Hermann J. Muller's research swan song: His long-term radiation dose-rate mutation investigations yielded no research publications, leaving questions of scientific misconduct.

The Certified in Disease Intervention (CDI) certification aims to reinforce and expand the expertise of those working at the community level to prevent the spread of infectious disease. To support the recertification process and promote continuing education of CDI competencies, a resource repository of relevant, high-quality materials is being developed. Interdisciplinary subject matter experts at Indiana University's School of Public Health-Bloomington, a Council on Education for Public Health-accredited and Association of Schools and Programs of Public Health member school, developed a systematic approach to identify and vet training resources. The result was an evidence-informed rubric for assessing multiple attributes of online training materials (including Course Overview and Introduction, Learning Objectives, Learning Assessment and Activities, Instructional Materials, Social Awareness, Scenario-Based Learning, Timeliness of Content, Course Technology, Learner Support, and Accessibility and Usability), a list of resources with evaluation outcomes, and metadata tags for populating a searchable database. The team successfully curated a robust repository of high-quality educational resources. However, the list is not exhaustive, and future work is needed to capture and evaluate more resources and update when new trainings are released for disease intervention professionals in the United States.

The hidden isotope fractionation: How temperature of pyrolysis coupled to compound-specific isotope analysis shapes δ¹³C results.

Primate exposure to anthropogenic pollutants: A role for comparative aging research Daniella E. Chusyd of Indiana University discusses how variation in rates of aging across species can help us understand the consequences of our chemical exposome. Aging research seeks to understand why health, physical function, and cognitive abilities change over time. While aging is often thought of simply as ‘getting older,’ biologically it reflects a gradual accumulation of damage and dysregulation across multiple systems. In 2013, researchers proposed the ‘hallmarks of aging’ framework, which identified core biological processes that contribute to age-related decline. (1) These include genomic instability (i.e., DNA damage), telomere (i.e., protective caps on chromosomes) shortening, epigenetic alterations (i.e., changes to how genes are regulated), impaired protein maintenance, deregulated nutrient sensing, disrupted energy production in mitochondria, accumulation of senescent cells, stem cell exhaustion, and altered communication between cells. More recent updates have incorporated impaired cellular recycling (i.e., autophagy), chronic inflammation, and imbalances in the microbiome as additional contributors. (2)

Whether performing an away rotation increases fellowship match success is important for both cost and logistic considerations. This study aimed to determine whether away rotations increase the likelihood of urogynecology and pelvic reconstructive surgery (URPS) match success. This was a retrospective analysis of 246 applications to Indiana University URPS fellowship between 2015 and 2019. Variables included match success, whether an away rotation was performed, demographic data, medical school and residency descriptors, examination scores, prior research, personal statements and letters of recommendation. Data were summarized for the overall cohort and by URPS match (yes/no). The t test was used to compare continuous variables with normal distributions, the Wilcoxon rank sum test for nonnormal distributions, and the χ2 test for categorical variables. Univariate logistic regression was performed to test the association of each variable collected with the URPS match. Backward selection was used to determine variables to include in multivariable logistic regression and identify independent predictors of URPS match success. Of the 245 applicants for whom match data were available, 80 performed a URPS away rotation and 171 (69.8%) successfully matched into a URPS fellowship. Approximately 34% (33.9%) who matched performed a URPS away rotation versus 30.1% (P value 0.57) who did not. Successful match applicants were younger (adjusted OR, 0.82; 95% CI, 0.73, 0.92), had presented at a URPS-related scientific meeting (adjusted OR, 2.53; 95% CI, 1.25-5.13), and had higher personal statement scores (adjusted OR, 1.11; 95% CI, 1.01-1.21). URPS away rotations do not increase the likelihood of match success. Conducting URPS research is likely a more productive focus.

What do you do, and how long have you been doing it? I research how technology fits or doesn't into the lives of neurodivergent people, particularly those with ADHD and autism. I'm currently a PhD student at Michigan State University, where I focus on making technologies more accessible and socially aligned. I came to this work through a master's in HCI at Indiana University and a stint as a product designer at Osmo. Across research, design, and academia, I've been working in and around HCI for about 7 years.

