Collin College

Purpose/Objective(s) Although charged-particle radiation therapy system treatments have existed for decades, device complications continue to cause adverse effects for patients. Furthermore, there is limited research on the type of malfunctions occurring. In this abstract, we aim to characterize the type of device problems and patient complications that have occurred to better understand complications stemming from radiation therapy delivery. We expect that radiation therapy device complications will cause inaccurate radiation doses to patients and expect there to be more mechanical complications than software complications with these devices. Materials/Methods Using the Food and Drug Administration (FDA) Manufacturer and User Facility Device Experience (MAUDE) database, we used statistical software (Python) to analyze all the data listed since 2016 coming from mandatory and voluntary reporters. We also assessed the Total Product Life Cycle report for medical charged-particle radiation therapy systems. Results From the MAUDE database, we showed that there were 44 injuries, 214 malfunctions, and 5 deaths since 2016. Common radiation exposure patient adverse effects were 11 radiation overdose, 3 unintended radiation exposure, and 2 radiation exposure unintended, and 2 radiation underdose events. The most common device complication events were 170 computer software problems, 28 use of device problems, 14 improper procedures, and 12 dose calculation errors, as shown in Table 1. Conclusion We show that, in the last 10 years, many patients suffered from injuries, malfunctions, and deaths due to charged-particle radiation therapy device complications with many receiving an inappropriate amount of radiation exposure. We also show that computer software problems are the most commonly reported device complications and were more commonly reported than mechanical device complications. The continued presence of these health consequences and device complications necessitates refinement, especially regarding software improvements of radiation therapy system treatments.

In the age of quantum communication and 6G, the safety of financial networks has become a more important concern as cyber threats become more sophisticated. This study examines how deep learning models can be used to improve vulnerability detection and prediction in quantum-secure financial settings. Based on the Quantum-Secure 6G Slicing Dataset, which combines the latest encryption methods, including ECC, AES, and QKD, the study employs preprocessing steps— data cleaning, normalizatio, label encoding, and SMOTE-based balancing to ensure optimal model performance. The suggested ANN was compared with other deep learning and machine learning methods, including Decision Trees (DTs), Convolutional Neural Networks (CNNs), and Logistic Regression (LR). The results of the experiments were found to be the highest with ANN, the accuracy, 92.25, precision, 89.58, recall, 93.99, and F1-score, 91.73, which indicates that the ANN outperforms the existing models. These findings indicate that the ANN will be a powerful and robust method for detecting fraud and forecasting attack vulnerability in quantum-secure financial networks, as it implies that the ANN is an effective way to capture complex nonlinear dependencies and achieve better generalization and predictive stability.

We present results from very long baseline interferometry (VLBI) observations of the nucleus in the lobe-dominated quasar 3C 207. These observations were completed at 8.4 GHz or 10.7 GHz (X band) from 1981 to 2010, spanning 29 yr. The nucleus of 3C 207 is the strongest and most variable in the 3CR complete sample of lobe-dominated quasars, which is under study to test relativistic jet models over a wide range of jet orientation angles. Images have typical resolutions of ∼0.5–1.0 mas and sensitivities of ∼0.1–0.2 mJy beam ^−1 . The VLBI core region shows multiple flux density outbursts from a stationary “true” core that feeds a “swinging component” ∼0.5 mas to the east. The position angle (PA) of the swinging component shows a long-term increase of ∼40°, with a short-term reversal of ∼10°. A one-sided, curved VLBI jet extends ∼25 mas eastward, with components spanning a PA range of ∼25°. The jet components have average apparent transverse velocities v _app ≈ 10 c . One component shows apparent acceleration from 7 c to 14 c at 2–3 mas from the true core, where the flow is redirected toward PA ∼ 90°. Another component shows marginal evidence for apparent deceleration. Individual jet components expand until reaching the recollimation zone. Our results are consistent with a physical model in which 3C 207 has quasi-periodic outbursts, jet precession by ballistic components on a conical surface with a small opening angle, and a recollimation process that modifies component motions and narrows the conical geometry on a scale of ∼100 pc.

We present results from very-long-baseline-interferometry (VLBI) observations of the nucleus in the lobe-dominated quasar 3C207. These observations were completed at 8.4 or 10.7 GHz (X-band) from 1981 to 2010, spanning 29 years. The nucleus of 3C207 is the strongest and most variable in the 3CR complete sample of LDQs, which is under study to test relativistic jet models over a wide range of jet orientation angles. Images have typical resolutions of 0.5-1.0 milliarcseconds (mas) and sensitivities of 0.1-0.2 mJy beam^-1. The VLBI core region has flux density outbursts at mean intervals of 7 yr; two of these are multiple outbursts from a stationary "true: core that feeds a "swinging component" 0.5 mas to the east. The position angle (PA) of the swinging component shows a long-term increase of 40°, with a short-term reversal of 10°. A one-sided, curved VLBI jet extends 25 mas eastward, with components spanning a PA range of 25°. The jet components have average apparent transverse velocities 10c. One component shows apparent acceleration from 7c to 14c at 2-3 mas from the true core, where the flow is redirected toward PA 90°. Another component shows marginal evidence for apparent deceleration. Individual jet components expand until reaching the recollimation zone. Our results are consistent with a physical model in which 3C207 has quasi-periodic outbursts, jet precession by ballistic components on a conical surface with a small opening angle, and a recollimation process that modifies component motions and narrows the conical geometry on a scale of 100 pc.

Traditional undergraduate science courses often prioritize content mastery over authentic engagement with the scientific process. Course-based research, also referred to as course-based undergraduate research experiences (CUREs), addresses this limitation by immersing students in authentic scientific practice. In course-based research, assessment practices can also mirror the authentic scientific practice, where extensive formative feedback supports refinement of skills and understanding. Here, we present two rubrics designed to support the teaching and assessment of science communication in a way that reflects how scientists prepare to disseminate their research findings. One rubric is for creating scientific posters and another for writing short-format manuscripts. Developed by approximately 100 faculty members who collaboratively implement CUREs through the Howard Hughes Medical Institute (HHMI) Science Education Alliance (SEA) program, these rubrics outline the authentic steps scientists take when preparing to communicate their research and provide performance levels that clarify expectations for both students and instructors. Together, these tools aim to further align undergraduate science education and authentic scientific practice.