Treatment Planning Research Articles

Unlocking the potential of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) in revolutionizing type 2 diabetes management: A comprehensive review

Diabetes mellitus (DM) is a long-term metabolic disorder caused by inadequate production and resistance to insulin. The prevalence of DM is rapidly increasing, with type 2 diabetes (T2D) accounting for more than 90% of cases. Despite new treatments, many patients with T2D do not meet their glycemic targets due to clinical inertia. This review provides an overview of glucagon-like peptide 1 receptor agonists (GLP-1 RAs) in the management of T2D. The review synthesizes data from clinical trials and meta-analyses on the efficacy, safety, and cost-effectiveness of GLP-1 RAs. It also discusses the mechanisms of action, classification, and barriers to adherence and persistence in therapy. GLP-1 RAs improve glycemic control by lowering A1C levels and promoting weight loss. They have cardioprotective effects and may reduce endothelial inflammation, oxidative stress, and blood pressure. Adherence to GLP-1 RAs is better with once-weekly injections, though gastrointestinal side effects and cost can affect persistence. Semaglutide and liraglutide have shown significant weight reduction, with semaglutide being particularly effective. GLP-1 RAs are cost-effective due to reduced healthcare costs associated with fewer hospitalizations and lower mortality rates. Safety concerns include gastrointestinal issues, pancreatitis, and rare cases of diabetic retinopathy and thyroid C-cell tumors. For clinical practice, GLP-1 RAs represent a valuable option not only for glycemic control but also for weight management and cardiovascular protection. Incorporating GLP-1 RAs into treatment plans can improve patient outcomes, and optimizing dosing regimens and addressing barriers such as cost and side effects are crucial to enhancing patient adherence and long-term treatment success.

Developing a Statewide Strategic Plan for Prevention, Early Identification, and Treatment of Psychosis.

Access to evidence-based programs that address early psychosis is a substantial public health concern. The authors describe the community-engaged, data-driven process that informed the development of the Massachusetts Strategic Plan for Early Psychosis, an effort to identify actionable priorities to build a system of prevention and care that responds to the needs of individuals who experience psychosis and their caregivers. A multiphase, mixed-methods approach was used to gather knowledge from young adults experiencing early psychosis and their caregivers, including two symposia with diverse stakeholders. Six overarching goals were identified, each with possible action steps and stigma reduction strategies: connecting and supporting individuals who experience psychosis and their families, promoting early identification of and intervention for psychosis through community education, providing specialized support to key community members, providing specialized support to medical and behavioral health care professionals, supporting specialized treatment teams in the delivery of evidence-based care, and developing a statewide system of psychosis services. Next steps for and operationalization of the statewide strategic plan for psychosis in Massachusetts will require a population health approach that engages the community through intersectoral and multisectoral strategies.

Radiomic Features Based Automatic Classification of CT Lung Findings for COVID-19 Patients.


The lung CT images of COVID-19 patients can be typically characterized by three differentfindings-
Ground Glass Opacity (GGO), consolidation and pleural effusion. GGOs have been shown to precede consolidations and has different heterogeneous appearance. Conventional severity scoring only uses total area of lung involvement ignoring appearance of the effected regions. This study proposes a baseline to select heterogeneity/radiomic features that can distinguish these three pathological lung findings.
Methods:
Four approaches were implemented to select features from a pool of 44 features. First one is a manual feature selection method. The rest are automatic feature selection methods based on Genetic Algorithm (GA) coupled with 1) K-Nearest-Neighbor (GA-KNN), 2) binary-decision-tree (GA-BDT) and 3) Artificial-Neural-
Network (GA-ANN). For the purpose of validation, an ANN was trained using the selected features and tested on a completely independent data set.
Results:
Manual selection of nine radiomic features was found to provide the most accurate results with the highest sensitivity, specificity and accuracy (85.7% overall accuracy and 0.90 area under receiver operating characteristic curve) followed by GA-BDT, GA-KNN and GA-ANN (accuracy 78%, 77.5% and 76.8%).
Conclusion:
Manually selected nine radiomic features can be used in accurate severity scoring allowing the clinician to plan for more effective personalized treatment. They can also be useful for monitoring the progression of COVID-19 and response to therapy for clinical trials.&#xD.