Clinical stage IIA/B embryonal carcinoma predominant nonseminomatous germ cell tumor: Minimizing treatment burden.

Karen Underhill. Bruno Schulz and Galician Jewish Modernity. Bloomington, IN: Indiana University Press, 2024. Pp. 278.

Wolfram syndrome type 1 (WS1) is a rare autosomal recessive disorder classically associated with diabetes mellitus (DM) and optic atrophy (OA). We aimed to characterize OA in WS1 and evaluate optical coherence tomography (OCT) and genetic biomarkers as tools for disease monitoring and prognostication. We conducted a retrospective chart review of genetically confirmed patients with WS1 seen at Washington University or Indiana University neuro-ophthalmology clinics between July 2017 and 2024. Data included demographics, clinical history, best corrected visual acuity (BCVA), OCT retinal nerve fiber layer (RNFL) and ganglion cell complex (GCC) thickness, and WFS1 mutation severity scores. Linear regression analyses assessed correlations between BCVA and clinical, structural, and genetic variables. Thirty-six patients (22 women, 14 men; median age 20 years) were identified. Median mutation severity score was 3.5. Vision loss occurred in 31 patients; in 3 patients it was the only major symptom, in 5 patients it preceded DM, and in 6 patients it occurred without DM. Mean and median BCVA were 20/125 and 20/80, respectively. BCVA correlated inversely with RNFL thickness (P = 0.0017, R2 = 0.14), GCC thickness (P = 0.0018, R2 = 0.29), and mutation severity score (P = 0.031, R2 = 0.14). OA was the most common and sometimes earliest WS1 manifestation. Correlations between BCVA, OCT metrics, and mutation severity score support their potential value as biomarkers and prognostic tools. Findings also support considering genetic screening for WFS1 mutations in patients presenting with otherwise unexplained OA.

Acute kidney injury in sepsis evolves over hours to days, yet most ICU models emphasize onset and provide limited insight into cardio-renal interactions. We developed AKI-twinX, an organ-structured, explainable digital twin that jointly forecasts acute kidney injury onset, acute kidney injury trajectory, and near-term mortality risk. The model learns renal and cardiovascular latent states with sparse feature gating and captures cross-organ coupling with attention. We trained AKI-twinX on MIMIC-IV sepsis using 5-fold cross-validation and evaluated it on an Indiana University Health cohort. Discrimination was consistent across systems (AUC: mortality 0.86-0.88, acute kidney injury onset 0.78-0.82, acute kidney injury trajectory 0.73-0.78). In vasopressor-treated windows, 12-hour systolic blood pressure forecasts tracked observed values (mean absolute error 8.5 mmHg). Counterfactual vasopressor withdrawal shifted predicted blood pressure downward and increased predicted risk, supporting sensitivity to clinically meaningful interventions. AKI-twinX enables trajectory-aware forecasting with bedside auditability in sepsis.

IntroductionPazopanib is a treatment option for patients who have advanced or metastatic soft tissue sarcoma. Although the FDA label recommends pazopanib 800 mg q24h as the initial dose, the starting dose in clinical practice can range from 200 mg to 800 mg q24h due to poor tolerability. This study describes a single institution's experience and outcomes of using pazopanib for patients with sarcoma.MethodsA retrospective chart review was conducted for adult patients with recurrent, advanced, or metastatic sarcoma who were prescribed pazopanib between 2014 and 2022 at Indiana University Simon Comprehensive Cancer Center.ResultsOf 148, 66 patients met the inclusion criteria. Median starting dose and frequency of pazopanib was 300 mg q24h (range, 200 mg q48h-800 mg q24h) with 200 mg q24h (48%) being the most common. Median time on pazopanib was 5.5 months (range, 0.62-100 months). Median progression free survival was 4.4 months (95% CI, 3.3-9.4), and median overall survival was 29 months (95% CI, 21.8-NE). The most common toxicities were fatigue (51%), hypertension (37%), nausea/vomiting (34%), and diarrhea (33%). Treatment interruptions occurred in 45% of patients, and dose reductions in 30% of patients.ConclusionsLower starting dose of pazopanib did appear to reduce the risk of various side effects compared to the PALETTE trial, suggesting a potential benefit of this dosing strategy given similar clinical efficacy. Further comparative studies examining lower initial dose vs full-dose initiation are needed to determine differences in patient tolerability and clinical outcomes.

