoustic waves, a channel of deexcitation that competes with the spontaneous emission of photons, and a direct mechanism for water luminescence. We further study the mergers of nanobubbles that lead to jet-flow structures at the collision interface. We introduce a time delay in the nucleation of nano-bubbles, a novel mechanism, responsible for the growth and stability of much larger or even micro-bubbles, possibly relevant to FLASH ultra-high dose rate (UHDR). Graphical abstract Molecular dynamic simulation of the interaction between two ionizing radiation-induced nano-bubble formed simultaneously.

We investigated the kinetics of simple and complex types of double-strand breaks (DSB) using our newly proposed mechanistic mathematical model for NHEJ DSB repair. For this purpose the simulated initial spectrum of DNA DSB, induced in an atomistic canonical model of B-DNA by low-energy single electron tracks, 100 eV to 4.55 keV, and the electrons generated by ultrasoft X rays (CK, AlK and TiK), were subjected to NHEJ repair processes. The activity elapsed time of sequentially independent steps of repair performed by proteins involved in NHEJ repair process were calculated for separate DSB. The repair kinetics of DSBs were computed and compared with published data on repair kinetics obtained by pulsed-field gel electrophoresis method. The comparison shows good agreement for V79-4 cells irradiated with ultrasoft X rays. The average times for the repair of simple and complex DSB confirm that double-strand break complexity is a potential explanation for the slow component of DSB repair observed in V79-4 cells irradiated by ultrasoft X rays.

Recent experimental data have revealed a wealth of information that provides an exceptional opportunity to construct a mechanistic model of DNA repair. The cellular response to radiation exposure starts with repair of DNA damage and cell signalling that may lead to mutation, or cell death. The purpose of this work was to construct a mechanistic mathematical model of DNA repair in mammalian cells. The repair model is based on biochemical action of repair proteins to examine the hypotheses regarding two or more components of double strand break (DSB)repair kinetics.The mechanistic mathematical model of repair proposed in this thesis is part of a bottom-up approach that assumes the cell is a complex system. In this approach radiation induces DNA damage, and the cellular response to radiation perturbation was modelled in terms of activating repair processes. A biochemical kinetic method based on law of mass action was employed to model the repair pathways. The repair model consists of a set of nonlinear differential equations that calculates and explains protein activity on the damage step by step. The model takes into account complexity of the DSB, topology of damage in the cell nucleus, and cell cycle.The solution of the model in terms of overall kinetics of DSB repair was compared with pulsed-field gel electrophoresis measurements. The repair model was integrated with the track structure model to calculate the damage spectrum and repair kinetics for every individual DSB induced by monoenergetic electrons, and ultrasoft X-rays. For this purpose we proposed a method to sample the protein repair actions for every individual DSB, and finally calculate the total repair time for that specific DSB. The DSB-repair kinetics for the number of DSB induced by 500 tracks of monoenergetic electrons and ultrasoft X-rays were calculated and compared with experimental results for cells irradiated with AlK, CK, and TiK ultrasoft X-rays.The results presented here form the first example of mechanistic modelling and calculations for NHEJ, HR and MMEJ repair pathways. The results, for the first time, quantitatively confirm the hypothesis that the complex type double strand breaks play a major role in the slow kinetics of DSB repair. The results also confirm that simple DSB located in the heterocromatin delay the repair process due to a series of processes that are required for the relaxation of the heterochromatin. The repair model established in this work provides a unique opportunity to continue this study of cellular responses to radiation further downstream that may have important implications for human risk estimation and radiotherapy.

Breast cancer (BC) remains one of the leading causes of cancer-related mortality among women, despite recent advances in Computer-Aided Diagnosis (CAD) systems. Accurate and efficient interpretation of multi-view mammograms is essential for early detection, driving a surge of interest in Artificial Intelligence (AI)-powered CAD models. While state-of-the-art multi-view mammogram classification models are largely based on Transformer architectures, their computational complexity scales quadratically with the number of image patches, highlighting the need for more efficient alternatives. To address this challenge, we propose Mammo-Mamba, a novel framework that integrates Selective State-Space Models (SSMs), transformer-based attention, and expert-driven feature refinement into a unified architecture. Mammo-Mamba extends the MambaVision backbone by introducing the Sequential Mixture of Experts (SeqMoE) mechanism through its customized SecMamba block. The SecMamba is a modified MambaVision block that enhances representation learning in high-resolution mammographic images by enabling content-adaptive feature refinement. These blocks are integrated into the deeper stages of MambaVision, allowing the model to progressively adjust feature emphasis through dynamic expert gating, effectively mitigating the limitations of traditional Transformer models. Evaluated on the CBIS-DDSM benchmark dataset, Mammo-Mamba achieves superior classification performance across all key metrics while maintaining computational efficiency.

Despite advancements in Computer-Aided Diagnosis (CAD) systems, breast cancer remains one of the leading causes of cancer-related deaths among women worldwide. Recent breakthroughs in Artificial Intelligence (AI) have shown significant promise in development of advanced Deep Learning (DL) architectures for breast cancer diagnosis through mammography. In this context, the paper focuses on the integration of AI within a Human-Centric workflow to enhance breast cancer diagnostics. Key challenges are, however, largely overlooked such as reliance on detailed tumor annotations and susceptibility to missing views, particularly during test time. To address these issues, we propose a hybrid, multi-scale and multi-view Swin Transformer-based framework (MSMV-Swin) that enhances diagnostic robustness and accuracy. The proposed MSMV-Swin framework is designed to work as a decision-support tool, helping radiologists analyze multi-view mammograms more effectively. More specifically, the MSMV-Swin framework leverages the Segment Anything Model (SAM) to isolate the breast lobe, reducing background noise and enabling comprehensive feature extraction. The multi-scale nature of the proposed MSMV-Swin framework accounts for tumor-specific regions as well as the spatial characteristics of tissues surrounding the tumor, capturing both localized and contextual information. The integration of contextual and localized data ensures that MSMV-Swin's outputs align with the way radiologists interpret mammograms, fostering better human-AI interaction and trust. A hybrid fusion structure is then designed to ensure robustness against missing views, a common occurrence in clinical practice when only a single mammogram view is available.

Lung cancer remains one of the leading causes of cancer-related mortality worldwide. A crucial challenge for early diagnosis is differentiating uncertain cases with similar visual characteristics and closely annotation scores. In clinical practice, radiologists rely on quantitative, hand-crafted Radiomic features extracted from Computed Tomography (CT) images, while recent research has primarily focused on deep learning solutions. More recently, Vision-Language Models (VLMs), particularly Contrastive Language-Image Pre-Training (CLIP)-based models, have gained attention for their ability to integrate textual knowledge into lung cancer diagnosis. While CLIP-Lung models have shown promising results, we identified the following potential limitations: (a) dependence on radiologists' annotated attributes, which are inherently subjective and error-prone, (b) use of textual information only during training, limiting direct applicability at inference, and (c) Convolutional-based vision encoder with randomly initialized weights, which disregards prior knowledge. To address these limitations, we introduce AutoRad-Lung, which couples an autoregressively pre-trained VLM, with prompts generated from hand-crafted Radiomics. AutoRad-Lung uses the vision encoder of the Large-Scale Autoregressive Image Model (AIMv2), pre-trained using a multi-modal autoregressive objective. Given that lung tumors are typically small, irregularly shaped, and visually similar to healthy tissue, AutoRad-Lung offers significant advantages over its CLIP-based counterparts by capturing pixel-level differences. Additionally, we introduce conditional context optimization, which dynamically generates context-specific prompts based on input Radiomics, improving cross-modal alignment.