Abstract 4132657: Age-related Differences in Peak Oxygen Uptake in Patients with Multimorbidity Undergoing Cardiac Rehabilitation

Background: Cardiac rehabilitation (CR) patients demonstrate a high burden of chronic conditions (CC). Both multimorbidity (MM; ≥2 coexisting CC) and aging negatively impact functional capacity, whereas CR improves performance. Age-related difference in cardiorespiratory fitness (CRF, peak oxygen uptake (VO 2 )) in CR patients with MM has not been studied. Hypothesis: We hypothesized improvement in CRF will be attenuated by MM with increasing age in CR patients. Aims: We aimed to identify age-related differences in CRF improvement in CR patients with MM. Methods: Patients ≥18 yrs old who attended ≥1 CR sessions from 1999-2017 and completed cardiopulmonary exercise stress test before and after CR were included. The prevalence of CC was assessed using Rochester Epidemiology Project records-linkage system. Age categories included: Younger (18-49 yrs), Midlife (50-64 yrs) and Older (≥65 yrs). CRF categories included: <0%, 0-8% and >8%. Analysis included Kruskal-Wallis and Chi-squared. Results: Of 622 patients, 75.4% were male. Mean age: 62.9±11.2 yrs; Younger (n=70) age: 42.7±6.8 yrs, 67.1% male; Midlife (n=283) age: 58.3±4.3 yrs, 78.8% male; Older (n=269) age: 73.0±5.2 yrs, 74.0% male. BMI was 29.9±5.4 kg/m 2 (Younger: 30.0±6.7, Midlife: 30.4±5.6, Older: 29.3±4.7; p<.05). Median (Q1, Q3) no. of CR sessions attended was 33(22, 35), with lower attendance in the Younger group: 25(16,34) vs Midlife: 33(20,35) and Older: 34 (29,35), p<.05. Overall, percent change in peak VO 2 was 11.5±21.6% (Younger: 18.8±22.8%, Midlife: 11.6±19.5%, Older: 9.4±23.1%; p<.01). The median number of CC was 7(5,8) with significant differences across groups: Younger: 5(3,7), Midlife: 6(5,8), Older: 7(6,9), p<.01. Number of CC increased with age across all CRF categories (p<.01, Fig. 1). However, there was no difference in number of CC within age groups across categories of CRF, Table 1. Conclusions: Our results suggest that MM increases with age in patients attending CR; however, within a given age group, improvement in CRF does not appear dependent on MM. The impact of age appears to have a greater impact than MM on CRF improvement. Individualized treatment plans should focus on the impact of age-related disparities on CRF improvement among CR patients with MM.

Drug Sensitivity Prediction Based on Multi-stage Multi-modal Drug Representation Learning.

Accurate prediction of anticancer drug responses is essential for developing personalized treatment plans in order to improve cancer patient survival rates and reduce healthcare costs. To this end, we propose a drug sensitivity prediction model based on multi-stage multi-modal drug representations (ModDRDSP) to reflect the properties of drugs more comprehensively, and to better model the complex interactions between cells and drugs. Specifically, we adopt the SMILES representation learning method based on the deep hierarchical bi-directional GRU network (DSBiGRU) and the molecular graph representation learning method based on the deep message-crossing network (DMCN) for the multi-modal information of drugs. Additionally, we integrate the multi-omics information of cell lines based on a convolutional neural network (CNN). Finally, we use an ensemble deep forest algorithm for the prediction of drug sensitivity. After validation, the ModDRDSP shows impressive performance which outperforms the four current industry-leading models. More importantly, ablation experiments demonstrate the validity of each module of the proposed model, and case studies show the good results of ModDRDSP for predicting drug sensitivity, further establishing the superiority of ModDRDSP in terms of performance.