To assess the effectiveness of virtual reality relaxation (VRR) as an intervention for reducing anxiety in adolescents and adults during the orthodontic bonding procedure. This prospective clinical study included 53 patients undergoing fixed orthodontic treatment at the Indiana University School of Dentistry. Participants were randomly assigned to one of two groups: the experimental group using VRR during orthodontic direct bonding or the control group receiving no distraction. Physiological measures (blood pressure [BP], heart rate [HR], and oxygen saturation) and psychological assessments (5-item State-Trait Anxiety Inventory-Short Form, and the Visual Analog Scale for Anxiety [VAS-A]) were collected at baseline (T0) and 30 minutes into the bonding procedure after bonding of one arch (T1). Repeated measures of analysis of variance were used to compare groups and timepoints. At T0, no statistically significant differences were found between the control and VRR groups across any psychological or physiological measures (P > .05). Although both groups showed significant reductions for physiological and psychological measures from T0 to T1 (P < .01), the VRR group reported significantly lower VAS-A scores, HR, and diastolic BP at T1 than the control group (P < .05). VRR effectively reduced psychological and physiological anxiety during orthodontic bonding procedures, supporting VRR as a promising adjunctive tool in managing dental anxiety during orthodontic care.

The purpose of this study is to evaluate whether a single, fine-tuned SLM can match or exceed the performance of LLMs across diverse clinical tasks, enabling hospitals to build tailored, privacy-preserving, efficient, and deployable language models that do not require managing multiple task-specific systems. We used SLMs of varying sizes and applied low-rank adaptation (LoRA) for fine-tuning across three clinical tasks: (1) medical report labeling, (2) DICOM series description harmonization, and (3) impression generation from findings. These tasks were constructed using two datasets: the public Open-i Indiana University Chest X-ray Dataset and an in-house brain MRI DICOM metadata dataset. We compared single-task SLMs, a multi-task SLM (representing our proposed configuration), and GPT-4o using zero-shot and few-shot prompting. We found OPT-350m to be the optimal SLM. In medical report labeling, the multi-task SLM achieved an F1 score of 0.894 compared to additional prompt-engineered GPT-4o's 0.728. In DICOM series description harmonization, the multi-task achieved an accuracy of 0.975 compared to additional prompt-engineered GPT-4o's 0.878. In impression generation from findings, the multi-task SLM achieved an average Likert scale score of 4.39 ± 1.00, compared to GPT-4o's 3.65 ± 1.00 (p = 0.0008). This study demonstrates that a single fine-tuned SLM can serve as a general-purpose clinical assistant, offering performance on par with or better than larger models. With lower resource requirements, greater customizability, privacy protection, and strong task generalization, fine-tuning one SLM to support multiple clinical tasks meets the practical demands of clinical AI deployment in both high-resource and resource-limited healthcare settings.

