Explore the vast range of titles available.

The abscopal effect is a phenomenon relating to the treatment of metastatic cancer in which localized irradiation to a tumor concurrently causes shrinkage of tumors distant from the area of treatment. Localized radiotherapy is thought to cause anti-tumor immunologic responses that lead to regression and remission of cancers distant to the initial location of treatment. We present a 47-year-old male with brain metastasis from non-small cell lung cancer (NSCLC) who went into remission following stereotactic radiosurgery treatment to a brain lesion, in the absence of systemic treatment. We discuss the novelty of this case and its importance to future research on the abscopal effect. Though it is difficult to distinguish the abscopal effect from spontaneous remission of non-targeted cancer, this report sheds insight on the potential for improving treatment for the leading cause of cancer death worldwide.

The worldwide average exposure to cosmic rays contributes to about 16% of the annual effective dose from natural radiation sources. At ground level, doses from cosmic ray exposure depend strongly on altitude, and weakly on geographical location and solar activity. With the analytical model PARMA developed by the Japan Atomic Energy Agency, annual effective doses due to cosmic ray exposure at ground level were calculated for more than 1,500 communities across Canada which cover more than 85% of the Canadian population. The annual effective doses from cosmic ray exposure in the year 2000 during solar maximum ranged from 0.27 to 0.72 mSv with the population-weighted national average of 0.30 mSv. For the year 2006 during solar minimum, the doses varied between 0.30 and 0.84 mSv, and the population-weighted national average was 0.33 mSv. Averaged over solar activity, the Canadian population-weighted average annual effective dose due to cosmic ray exposure at ground level is estimated to be 0.31 mSv.

Purpose To examine dosimetric and clinical outcomes for Cs-131 radioactive seed implant compared to Pd-103 and I-125.  Background/Significance Cs-131 is a novel isotope with relatively short half-life (9.7 days) that may have clinical advantages in seed implant treatments of prostate cancers. There may be a shorter duration of symptoms and increased PSA control rates. Methods We performed a retrospective study in which clinical and dosimetric outcomes were compared for 186 prostate implants performed over a ten-year time period at three different Ascension hospitals. Isotopes that were used included Cs-131 (n=66; half-life 9.7 days), I-125 (n=60; half-life 60 days), and Pd-103 (n=60; half-life 17 days) Results The implants used standard radiation dosages. These were 145 Gy for I-125 alone or 109 Gy when combined with external beam radiation. In the case of Cs-131 used alone, the dose was 115 Gy or 85 Gy when combined with an external beam. For Pd-103, 125 Gy was used for monotherapy and 90 Gy when combined with an external beam. The Cs-131 dosimetry was found to be similar to I-125 and Pd-103 on a quantitative basis. However, there was better homogeneity, and the delivered activity per seed and the number of seeds employed were greater compared to other isotopes. We compared the corrected total source strengths (i.e. normalized to sample mean values) and were able to demonstrate similar distributions for the three isotopes. Dosimetric analysis also suggested there was superior homogeneity with Cs-131. The median PSA value at 60 months was 0.11 ng/ml. There were only a few PSA failures in the three groups of cases, nonetheless, the Cs-131 had the fewest.  Conclusions One attractive option for men with early-stage prostate cancer is interstitial brachytherapy. The use of the shorter-acting Cs-131 isotope may be expected to have dose-related side effects that resolve more rapidly. This series suggests a trend for improved PSA control outcomes for Cs-131 patients compared with I-125 and Pd-103.

Patients with neurological diseases are vulnerable to cerebral ischemia, which can lead to brain injury. In the intensive care unit (ICU), neuromonitoring techniques that can detect flow reductions would enable timely administration of therapies aimed at restoring adequate cerebral perfusion, thereby avoiding damage to the brain. However, suitable bedside neuromonitoring methods sensitive to changes of blood flow and/or oxygen metabolism have yet to be established.Near-infrared spectroscopy (NIRS) is a promising technique capable of noninvasively monitoring flow and oxygenation. Specifically, diffuse correlation spectroscopy (DCS) and time-resolved (TR) NIRS can be used to monitor blood flow and tissue oxygenation, respectively, and combined to measuring oxidative metabolism. The work presented in this thesis focused on advancing a DCS/TR-NIRS hybrid system for acquiring these physiological measurements at the bedside.The application of NIRS for neuromonitoring is favourable in the neonatal ICU since the relatively thin scalp and skull of infants has minimal effect on the detected optical signal. Considering this application, the validation of a combined DCS/NIRS method for measuring the cerebral metabolic rate of oxygen (CMRO2) was investigated in Chapter 2. Although perfusion changes measured by DCS have been confirmed by various flow modalities, characterization of photon scattering in the brain is not clearly understood. Chapter 3 presents the first DCS study conducted directly on exposed cortex to confirm that the Brownian motion model is the best flow model for characterizing the DCS signal. Furthermore, a primary limitation of DCS is signal contamination from extracerebral tissues in the adult head, causing CBF to be underestimated. In Chapter 4, a multi-layered model was implemented to separate signal contributions from scalp and brain; derived CBF changes were compared to computed tomography perfusion.Overall, this thesis advances DCS techniques by (i) quantifying cerebral oxygen metabolism, (ii) confirming the more appropriate flow model for analyzing DCS data and (iii) demonstrating the ability of DCS to measure CBF accurately despite the presence of a thick (1-cm) extracerebral layer. Ultimately, the work completed in this thesis should help with the development of a hybrid DCS/NIRS system suitable for monitoring cerebral hemodynamics and energy metabolism in critical-ill patients.