Deep learning prediction of scenario doses for direct plan robustness evaluations in IMPT for head-and-neck

Objective. Intensity modulated proton therapy (IMPT) is susceptible to uncertainties in patient setup and proton range. Robust optimization is employed in IMPT treatment planning to ensure sufficient coverage of the clinical target volume (CTV) in predefined scenarios, albeit at a price of increased planning times. We investigated a deep learning (DL) strategy for dose predictions in individual error scenarios in head and neck cancer IMPT treatment planning, enabling direct evaluation of plan robustness. The model is able to differentiate between scenarios by using embeddings of the scenario index.Approach. To accommodate resolution disparities in planning CT-scans and accommodate the setup error scenarios, we introduced scenario-specific isocentric distance maps as inputs to the DL models. For 392 H&N cancer patients, high-quality 9-scenario ground truth (GT) robust plans were generated with wish-list driven fully automated multi-criteria optimization. The scenario index is converted to one-hot-vector that is used to derive the scenarios embeddings through the training of the DL model, aiding the model to predict a scenario specific dose distribution.Main results. The model achieved within 1%-point of agreement with the GT the predictedV95%of the voxelwise minimum dose for CTV Low and CTV High for 96% and 75% respectively of the test patients. Considering all robustness scenarios, median differences were 0.035%-point for CTV HighV95%, 0.11%-point for CTV LowV95%, 0.29 GyE for parotidsDmean, 0.7 GyE for submandibular glandsDmeanand 0.9 GyE for oral cavityDmean. Prediction of full 3D dose distributions for all scenarios took around 14 s.Significance. Predicting individual scenarios for robust proton therapy using DL dose prediction is feasible, enabling direct robustness evaluation of the predicted scenario doses.

Abstract 4147627: A Pharmacovigilance Study of Bempedoic Acid: Balancing Risks and Benefits

Introduction: Bempedoic acid has been approved by FDA since Feb 2020 as an adjunct to maximally tolerated statin therapy for the reduction of LDL-C, specifically in individuals afflicted by atherosclerotic cardiovascular disease (ASCVD) and heterozygous familial hypercholesterolemia (HeFH). Bempedoic acid is also a viable alternative for patients intolerant to statins. Most recently, it has been approved to use for primary prevention of hyperlipidemia, alone or in combination with a statin. It is important to study its side effect profile as it is anticipated to be used more frequentlynin the future. Methods: We explored reported adverse events (AEs) of bempedoic acid using FDA Adverse Event Reporting System (FAERS). A total of 1202 adverse effects were reported until March 31, 2024. 69.97% AEs were reported by healthcare professionals (HCPs), which were further analyzed. AEs were analyzed in men (32.53%) vs women (53.24%) and in three age groups of 18-64, 65-85, and >85 years. Results: A total of 586 (48.7%) events were serious adverse effects. Most common side effects were myalgia which included muscle pain, extremity pain and muscle spasm (25.04%), arthralgia including back pain and bone pain (18.47%), nausea (6.6%), headache(5.5%), gout (4.4%). Among the patients with adverse effects, 4.75% had cardiac side effects which included atrial fibrillation (12%), palpitation (29%), angina pectoris (18%), Myocardial Infarction (5%). Discussion: A thorough understanding of bempedoic acid's complex pharmacology empowers healthcare professionals to tailor treatment plans based on individual patient requirements. This enables informed decision-making in prescribing, optimizing dosage regimens, and mitigating potential adverse reactions. Most studies have been relatively short term in order to confirm the long term side effect profile. It would be beneficial to be cautious in using this medication for patients with above mentioned clinical conditions like myalgia, arthralgia, gout, rhabdomyolysis, atrial fibrillation, angina pectoris. Conclusion: Long term pharmacovigilance is needed to assess the safety profile of bempedoic acid over extended use. With the current increasing use of bempedoic acid, there is a need for more data on safety profiles. In future, further randomized trials are required to better assess the safety profile.