In the United States, Black adults experience disproportionately high rates of hypertension and are nearly twice as likely to die due to a cardiovascular disease–related complication as white adults. These disparities are related to systemic barriers — including food, housing, and economic insecurity — that impede access to consistent, high-quality care. To address these structural inequities, the Indianapolis Health Equity, Access, Outreach, and Treatment Collaborative launched the Convenient Home Evaluation for Cardiovascular Health and Individual Tracking program in June 2023. It integrated home blood pressure monitoring, patient education, community health worker support, and virtual medication management to improve hypertension control among residents of high-risk Indianapolis neighborhoods. Adapted from the American Heart Association’s “Check. Change. Control.” model, the 4-month program was designed to help patients build the knowledge, skills, and support needed to manage their cardiovascular health. The program operated across six Indiana University (IU) Health primary care facilities strategically positioned near anchor communities selected based on high rates of cardiovascular disease, social vulnerability, and racial disparity. Participants received a free blood pressure cuff and training on accurate home measurement, along with automated text reminders and a 4-month educational curriculum covering hypertension basics, including nutrition, stress reduction, sleep, and physical activity. Through outreach and home visits, community health workers provided social support, assessed social determinants of health needs, and connected participants with additional resources. Community health workers escalated urgent cases to IU Health’s social work team for rapid intervention, while IU Health pharmacists offered virtual support for participants with uncontrolled blood pressure. During the program’s first 18 months, more than 60% of participants who completed the program achieved blood pressure control (blood pressure <140/90 mmHg), with average reductions of 7 mmHg systolic and 4 mmHg diastolic. Data on maintenance of blood pressure control were not collected beyond the 4-month period. In addition, data were not collected on how many referrals to community resources to address social needs resulted in services provided, though plans are under way to address that gap. Nevertheless, the authors’ findings highlight the potential of the Convenient Home Evaluation for Cardiovascular Health and Individual Tracking (CHECK-IT) program as a scalable, community-centered model addressing both medical and social drivers of hypertension disparities.

Diagnosing diseases from medical images and reporting them at the paragraph level is a significant challenge for deep learning-based autonomous systems. Existing work primarily focuses on achieving high accuracy, often paying less attention to the computational cost of training and testing. The goal of this work is to build a low computational cost and high-performance hybrid encoder-decoder architecture capable of producing autonomous medical reports. On the encoder side of our architecture, called FAST-MRG, features are extracted from images with a transformer-based encoder enriched with distillation techniques, while on the decoder side, a generative pre-training transformer generates paragraph-level text using the extracted features. Numerical analysis with word matching evaluation metrics, temporal analysis and observational analysis were performed to measure the success of the architecture. Our hybrid encoder-decoder architecture was trained and tested using chest X-ray images and reports from the Indiana University Chest X-ray collection dataset. The FAST-MRG architecture achieved scores of 0.373, 0.226 and 0.332 on the Bleu-1, Meteor and Rouge evaluation metrics, respectively. It also has an average time efficiency of 66% compared to previous work using similar GPU environments. The study demonstrates through experiments that meaningful reports are produced that can support doctors in diagnosis and treatment processes. In the study, the results are presented not only with measurable average values but also with a density distribution graph and the test results are analyzed in depth. With its low runtime and high performance, the proposed architecture can serve as a basis for future work.

615 Background: Surveillance of patients with NSGCT following first-line chemotherapy typically relies on a combination of STM and imaging. While STM are sensitive, a proportion of relapses occur without STM elevation. We aimed to characterize the incidence of progression with normal STM in patients with NSGCT who had initial elevated STM prior to first-line chemotherapy. Methods: The Indiana University (IU) germ cell tumor database was reviewed for patients with metastatic NSGCT who had elevated tumor markers prior to first-line chemotherapy (AFP &gt;25 ng/mL or β-hCG &gt;5 mIU/mL) and subsequently had disease progression. Patients with radiographic relapse in the absence of STM were classified as “normal STM relapse.” Clinicopathologic variables, sites of relapse, and histology at progression were compared between elevated and normal STM groups at progression using chi-square or Fisher’s exact test, as appropriate. Results: A total of 725 patients were eligible. Primary tumor site was testis/retroperitoneal in 88.4% and mediastinal in 11.6%. A total of 82 (11.3%) progressed/relapsed with normal STM. Among the 82 patients who progressed with normal STM, sites of progression were lymph nodes in 60 (73.2%), lungs in 44 (53.7%), brain in 11 (13.4%), liver in 6 (7.3%), and bones in 3 (3.7%). Histologic evaluation at relapse revealed similar rates of viable germ cell tumor between patients with normal STM and elevated STM at progression (20.7% vs. 16.5%, p = 0.349). However, teratoma (40.2% vs. 11.8%, p &lt; 0.001) and malignant transformation of teratoma (14.6% vs. 2.6%, p &lt; 0.001) were significantly more common among patients with normal STM relapse. Conclusions: Among patients with metastatic NSGCT who have elevated STM at diagnosis and progress after first-line chemotherapy, a clinically meaningful subset will have radiographic progression with normal STM distinguished by increased prevalence of teratoma and malignant transformation. These findings underscore the role of routine imaging in addition to STM in post-therapy follow-up of NSGCT. Pathology at Progression STM Normal (n = 82) STM Elevated (n = 643) p-value GCT 17 (27.4%) 106 (53.3%) 0.349 Teratoma 33 (53.2%) 76 (38.2%) &lt;0.001 Malignant Transformation 12 (19.4%) 17 (8.5%) &lt;0.001 No Pathology 20 444