Radon decays to a long-lived isotope ²¹⁰Pb with a half-life of about 22 years. Measuring concentrations of ²¹⁰Pb in household dust could be an alternative method of determining indoor radon levels. This novel method for estimating long-term radon concentration was explored in over a hundred Canadian residential homes. The results demonstrate that ²¹⁰Pb concentrations in household dust relate reasonably well to radon concentrations in homes.

Radon decays to a long-lived isotope super(210)Pb with a half-life of about 22years. Measuring concentrations of super(210)Pb in household dust could be an alternative method of determining indoor radon levels. This novel method for estimating long-term radon concentration was explored in over a hundred Canadian residential homes. The results demonstrate that super(210)Pb concentrations in household dust relate reasonably well to radon concentrations in homes.

Radon decays to a long-lived isotope 210 Pb with a half-life of about 22 years. Measuring concentrations of 210 Pb in household dust could be an alternative method of determining indoor radon levels. This novel method for estimating long-term radon concentration was explored in over a hundred Canadian residential homes. The results demonstrate that 210 Pb concentrations in household dust relate reasonably well to radon concentrations in homes.

Large Language Models (LLMs) offer significant potential for clinical symptom extraction, but their deployment in healthcare settings is constrained by privacy concerns, computational limitations, and operational costs. This study investigates the optimization of compact LLMs for cancer toxicity symptom extraction using a novel iterative refinement approach. We employ a student-teacher architecture, utilizing Zephyr-7b-beta and Phi3-mini-128 as student models and GPT-4o as the teacher, to dynamically select between prompt refinement, Retrieval-Augmented Generation (RAG), and fine-tuning strategies. Our experiments on 294 clinical notes covering 12 post-radiotherapy toxicity symptoms demonstrate the effectiveness of this approach. The RAG method proved most efficient, improving average accuracy scores from 0.32 to 0.73 for Zephyr-7b-beta and from 0.40 to 0.87 for Phi3-mini-128 during refinement. In the test set, both models showed an approximate 0.20 increase in accuracy across symptoms. Notably, this improvement was achieved at a cost 45 times lower than GPT-4o for Zephyr and 79 times lower for Phi-3. These results highlight the potential of iterative refinement techniques in enhancing the capabilities of compact LLMs for clinical applications, offering a balance between performance, cost-effectiveness, and privacy preservation in healthcare settings.

This study introduces a novel teacher-student architecture utilizing Large Language Models (LLMs) to improve prostate cancer radiotherapy symptom extraction from clinical notes. Mixtral, the student model, initially extracts symptoms, followed by GPT-4, the teacher model, which refines prompts based on Mixtral's performance. This iterative process involved 294 single symptom clinical notes across 12 symptoms, with up to 16 rounds of refinement per epoch. Results showed significant improvements in extracting symptoms from both single and multi-symptom notes. For 59 single symptom notes, accuracy increased from 0.51 to 0.71, precision from 0.52 to 0.82, recall from 0.52 to 0.72, and F1 score from 0.49 to 0.73. In 375 multi-symptom notes, accuracy rose from 0.24 to 0.43, precision from 0.6 to 0.76, recall from 0.24 to 0.43, and F1 score from 0.20 to 0.44. These results demonstrate the effectiveness of advanced prompt engineering in LLMs for radiation oncology use.

Natural Language Processing (NLP) is a key technique for developing Medical Artificial Intelligence (AI) systems that leverage Electronic Health Record (EHR) data to build diagnostic and prognostic models. NLP enables the conversion of unstructured clinical text into structured data that can be fed into AI algorithms. The emergence of the transformer architecture and large language models (LLMs) has led to remarkable advances in NLP for various healthcare tasks, such as entity recognition, relation extraction, sentence similarity, text summarization, and question answering. In this article, we review the major technical innovations that underpin modern NLP models and present state-of-the-art NLP applications that employ LLMs in radiation oncology research. However, these LLMs are prone to many errors such as hallucinations, biases, and ethical violations, which necessitate rigorous evaluation and validation before clinical deployment. As such, we propose a comprehensive framework for assessing the NLP models based on their purpose and clinical fit, technical performance, bias and trust, legal and ethical implications, and quality assurance, prior to implementation in clinical radiation oncology. Our article aims to provide guidance and insights for researchers and clinicians who are interested in developing and using NLP models in clinical radiation oncology.