CLINICAL CASE OF CONSERVATIVE MANAGEMENT OF INFEROMEDIAL ORBITAL WALL FRACTURE

Justification. Orbital fractures are one of the most common injuries of the middle zone of the face, second only to damage to the bones of the nose in adults and the lower jaw in children. An increase in the frequency of facial skeletal injuries, the complexity of diagnosis, multidisciplinary interaction, ambiguity and the lack of uniform algorithms for the treatment and rehabilitation of such patients increase interest in studying the topic of orbital fractures, require improved approaches in diagnosis and treatment.Objective: Demonstration of a clinical case of conservative management of an inferomedial orbital fracture.Methods. While getting off the bus, the patient tripped and hit the left side of her face on the curb. She sought medical help. A diagnosis was made: Inferomedial orbital fracture, surgical treatment was proposed. The patient refused surgery. Conservative treatment was carried out.Results. After 3 months from the moment of diagnosis, almost complete restoration of functions and absence of aesthetic deformation was achieved. Conclusions. The optimal treatment plan is often variable and depends on the clinical situation, extent of injury, and patient comfort level.

Investigating the Presence of Autistic Traits and Prevalence of Autism Spectrum Disorder Symptoms in Anorexia Nervosa: A Systematic Review and Meta-Analysis.

The present meta-analysis aims to assess whether individuals with anorexia nervosa (AN) demonstrate elevated autistic traits, to explore potential associations between autistic traits and eating disorder symptoms; as well as to estimate the prevalence of a positive screen for autism spectrum disorder (ASD) assessed via Autism Diagnostic Observation Schedule (ADOS), in AN. A systematic literature search was conducted in PsycINFO, MEDLINE, and Web of Science in August 2023 and later updated in April 2024 to identify relevant studies. Twenty-two studies with 1172 AN patients and 2747 healthy controls (HCs) met the inclusion criteria. There was a significant difference between AN and HC groups in autistic traits (g = 0.88, CI = 0.65-1.12), and a significant but modest correlation was found between autistic traits and severity of eating disorder symptoms (r = 0.28, CI = 0.11-0.44). Proportion meta-analysis indicated that 29% (CI = 0.19-0.38) of children and adults scored above the cut-off for ASD. The current findings suggest that AN frequently overlaps with both autistic traits and autistic symptomatology. Therefore, it is essential to evaluate autism and autistic traits in individuals with AN to tailor individualized treatment plans.

AI bias in lung cancer radiotherapy

Aim: In lung cancer research, AI has been trained to read chest radiographs, which has led to improved health outcomes. However, the use of AI in healthcare settings is not without its own set of drawbacks, with bias being primary among them. This study seeks to investigate AI bias in diagnosing and treating lung cancer patients. The research objectives of this study are threefold: 1) To determine which features of patient datasets are most susceptible to AI bias; 2) to then measure the extent of such bias; and 3) from the findings generated, offer recommendations for overcoming the pitfalls of AI in lung cancer therapy for the delivery of more accurate and equitable healthcare. Methods: We created a synthetic database consisting of 50 lung cancer patients using a large language model (LLM). We then used a logistic regression model to detect bias in AI-informed treatment plans. Results: The empirical results from our synthetic patient data illustrate AI bias along the lines of (1) patient demographics (specifically, age) and (2) disease classification/histology. As it concerns patient age, the model exhibited an accuracy rate of 82.7% for patients < 60 years compared to 85.7% for patients ≥ 60 years. Regarding disease type, the model was less adept in identifying treatment categories for adenocarcinoma (accuracy rate: 83.7%) than it was in predicting treatment categories for squamous cell carcinoma (accuracy rate: 92.3%). Conclusions: We address the implications of such results in terms of how they may exacerbate existing health disparities for certain patient populations. We conclude by outlining several strategies for addressing AI bias, including generating a more robust training dataset, developing software tools to detect bias, making the model’s code open access and soliciting user feedback, inviting oversight from an ethics review board, and augmenting patient datasets by synthesizing the underrepresented data.

Real-world outcomes of everolimus-based treatment in a Taiwanese cohort with metastatic HR+/HER2- breast cancer.

