United States

Background
Corpus callosum infarction is a rare presentation of ischemic stroke, typically manifesting with atypical symptoms such as ideomotor apraxia or interhemispheric disconnection.
Moyamoya disease, a progressive cerebrovascular occlusive disorder, most often results in ischemic events in the MCA and ACA territories.
Roger’s Syndrome, or Thiamine-Responsive Megaloblastic Anemia (TRMA), is a rare autosomal recessive condition characterized by early-onset diabetes, sensorineural deafness, and megaloblastic anemia.
To our knowledge, this is the first reported case of a patient with both Moyamoya disease and Roger’s Syndrome, highlighting potential synergistic risk factors for stroke.

Case Presentation
A 22-year-old female with a history of TRMA, Moyamoya disease, type 1 diabetes mellitus, a left cochlear implant, and multiple prior strokes presented with left leg weakness, aphasia, altered mental status, and new-onset urinary incontinence.

The patient had a recent insulin dose reduction and was found to be in diabetic ketoacidosis (glucose 419 mg/dL, bicarbonate 9.6 mmol/L).
CT of the brain demonstrated hypoattenuation in the genu of the corpus callosum, indicating an acute/subacute infarct, along with chronic bilateral frontal infarcts. MRI was limited due to cochlear implant artifact.

Cerebral angiography revealed progression of Moyamoya disease with occlusion of the ICA, MCA, and A1 segment, and extensive collateralization.
She underwent a combined revascularization procedure with a right STA-MCA bypass and bilateral indirect onlay craniotomies.
Postoperative imaging confirmed evolving infarcts in MCA and ACA territories. Rehabilitation was initiated with moderate functional gains; the patient was ambulating 50 feet with moderate assistance by discharge.

Discussion
This case illustrates an unusual presentation of Moyamoya disease involving the corpus callosum, resulting in a form of Callosal Disconnection Syndrome.
While stroke is a known complication of Moyamoya, infarction of the corpus callosum remains uncommon.
The co-occurrence of TRMA may represent a novel risk enhancer, potentially through mechanisms such as chronic metabolic stress, thiamine deficiency–related microangiopathy, and diabetic endothelial dysfunction.
Diabetic ketoacidosis likely exacerbated cerebral ischemia by further compromising vascular integrity. This case underscores the importance of multidisciplinary management and vigilance for cerebrovascular events in patients with overlapping rare genetic and metabolic conditions.

AMA Research Challenge 2025 

Roger’s Syndrome and Callosal Disconnection Syndrome in a Stroke Patient: A Case Report

Welcome to the **largest national, multi-specialty research** event featuring research from more than 750 medical students, residents, fellows, and international medical graduates.

Use the buttons below or navigation menu to access the event.
<br><br><br>

[![](https://assets.underline.io/markdown_image/1/image/20e1e77aa954efeac7a15d84c85eb383.jpg)](https://underline.io/events/468/sessions?searchGroup=all)
<br>
[![](https://assets.underline.io/markdown_image/1/image/068522d81aca0b68a369787c7dfccd1b.jpg)](https://underline.io/events/468/posters?searchGroup=all)
<br>
[![](https://assets.underline.io/markdown_image/1/image/d3971ece79d5e5575b178bdc2c8bbfae.jpg)](https://underline.io/events/468/help-desk)
<br><br>

#### Event Details

The 2024 American Medical Association Research Challenge includes a poster symposium hall and a semifinals hall with the top scored research. **Log in at any time throughout the event to:**

* **View posters and presentations** on a variety of topics and specialties
* **Score posters in the semifinals** to help decide the top five finalists (AMA member exclusive opportunity)
* **Connect with participants** and provide feedback to help advance their research

At the conclusion of this event, the AMA will announce the top five finalists who will advance to the AMA Research Challenge finals, present their research to an elite panel of judges, and compete to win a $10,000 grand prize presented by Laurel Road.<br><br>

### Grand Prize Sponsor

AMA would like to thank Laurel Road for sponsoring a $10,000 grand prize for the winner of the AMA Research Challenge. Laurel Road is not involved in winner selection process. The winner will be announced on February 20, 2025 during the AMA Research Challenge finals. Payment of the grand prize will be directly from Laurel Road. See the AMA Research Challenge Official Rules for terms and conditions.