IntroductionEach year, an estimated 1700 children should be diagnosed with cancer in western Kenya, with leukaemia making up nearly one-third of cases. However, far fewer are actually diagnosed, highlighting significant delays or errors in diagnosis. Flow cytometry, which the WHO considers essential for leukaemia diagnosis, remains underused across sub-Saharan Africa due to high costs, outdated equipment and a lack of trained personnel. In Kenya, decades-old cytometers have been adapted for leukaemia detection, but these systems are now outdated. Newer platforms, such as simplified single-tube multiparametric assays, provide a scalable and sustainable alternative. This study presents a protocol to evaluate the accuracy of diagnosis and the potential for implementing a streamlined flow cytometry assay using peripheral blood, supported by a regional educational initiative.Methods and analysisThis prospective, mixed-methods implementation study has three aims: (1) to assess the concordance between the Beckman Coulter ClearLLab 10C gold standard 4-tube assay and the streamlined ClearLLab LS 1-tube assay using paired bone marrow and peripheral blood samples; (2) to evaluate the feasibility of peripheral facility referrals and transport logistics with couriered peripheral blood samples from referring sites across western Kenya; and (3) to measure training effectiveness and knowledge gain through a multimodal educational programme using the Project ECHO (Extension for Community Healthcare Outcomes) model. Up to 300 patients at Moi Teaching and Referral Hospital in Eldoret, Kenya, will be enrolled in Aim 1. A separate sample of 100 patients from peripheral facilities will be included in Aim 2. Surveys, knowledge assessments and structured interviews will be used to evaluate training impact under Aim 3. Diagnostic concordance, sensitivity, specificity and knowledge gain will be measured through appropriate quantitative and qualitative methods.Ethics and disseminationThe protocol has received approval from institutional ethics committees at Moi University, MTRH and Indiana University. De-identified data will be analysed and shared through peer-reviewed publications, stakeholder presentations and educational platforms.

Introduction: Cefiderocol is typically reversed for treatment of multi-drug resistance gram negative pathogens. Package insert dosing of cefiderocol in patients with impaired renal function often wastes part of a vial. Investigating alternative dosing strategies is crucial to optimize the use of cefiderocol, potentially reducing excess costs while maintaining therapeutic efficacy. CREDIBLE-CR trial showed comparable microbiological outcomes but higher mortality in the cefiderocol group compared to the best available therapy group. The infectious diseases pharmacists at Indiana University Health (IUH) developed an alternative dosing based on pharmacodynamic modeling studies. Methods: This was a retrospective cohort study of adult patients admitted from January 1, 2019, to July 31, 2024, with carbapenem-resistant gram-negative infections using an alternative dosing scheme of cefiderocol. The alternative dosing was derived from Monte Carlo simulations and resulted in 1 gram every 6-12 hours per patient’s renal function. The two study groups were patients that would have been ineligible for CREDIBLE-CR (Group A) and those who would have been eligible (Group B). The primary endpoint was a composite of 28-day all-cause mortality or treatment failure. Our objective aimed to describe patient outcomes in adults who received alternative cefiderocol dosing in carbapenem-resistant organisms. Results: Composite of 28-day morality or treatment failure occurred in 10/30 (33.3%) patients in group A and 12/28 (42.9%) patients in group B, (p=0.445). Treatment failure occurred in 7 patients (23.3%) and 6 patients (21.4%) in groups A and B, respectively (p=0862). The primary indication for cefiderocol use in both groups was pneumonia (38%) and blood stream infections (34%) caused by Pseudomonas aeruginosa(52%) followed by Acinetobacter baumannii (14%). Conclusions: This was the first study of an alternative dosing of cefiderocol in a real-world setting. Our proposed dosing may offer potential cost savings for health systems by optimizing the vial utilization compared to package insert dosing.