Everolimus was the first orally targeted therapy for certain cancers. It was introduced before CDK4/6 inhibitors and is widely used to treat advanced hormone receptor-positive (HR+)/human epidermal growth factor receptor 2-negative (HER2-) breast cancer. This study presents comprehensive findings including updated data and long-term survival analyses focusing on patients with HR+/HER2- metastatic breast cancer who received everolimus-based treatment. The objectives were to assess the impact of everolimus on overall survival (OS) and progression-free survival (PFS) by treatment line, and to evaluate its role in therapeutic strategies in a real-world setting. We included 299 women aged over 20 years with histologically confirmed HR+/HER2- breast cancer who received everolimus-based treatment from multiple medical centers in Taiwan. Survival curves were generated using the Kaplan-Meier method, with the log-rank test for comparisons. Univariate and multivariate analyses were performed using a Cox proportional hazards regression model. Adverse effects were graded according to the Common Terminology Criteria for Adverse Events version 5.0. The median PFS was 5.6 months, and the median OS was 60.1 months. Patients receiving everolimus treatment in three or more lines and those who underwent chemotherapy prior to everolimus-based treatment had a significantly shorter PFS but longer OS. Patients with liver and central nervous system metastases had significantly shorter PFS and OS. The disease control rate was 51.5%, and the overall response rate was 8.0%. These findings support current guidelines and advocate for the inclusion of everolimus in treatment plans for patients with metastatic HR+/HER2- breast cancer, particularly in late-line treatment, with careful consideration of the benefit-risk profile for each patient.

Current Concepts of the Management of Painful Traumatic Peripheral Nerve Neuromas

Painful neuromas are a complex clinical condition that results in notable disability and functional impairment after injury to a peripheral nerve. When regenerating axons lack a distal target, they form a stump neuroma. Up to 60% of neuromas are painful because of mechanical sensitivity and crosstalk between nerve fibers. Clinical evaluation includes a thorough history and physical examination followed by directed diagnostic imaging and procedures to assess pain generators and their effect on quality of life. Nonsurgical management options may include pharmacological interventions, desensitization strategies, injections, and therapies to reduce pain perception and improve function. Surgical interventions, such as nerve reconstruction by direct repair or grafting, redirection to alternative targets, and containment of regenerating axons by relocation into innervated tissues or in grafts, are considered when conservative measures fail. A comprehensive and individualized treatment plan is crucial for optimizing patient outcomes with painful neuromas. The plan should consider the underlying pathology, pain generators, and psychosocial factors contributing to the patient's pain.

Feasibility of creating a daily adaptive plan using automatic DIR-created target and OARs contours in patients with prostate cancer magnetic-resonance-guided adaptive radiotherapy.

The purpose of this study was to evaluate the feasibility of treatment plans for prostate cancer with magnetic resonance (MR)-guided online adaptive radiotherapy, which are generated using deformable image registration (DIR)-created contours of the targets and organs. Totally, 150 fractions from 30 prostate cancer patients implanted with a hydrogel spacer and treated with the MR-Linac were studied. Reference treatment plans that satisfied all institutional dose constraints were initially created on planning MRI. The adaptive treatment plans were created on daily MRI based on the reference plan using the DIR-created contours, ensuring all dose constraints were met. Subsequently, a clinician manually created reference contours for each daily MRI. Finally, the dose volume histogram indices of the plan generated with DIR-created contours were re-evaluated with clinician created contours. The evaluated contours included the bladder wall, rectum wall, sigmoid, small bowel and planning target volume (PTV) for dose prescription. The PTV for dose prescription met the dose constraints in all fractions. The bladder and rectum walls met the dose constraint of maximum dose (D0.03 cc) in all fractions. Five patients failed to meet the sigmoid and small bowel dose constraints, with the largest deviation being 13.3% exceedance at D2 cc in the small bowel added 3mm margin. This study suggests that most treatment plans created without modifying the DIR-created contours are clinically viable. However, dislodgements of the small bowel and sigmoid may exceed the extent of DIR propagation from the reference plan contours, and it is recommended that these contours be verified.