[![](https://assets.underline.io/uploads/markdown_image/1/image/611e331d4a0eba0920f240cfc04b4d77.png)](https://www.laurelroad.com/)

<br><br>

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

**There is no registration fee for this event! **

If you have used the AMA Research Challenge platform previously, please click the ‘Log-in’ button below. If you do not remember your password, please click “forgot password” and follow prompts to reset your password. After resetting your password, please refresh your webpage to access the platform.

Register to explore research in a variety of topics. AMA members get exclusive access to score posters in the semifinals. 

**Please use this link [here](https://www.ama-assn.org/member-benefits/events/ama-research-challenge) to register for free.**

Registration Is Required

AMA Research Challenge 2024 

Join the AMA Research Challenge poster symposium and semifinals to explore top research posters and score posters in the semifinals (an AMA member exclusive opportunity).

poster

Congratulations to all the researchers who have qualified for the AMA Research Challenge virtual poster symposium and semifinals!

Welcome to the **largest national, multi-specialty research event** featuring research from nearly 1,000 medical students, residents, fellows, and international medical graduates across six topic areas - basic science, clinical and translational research, clinical vignettes, health systems science, medical education, and public health and health policy.

#### Access the Event
Use the buttons below or navigation menu to access the event. 

[![](https://assets.underline.io/markdown_image/1/image/d4076f2146939317a29eb9f69c53ad41.png)](https://underline.io/events/506/sessions?searchGroup=all)
[![](https://assets.underline.io/markdown_image/1/image/6bcee0b523a4874c4292ca0415d88dde.png)](https://underline.io/events/506/posters?searchGroup=all)

#### Event Details
The 2025 AMA Research Challenge includes the following symposium halls:
* **Semifinals hall:** Features research from 54 medical students, residents and international medical graduates that received top scores during first round scoring. 
* **Poster symposium hall:** Features research from nearly 1,000 medical students, residents and international medical graduates categorized by topics and subtopics.

**Here’s how to participate:**
* **View posters and presentations** on a variety of topics and specialties
* **Score posters in the semifinals** to help decide the top five finalists (AMA member exclusive opportunity)
* **Connect with presenters** and provide feedback to help advance their research
* **New this year: Watch live poster presentations** by topic area and engage directly with researchers

At the conclusion of this event, the AMA announced the top five finalists who advanced to the AMA Research Challenge final round, presented their research to an elite panel of judges, and competed to win a $10,000 grand prize presented by Laurel Road.

**Addison Shenk, a fourth-year medical student from Edward Via Virginia College of Osteopathic Medicine, was the winner of the 2025 AMA Research Challenge and was awarded the grand prize of $10,000, presented by Laurel Road.**

#### A message from Dr. Bobby Mukkamala, AMA President
<div style="position: relative; width: 100%; padding-bottom: 56.25%; height: 0; overflow: hidden;">
  <iframe 
    src="https://underline.io/embed/136894-ama-video?t=0"
    title="AMA Welcome Video"
    allowfullscreen
    style="position: absolute; top: 0; left: 0; width: 100%; height: 100%; border: 0;">
  </iframe>
</div>


#### Recognized Institutions
In 2025, 1,400 abstracts were submitted across six topic areas representing 174 medical schools and more than 150 residency and health care institutions. [See if your institution is represented this year!](https://drive.google.com/file/d/1eFrTqCmCeobx9pr5drJK2hllvsIpFNOY/view?usp=drive_link)

#### Grand Prize Sponsor
AMA would like to thank Laurel Road, a brand of KeyBank, for sponsoring a $10,000 grand prize for the winner of the AMA Research Challenge. Laurel Road and KeyBank are not involved in winner selection process. The winner will be announced during the AMA Research Challenge final round. Payment of the grand prize will be directly from Laurel Road/KeyBank. See the [AMA Research Challenge Official Rules for terms and conditions](https://www.ama-assn.org/system/files/research-challenge-official-rules.pdf).