IntroductionThe over 14 million African children who are HIV-exposed but uninfected (CHEU) are at risk for poor health outcomes, including neurodevelopmental conditions such as autism; however, no study to date has examined autism in CHEU in Africa, where the vast majority of these children live. Scalable diagnostic and neurobehavioural tools, including powerful, low-cost approaches such as eye-tracking, for detection and study of mechanistic neural processes are necessary to advance autism research in these settings. The objective of this study is to examine autism diagnostic outcomes and eye-tracking biomarkers in relation to CHEU while at the same time building capacity for neuro-health research in Kenya.Methods and analysisThis study will leverage a longitudinally assessed cohort of CHEU and children who are HIV-unexposed and uninfected (CHUU) with well characterised HIV-related and contextual exposures. We will first determine and compare autism diagnostic outcomes between young CHEU and CHUU across a large cohort (n=850) of Kenyan children using research-grade autism assessment tools, and, second, determine whether neurobehavioural eye-tracking markers predict autism outcomes across this cohort.Ethics and disseminationHuman subjects approvals have been obtained from Moi University Institutional Review and Ethics Committee (IREC; IREC/909/2024; Approval #0004835), Kenya’s National Commission for Science, Technology and Innovation (NACOSTI; Reference #NACOSTI/P/25/415028), the Institutional Review Board of the Indiana University School of Medicine (Protocol #23171), with reliance agreements executed with Purdue University and Boston University. Dissemination of findings will occur through multiple channels within the research and clinical community, including peer-reviewed journal publications and conference abstracts and presentations. As part of capacity building efforts, the research team will also communicate study results to policy makers, the lay public and other health systems involved in the care of young children with disabilities via study-hosted workshops and conferences.

Introduction: Solar eclipses are often associated with mass gatherings along the path of totality. Understanding the effects of such events on emergency services is crucial for preparedness and resource allocation. This study aims to investigate preparations for and the impact of the April 8, 2024, total solar eclipse in two metropolitan areas: Indianapolis and Bloomington, Indiana, USA. Methods: Organizations involved in emergency management and public safety were sent a questionnaire assessing their preparedness efforts as well as the expected and actual impact that the eclipse had on each organization. Data was collected and managed using REDCap electronic data capture tools hosted at Indiana University School of Medicine. Results: Twenty-seven organizations were sent the questionnaire. Fifteen responses were obtained. Eight self-identified as hospital/healthcare, one as fire department, one as school/university, zero as federal or state government, one as local government, three as emergency medical services, zero as police, and one as other (emergency management/fire department) The median overall anticipated impact before the eclipse was three on a 1-5 Likert scale, with increased traffic being the most anticipated impact The median perceived impact after the eclipse was two (1-5 Likert scale), with increased population, traffic, and potential for mass casualties being the most impacted. Key themes extrapolated from the qualitative data include: communication and coordination between and within organizations is key to effective preparation and response; exercise planning for real-world events has intrinsic value beyond the actual event; overall impacts to hospitals were less than expected; overall preparedness efforts likely decreased any significant impacts. Conclusion: Solar eclipses pose unique problems and require multi-agency coordination to prepare for large influxes of visitors. While impacts on emergency services may not be as profound as initially anticipated, active preparation for such events has intrinsic value and may mitigate negative consequences.