Multicriteria optimization achieves superior normal tissue sparing in volumetric modulated arc therapy for gastric cancer.

To evaluate the benefits of volumetric modulated arc therapy (VMAT) based on multicriteria optimization (MCO) for gastric cancer patients, particularly the protection of serial organs at risk (OARs) that overlap with the target volume. MCO and single-criterion optimization (SCO) VMAT plans were conducted among 30 gastric cancer patients, with a prescription dose of 50.4Gy delivered in 28 fractions. All treatment plans underwent review, and a comparison was made between the active planning time and different dose-volume parameters. Both the MCO and SCO VMAT plans achieved the target dose coverage, with no significant difference in the conformity index (CI) for the planning target volume (PTV), at median CI values of 0.887 and 0.891, respectively (P = 0.417). The MCO plans showed a slight but significant increase in the homogeneity index of the PTV, with a median increase of 0.029 (P < 0.001). Additionally, the MCO plans resulted in a lower D2% to the small intestine and duodenum, with reductions of 3.43Gy and 0.3Gy, respectively (P < 0.05). Furthermore, the Dmax to the small intestine correlated moderately with the overlapping volume between the small intestine and the target volume (ρ = 0.42, P = 0.023). Except for the mean dose to the liver, the MCO plans performed better in terms of dose indicators for other OARs. Moreover, compared to the SCO plans, the median active planning time in the MCO plans was significantly reduced by 53.2min (P < 0.0001). MCO can effectively help the physicians to quickly select an optimal treatment plan for patients with gastric cancer. It has been shown that MCO VMAT plans can significantly reduce the dose to OARs and shorten the active planning time, with acceptable target coverage. In addition, these plans take less dosimetric time, thereby streamlining the treatment planning process.

NCOG-04. MEMORY TRACT SPARING USING DIFFUSE TENSOR IMAGING IN RADIATION PLANNING OF PRIMARY BRAIN TUMORS

Abstract INTRODUCTION Radiation therapy (RT) is a critical treatment modality for both primary and metastatic brain tumors, yet approximately 30% of patients experience cognitive decline post-radiotherapy. This cognitive toxicity is linked to low radiation doses affecting neural stem cells in the hippocampal dentate gyrus. Hippocampal Avoidance (HA)-Whole Brain Radiotherapy (WBRT) combined with Memantine has shown promising outcomes in preserving cognitive function and quality of life (QOL) in patients with brain metastases, suggesting it should be the standard of care for those with good performance status and no hippocampal metastases. METHODS We conducted a prospective trial approved by the IRB (SMC0307-23), including patients aged 18 and above with primary brain tumors post-resection or biopsy. Exclusion criteria included multifocal glioma crossing to the other hemisphere. RT was delivered to a total dose of 60Gy in 30 fractions with concomitant TMZ. Diffusion tensor imaging (DTI) was performed to map hippocampal-associated white matter tracts. Memory fiber tracts and hippocampi were contoured and integrated into RT planning using Eclipse treatment planning software. Dosimetric analyses compared two plans: one with memory fiber constraints and one without. RESULTS Twelve patients with low-grade gliomas were included, with successful contouring of memory fibers and hippocampi. VMAT treatment plans met dose constraints for memory fibers, with an average mean dose of 10.1 Gy. The average MoCA score before RT was 27.1 and 26.4 at six months post-treatment, with a p-value of 0.07. Excluding one patient, the scores were 27.1 and 26.6, respectively (p=0.13). CONCLUSION MRI planning using DTI for memory fiber detection and incorporation into RT planning via VMAT techniques enables hippocampal and associated white fiber sparing, potentially preserving cognitive function. Preliminary cognitive data are promising, supporting the need for further validation in a larger cohort.

