Citations

Abstract Background Markerless tumor tracking (MTT) using single‐energy (SE) kilovoltage (kV) imaging has been proposed as a technique for lung tumor motion management. However, bony structures can obscure the tumor and make tracking challenging. Dual‐energy (DE) subtraction imaging can suppress the bone and therefore enhance tumor visibility, potentially improving tracking accuracy. Purpose To develop a machine learning model that predicts the reliability of MTT for SE and DE imaging using phantom and patient data. Methods Images were collected at high and low energy (120/60 kV) using fast‐kV switching on dynamic thorax phantom with 5–15 diameter targets and 20 lung cancer patients, including one patient with two tumor locations. DE images were then generated off‐line using weighted logarithmic subtraction. Separately, a template‐based tracking algorithm was used to track the tumor on both phantom and patient images. For phantom data, the ground truth (GT) was defined by the cos 4 waveform programmed into the motion controller. Since the GT was unknown for patient data, it was estimated using a Kalman filter and further refined using visual inspection. The tracking success rate (TSR) was subsequently calculated. Based on TSR, patients were stratified into three tracking groups: good, moderate, and poor. A logistic regression (LR) model was implemented with tracking features, including match score, peak to side‐peak ratio (PSR), x ‐ and y ‐velocity. To address the distribution differences between phantom and patient data, a domain indicator and interaction features were also incorporated. The model was trained on phantom and 17 patient data using leave‐one‐out cross‐validation, and its performance was evaluated on three held‐out patients, one from each tracking group. Separate LR models were developed for each imaging modality, resulting in SE‐LR and DE‐LR models. For each held‐out patient, the confusion matrix, sensitivity, and specificity were reported. Results For both SE‐LR and DE‐LR models, PSR has the largest coefficient value, indicating that it is the most influential factor when determining the reliability of tracking. At a specificity of 95%, the DE‐LR model achieved significantly higher sensitivity than the SE‐LR model (0.908 vs. 0.731, p < 0.01 for phantom data and 0.657 vs. 0.589, p < 0.01 for patient data). In the three held‐out patients, the DE‐LR model outperformed the SE‐LR model across all metrics. The largest gains were found in the patient with good tracking quality. Conclusions The proposed LR models successfully predicted the reliability of SE‐ and DE‐MTT. Across phantom and patient datasets, the DE‐LR model consistently demonstrated improved performance compared with the SE‐LR model, suggesting that DE imaging may be beneficial for MTT applications.

Background: The role of venous Doppler ultrasound in assessing venous congestion in the setting of post-lung transplantation surgery is unknown. We performed a feasibility study on the use of venous Doppler ultrasound in the immediate post-operative period after lung transplantation in a retrospective, single-center observational evaluation. Patients had at least two venous congestion Doppler ultrasound exams involving interrogation of IVC diameter, hepatic vein, portal vein and renal vein Doppler assessments. Results: A total of 9 patients were included in our study with a median age of 60 years, 66.7% being male, and a median BMI of 20.9. All patients had VExUS grade ≥1 on their first venous Doppler exam. All nine patients had a decrease in portal vein pulsatility from initial to final exams. Correlation between portal venous pulsatility and net fluid balance was not statistically significant R= 0.33 (P= 0.39). Other parameters, including VExUS grade, IVC diameter, HV and RV Doppler changes from initial to final ultrasound exams showed variable changes in the post-operative period. Conclusion: Venous congestion Doppler ultrasound assessments appear feasible in immediate postoperative lung transplant patients. A multitude of factors beyond fluid status likely affect waveforms.