[![](https://assets.underline.io/uploads/markdown_image/1/image/611e331d4a0eba0920f240cfc04b4d77.png)](https://www.laurelroad.com/)

#### Code of Conduct
<html>
AMA has a code of conduct for its meetings – we ask all attendees to adhere to this code to participate in this year’s meeting. Please read and <a href="https://www.ama-assn.org/about/code-conduct" target="_blank">learn more here</a> about your conduct requirements.
<br><br>
	</html>
	

#### Semifinals Judges

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Need help? Contact us at <support@underline.io>.

If you have used the AMA Research Challenge platform previously, please click the ‘Log-in’ button below. If you do not remember your password, please click “forgot password” and follow prompts to reset your password. After resetting your password, please refresh your webpage to access the platform. 

Register to explore research in a variety of topics. AMA members get exclusive access to score posters in the semifinals. 

**Please use [this link here](https://www.ama-assn.org/member-benefits/events/ama-research-challenge) to register for free. **

Abstract Title
Topological Characterization of the Glioblastoma Microenvironment Reveals Perivascular Niche Interactions

Background
Differential response to treatment in many tumors can be attributed to intratumoral heterogeneity. This describes diversity of features on the molecular, cellular, and tissue levels that contribute to driving tumor dynamics and fueling disease. In glioblastoma (GBM), a deadly and aggressive brain tumor, extensive cellular and genetic intratumoral heterogeneity play a major role in the inevitable resistance to treatment. Routine histopathological analysis of GBM reveals an extremely diverse cancer ecosystem inclusive of marked nuclear atypia, microvascular proliferation, and foci of necrosis while large-scale genomic analyses have demonstrated an accumulation of genetic changes within cancer cells. However, interactions between genetically distinct GBM cells and the surrounding tumor microenvironment (TME) are incompletely understood. Considering the dynamic GBM microenvironment and heterogeneous clonal populations that arise within, the high degree of both molecular and cellular diversity may be reflecting a level of cooperation. Wherein, genetic heterogeneity confers an evolutionary advantage to tumor cells in the face of fluctuating microenvironmental selective pressures.

Methods
In a cohort of matched primary and recurrent GBM samples, we integrated immunofluorescent staining, fluorescence-in-situ-hybridization techniques, reflectance imaging, species distribution modeling, and computational image analysis methods alongside spatial RNA sequencing to achieve multiplexed single cell spatial profiling. With this method, we have evaluated the TMEs for various histopathological niches, extracellular matrix (ECM) organization, hot-spot mutations, and oncogenic amplifications in the same tissue and at the same histological coordinates. 

Results
Our study found that specific textures of the ECM were linked to specific genetic and phenotypic characteristics of tumor cells. We observed that “string-like” ECM textures, possibly representing neuron-rich environments, serve as selective niches for tumor cells with TERT promoter mutations and EGFR amplifications. Additionally, reflectance imaging of the ECM revealed structured organization of perivascular niches in recurrent GBM which were enriched in collagen and immunosuppressive macrophages. Results from the single-cell spatial transcriptomics further confirmed the presence of the niche-specific macrophage populations and identified interactions between endothelial cells, immunosuppressive macrophages, and perivascular fibroblasts. 

Conclusion
Our findings demonstrate how TME elements may serve as markers indicative of spatial heterogeneity and how combined analysis of spatial diversity can be used to uncover novel niche interactions. Understanding the molecular and cellular dependencies within the GBM microenvironment is crucial for devising targeted therapeutic strategies. In this work we expand the knowledge base of spatially resolved GBM characterization by evaluating the interplay among tumor cell populations, TME members, and ecological features that impact GBM pathophysiology.


Topological Characterization of the Glioblastoma Microenvironment Reveals Perivascular Niche Interactions

Background
Neurons decentralize translation to generate, maintain, and rapidly remodel distal cellular sub-compartments. Neurodevelopment, synaptic plasticity, and axonal repair all depend vitally on this process, with disruptions resulting in neurological disease. Local protein synthesis is a ubiquitous feature of neuronal processes. However, how neurons intrinsically regulate local translation is poorly understood. Given our recent discovery of mRNAs that hitchhike on lysosomes for long-distance transport in neurons, we sought to probe the capacity of lysosomes to support local protein synthesis.