Integrating Artificial Intelligence Across Medical Clinics: Strengthening Collaborative Efforts for Improved Patient Outcomes

The integration of Artificial Intelligence (AI) across medical clinics holds transformative potential for enhancing patient outcomes through improved collaboration. This review examines the applications of AI in fostering inter-clinic connectivity, with a focus on diagnostics, treatment planning, and patient management. By utilizing AI-powered data-sharing platforms, predictive analytics, and telemedicine solutions, clinics can collaborate more efficiently, ensuring continuity of care and reducing diagnostic errors. Despite these advancements, challenges persist, including data privacy, interoperability, and resistance to AI adoption among clinical staff. This review highlights case studies showcasing successful AI-enabled collaborations and proposes future directions to address current limitations. Overall, AI's role in linking medical clinics underscores its potential to revolutionize patient care by bridging gaps in communication and decision-making processes.

Unveiling the power of artificial intelligence for image-based diagnosis and treatment in endodontics: An ally or adversary?

Artificial intelligence (AI), a field within computer science, uses algorithms to replicate human intelligence tasks such as pattern recognition, decision-making and problem-solving through complex datasets. In endodontics, AI is transforming diagnosis and treatment by applying deep learning algorithms, notably convolutional neural networks, which mimic human brain function to analyse two-dimensional (2D) and three-dimensional (3D) data. This article provides an overview of AI applications in endodontics, evaluating its use in 2D and 3D imaging and examining its role as a beneficial tool or potential challenge. Through a narrative review, the article explores AI's use in 2D and 3D imaging modalities, discusses their limitations and examines future directions in the field. AI significantly enhances endodontic practice by improving diagnostic accuracy, workflow efficiency, and treatment planning. In 2D imaging, AI excels at detecting periapical lesions on both periapical and panoramic radiographs, surpassing expert radiologists in accuracy, sensitivity and specificity. AI also accurately detects and classifies radiolucent lesions, such as radicular cysts and periapical granulomas, matching the precision of histopathology analysis. In 3D imaging, AI automates the segmentation of fine structures such as pulp chambers and root canals on cone-beam computed tomography scans, thereby supporting personalized treatment planning. However, a significant limitation highlighted in some studies is the reliance on in vitro or ex vivo datasets for training AI models. These datasets do not replicate the complexities of clinical environments, potentially compromising the reliability of AI applications in endodontics. Despite advancements, challenges remain in dataset variability, algorithm generalization, and ethical considerations such as data security and privacy. Addressing these is essential for integrating AI effectively into clinical practice and unlocking its transformative potential in endodontic care. Integrating radiomics with AI shows promise for enhancing diagnostic accuracy and predictive analytics, potentially enabling automated decision support systems to enhance treatment outcomes and patient care. Although AI enhances endodontic capabilities through advanced imaging analyses, addressing current limitations and fostering collaboration between AI developers and dental professionals are essential. These efforts will unlock AI's potential to achieve more predictable and personalized treatment outcomes in endodontics, ultimately benefiting both clinicians and patients alike.

Inter-center comparison of proton range verification prototypes with an anthropomorphic head phantom**F. Hueso-González and J. Berthold share first authorship. J. Verburg and C. Richter share last authorship. P. Wohlfahrt was essential for the work conception, treatment planning design and equalization, and for the transatlantic transport of the phantom.

Objective. To compare in reproducible and equalized conditions the performance of two independent proton range verification systems based on prompt gamma-ray detectors from two different proton therapy centers.Approach. An anthropomorphic head phantom with calibrated stopping power, serving as ground truth, was irradiated with comparable treatment plans, spot positions and energies in both facilities. Clinical beam current, tumor contour and dose were used. The absolute range measurement was compared to the expected value according to the ground truth. The statistical precision was assessed by repeating each measurement ten times. Sensitivity to relative range shifts was evaluated by introducing 2 mm and 5 mm plastic slabs on half of the field.Main results. The resulting absolute range accuracy was within 2.4 mm in all cases. Relative range shifts were detected with deviations lower than 14%.Significance. The performance of both systems was deemed worthy of clinical application for the detection of range deviations. This study represents the first comparison of independent prompt gamma-ray-based proton range verification systems under equalized conditions with realistic treatment fields and beam currents.