Background Tacrolimus dosing after kidney transplantation is complex, highly individualized, and prone to variability, which can impact graft outcomes. While machine learning (ML) approaches have been used primarily to predict tacrolimus concentrations, large language models (LLMs) may enable protocol-constrained clinical decision support by generating dosing recommendations aligned with established treatment guidelines. Methods We developed TacroDose AI, a protocol-constrained LLM based on GPT-4, to generate tacrolimus dosing recommendations aligned with institutional guidelines. Using 300 structured simulated clinical scenarios, we evaluated protocol adherence, error patterns, and reproducibility of model-generated dosing recommendations. Following review of initial outputs, a refined prompt ( TacroAI 2.0 ) was implemented to strengthen protocol constraints and structured output verification, and performance was reassessed. Results TacroDose AI achieved 72.9% protocol adherence, with 27.1% protocol deviations, including 13.9% sentinel errors with potential clinical significance. Prompt refinement substantially improved performance: TacroAI 2.0 achieved 91.7% adherence, reduced sentinel errors to 1.0%, and improved reproducibility from 77.3 to 86.7% (all p &lt; 0.01). Although rounding discrepancies increased, these represented minor numerical differences without clinical relevance. Performance improvements were observed across subtherapeutic, therapeutic, and supratherapeutic tacrolimus scenarios. Conclusion Protocol-constrained LLMs can produce guideline-concordant tacrolimus dosing recommendations with high adherence and improved reliability in simulated clinical settings. These findings highlight a potential framework for integrating generative AI with rule-based clinical protocols to support safe medication management. With further validation using real-world data, such systems could be integrated into electronic health records to support clinician-supervised, protocol-based immunosuppressive dosing in transplant care.

Multiple sclerosis (MS) is an immune-mediated chronic inflammatory and degenerative disease of the central nervous system (CNS). Typically occurring in young and middle-aged individuals, untreated MS can have high rates of disability and recurrence, thereby imposing a significant burden on the patient, their family, and society. Many factors are implicated in the etiology of MS, with the relationship between Epstein-Barr Virus (EBV) infection and the development of MS being the subject of extensive research recently. When the human body experiences a decline in immune function, it may trigger reactivation of EBV, and this reactivation is also believed to increase the risk of onset or relapse of MS. Currently, the phenomenon of cross-reactivity resulting from molecular mimicry following EBV infection (including reactivation status) is theorized as a contributing etiology. EBV-mediated abnormalities in T cells and B cells also play a key role in the development of MS. However, the underlying mechanisms have not been thoroughly understood. Meanwhile, the limited availability of effective treatment options for MS, in particular MS progression, underscores the urgent need for novel therapeutic strategies. Here, we discuss the pathophysiological mechanisms underlying MS, specifically emphasizing the relationship between EBV infection and the disease pathology. Furthermore, we introduced relevant pharmacological targets in order to propose a broader range of therapeutic alternatives for individuals diagnosed with MS.

ABSTRACT Richter transformation of Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) into classic Hodgkin lymphoma (CHL-RT) is rare and remains incompletely understood. Two histologic subtypes are recognized: type 1 (CLL/SLL with scattered Hodgkin/Reed-Sternberg (HRS) cells) and type 2 (HRS cells within a polymorphous inflammatory background). In this multi-institutional study of 77 patients with CHL-RT (27 type 1 and 50 type 2), we characterized immune evasion markers, PD-L1 / PD-L2 copy number alterations, tumor microenvironment, and performed targeted next-generation sequencing on 37 CLL/SLL samples. HRS cells in CHL-RT displayed immune-evasion phenotypes similar to de novo CHL, though PD-L1 expression was lower in type 1 cases. PD-L1/PD-L2 gain/polysomy were frequent (83.3%). CLL/SLL with CHL-RT harbored increased mutations in XPO1, FBXW7, BIRC3, TRAF3 , and HLA-A versus reference CLL/SLL. Similar mutational profiles, demographics, and survival outcomes support a biological continuum between type 1 and type 2 CHL-RT, with distinct genetic features in CLL/SLL predisposing to CHL transformation. STATEMENT OF SIGNIFICANCE Classic Hodgkin lymphoma-type Richter transformation arises from convergent Hodgkin-like immune evasion on a genetically primed CLL/SLL background. Through integrated genomic and tumor microenvironment analyses in a large multicenter cohort, we establish this as a biologically coherent entity and provide rationale for PD-L1/PD-L2-directed therapeutic strategies.