Relevance of optic nerve B-mode sonography in multiple sclerosis diagnosis and follow-up

Abstract Background Among young people, multiple sclerosis (MS) accounts for the majority of non-traumatic neurological impairments. Almost all MS patients develop optic neuritis as the illness progresses. In MS, the optic nerve (ON) is vulnerable to demyelinating injury and is considered “eloquent” in comparison with the brain. Thus, it serves as a clinical model for MS. Since the eye is a suitable target for ultrasonography due to its high-water content, non-invasive trans-orbital sonography (TOS) can determine the optic nerve sheath diameter (ONSD). This research aimed to determine the effectiveness of ON ultrasonography as a simple, low-cost method for diagnosing and following up MS patients, facilitating their treatment plan. It correlates the findings of ONSD using TOS and visual-evoked potential (VEP), the number of magnetic resonance imaging (MRI) black holes, and other clinically significant parameters in 30 relapsing–remitting MS (RRMS) patients versus 30 controls. Results We examined a total of 120 eyes. The ONSD values at 3 mm, 5 mm, and myelination index were significantly lower in the patients (4.7 ± 0.3, 5.5 ± 0.3, and 0.86 ± 0.03, respectively) versus the controls (5.6 ± 0.3, 6.2 ± 0.4, and 0.90 ± 0.03, respectively) (p &lt; 0.001 in each). The mean P 100 latency was significantly delayed in the patients (125.5 ± 6.1 ms) than the controls (100.2 ± 3.3 ms), (p &lt; 0.001). The mean VEP amplitude was significantly lower in the patients (4.9 ± 0.7 μV) than the controls (6.8 ± 0.8 μV) (p &lt; 0.001). There was a highly significant positive correlation between the ONSD and P 100 amplitude, and highly significant negative correlations between the ONSD and disease duration, number of optic neuritis attacks, EDSS, MRI black holes number, and P 100 latency. Conclusions Trans-orbital sonography is a practical and easily accessible approach for assessing ON atrophy in MS. It can measure axonal loss and brain atrophy indirectly and is a reliable paraclinical diagnostic tool, suggesting that ONSD could be a biomarker of disease activity.

RADT-41. RADIOSURGERY FOR GLOMUS TUMORS: IMPACT OF MRI-PET FUSION ON TREATMENT PLANNING AND OUTCOMES

Abstract Paragangliomas present a challenge for stereotactic radiosurgery (SRS) treatment planning, particularly regarding the delineation of tumor boundaries. We propose that integrating 68Ga-DOTATATE PET scans alongside MRI in SRS planning will facilitate superior target identification and tumor coverage compared to reliance on MRI alone. A retrospective analysis of 9 patients diagnosed with paraganglioma who underwent MRI and 68Ga-DOTATATE PET imaging over a 20-month period was conducted. Images were imported into GammaPlan treatment planning software, partitioned into separate sessions: one with MRI and another with MRI-PET fusion. A single evaluator contoured paragangliomas as gross tumor volume (GTV) initially using MRI alone and subsequently with PET/MRI fusion. The pre- and post-PET fusion GTV volumes were compared. The standard deviation of the percentage increases in volume was calculated to assess variability across the sample. Of included 9 patients, 7 were females and median age was 58.3 years. Four patients had prior surgery. Seven patients were treated with Gamma Knife, while 2 with LINAC. Two patients were treated with single fraction SRS while 7 with hypofractionated SRS. Across all treated patients (n=9), GTV volumes significantly increased with PET fusion, with an average volume increase of 26.02% (n=8). One patient reported dryness of throat after SRS. Significant PET avidity over structures like temporal bone, internal jugular vein, petrous internal carotid artery and sigmoid sinus-IJV junction were included in PET/MRI fusion GTVs, which were missed on isolated MRI contours. At median 6-month follow up, all patients had a stable radiological response. Integration of 68Ga-DOTATATE PET scans with MRI enhances the visualization of paragangliomas in patients undergoing SRS. Further extensive studies are warranted to validate this observation and to better understand the variability in treatment response observed across the patient population. Our practice has initiated routine utilization of DOTATATE-PET imaging in paraganglioma patients undergoing SRS planning.