ABSTRACT As emergency department (ED) crowding and patient boarding reach critical levels nationwide, the role of the emergency medicine pharmacist (EMP) must be both defined and safeguarded. While EMPs are increasingly valued for their contributions to acute care, their responsibilities often unintentionally expand during periods of operational strain, particularly when admitted patients remain in the ED for prolonged durations. EMPs should ideally be resourced to remain focused on time‐sensitive, high‐risk medication management where their expertise is most impactful and not be used as default providers for all pharmacotherapy needs during ED boarding. This perspective outlines our position on the ideal scope of EMP practice during times of congestion, advocating for clearly defined role boundaries, strategic collaboration with other pharmacy team members, and system‐level support to prevent dilution of emergency care. EMPs must be empowered to prioritize critical ED workflows, support resuscitations, and lead initiatives in education, protocol development, and patient safety, without being overextended by responsibilities that can be absorbed by other team members.

Abstract TRIM29 (Tripartite Motif-Containing Protein 29), also known as ATDC (Ataxia-Telangiectasia Group D-Complementing Protein), is an E3 ubiquitin ligase frequently upregulated in muscle-invasive bladder cancer (MIBC). We have previously shown that TRIM29 overexpression in a genetically engineered mouse model (GEMM) is sufficient to drive development of MIBC. We have also shown that TRIM29 is enriched in basal bladder cancer subtypes and is part of a TP63-regulated program that promotes migration and invasive progression. Prior studies suggest that Keratin 5/14 (K5+/K14+) expressing basal-stem progenitor cells in the urothelium are the cells of origin for MIBC in mice. However, whether Trim29 expression in these basal urothelial cells is required for MIBC development has not been established. In this study, we find that Trim29 expression in K5+ basal cells is linked to the inflammatory response and is critical for bladder tumor formation and invasive progression. We show that urothelial TRIM29 expression is enhanced during inflammation in humans and under inflammatory and genotoxic stress in mice. Knockout (KO) of Trim29 in the K5+ basal urothelium of a GEMM exposed to a bladder-specific chemical carcinogen blocked MIBC development. Further, urothelial Trim29 KO enriches immune cell recruitment to the bladder and upregulates STING and inflammatory signaling in mouse and human bladder cancers. These data reveal a critical role for TRIM29 in immunomodulation of the bladder tumor microenvironment (TME), which may act in concert with pro-migratory basal gene programs to expedite MIBC development.

Background The clinical heterogeneity of systemic lupus erythematosus exceeds the resolution of conventional disease activity instruments. Artificial intelligence offers the analytical infrastructure to engage with this complexity, yet the translation of AI models into clinical practice remains limited. Methods This review critically appraises the current evidence for AI applications across the SLE clinical workflow, including computational phenotyping, diagnosis, disease activity monitoring, organ-specific predictive modelling, and treatment personalization. Studies were evaluated for external validation, prospective testing, algorithmic fairness, explainability, and regulatory status. Findings AI applications across the SLE clinical workflow show uneven methodological maturity. Diagnostic and monitoring tools include the SLERPI index, with multinational external validation, and an LSTM flare-prediction model (C-index 0.897). Treatment personalization is anchored by serum IgA2 anti-dsDNA, the only SLE response biomarker validated across independent trials, and by multi-stain deep learning on renal biopsies (AUC 0.84, three external centers). Predictive modelling for organ-specific manifestations has progressed unevenly: cardiovascular risk stratification (SLECRISK) and a multicenter thrombocytopenia model are the most mature, while neuropsychiatric, gastrointestinal, and ocular applications remain single-center proofs of concept. External validation is the exception across SLE prediction models, no AI tool has received regulatory clearance, and populations most affected by SLE remain underrepresented in training data. Conclusions AI demonstrates genuine analytical capability in SLE but the translation gap is defined by insufficient validation, limited explainability, and absent equity evidence. Closing it will require fewer published models and more validated tools, optimization for clinical impact rather than discrimination alone, and demonstration of performance equity across the demographic spectrum of the disease.