Methods
To study local translation, we used iPSC-derived neurons to recapitulate human neuronal biology. Using a derivative of puromycin, we labeled translation sites in neurites, co-localizing these sites to lysosomes. We developed and used a novel spatially-resolved transcriptomics method and a lysosome-targeted genetically-encoded protein synthesis inhibitor, mapping the landscape of mRNA transcripts and nascent protein made on the surface of the lysosome. We fed isotope-labeled peptide to neurons and traced their incorporation into new proteins, quantifying the contribution of lysosome-derived amino acids to local protein synthesis. We then conducted proteomics of the lysosome-ribosome interface, identifying potential regulatory complexes for lysosome-associated translation. Finally, we used bio-orthogonal nascent protein and lysosome co-detection along with live cell nascent protein imaging to characterize the regulatory dynamics of local protein synthesis at the lysosome surface.

Results
We find that lysosomes are the primary sites of protein synthesis in neuronal processes. Lysosome-associated ribosomes support the translation of a wide variety of neurite-enriched transcripts, including those encoding ribosomal subunits, mitochondrial proteins, and synaptic proteins. Lysosome-derived amino acids are important fuel sources for translating lysosome-localized transcripts, and deacidification of the lysosome impairs their recycling into new proteins at the lysosome surface. The mTORC1 pathway is enriched at the lysosome-ribosome interface, and mTOR inhibition mirrors the translational effects of a lysosome-targeted protein synthesis inhibitor. Finally, lysosomes associated with active translation co-localize with mTORC1 in neuronal processes.

Conclusion
Our data unravel a functional model linking lysosomal signaling and metabolism with RNA co-trafficking, revealing a new layer of spatiotemporal translational control by mTORC1. These findings also implicate deficient local translation as a potential pathomechanism in neurological diseases with prominent lysosomal dysfunction, such as Alzheimer’s, Parkinson’s, FTD, and ALS. More broadly, the spatial coincidence of anabolism and catabolism presented here provides a new platform for exploring diverse modes of subcellular metabolic regulation.


A lysosome-associated ribosome-mTORC1 assembly regulates local protein synthesis in neurons

Abstract Title: Bidirectional Risks and Immune Dysregulation in TB and Type 2 Diabetes

Background:
Pulmonary Tuberculosis (TB) and Type 2 Diabetes Mellitus (T2DM) are major global health burdens that independently impair immune function. When present concurrently, they exacerbate disease severity, reduce treatment effectiveness, and increase mortality. Recent literature suggests a growing bidirectional relationship between these conditions.

Methods:
A literature review was conducted using PubMed, focusing on articles related to cytokine signaling, immune modulation, T2DM, and tuberculosis.

Results: 
T2DM contributes to impaired immunity primarily through hyperglycemia-induced suppression of cytokines such as tumor necrosis factor-alpha (TNF-α) and various interleukins (IL-1β, IL-6, IL-12). This dysregulation compromises macrophage and T-cell responses, increasing the risk of TB activation and reducing treatment efficacy. The rising incidence of both diseases in the United States highlights the urgent need to understand this mechanistic overlap to improve therapeutic strategies.

Conclusion:
T2DM serves as an independent risk factor for TB due to its immunosuppressive metabolic environment. Recognizing these mechanisms can inform clinical decisions and improve management of co-affected patients.


Bidirectional Risks and Immune Dysregulation in TB and Type 2 Diabetes​

TITLE: Histotripsy for the Treatment of Oral Tumors: A Canine Ex-Vivo Feasibility Study

AUTHORS: Dana Wang (2), Victor A. Lopez (1), Jessica Gannon (1), Sheryl Coutermarsh-Ott (3), Jeremy A. Brown (4), Adam Maxwell (1), Joanne Tuohy (5), Eli Vlaisavljevich (1)

(1 )Biomedical Engineering and Mechanics, Virginia Tech 
(2) Virginia Tech Carilion School of Medicine, Virginia Tech 
(3) Biological Sciences and Pathobiology, Virginia-Maryland College of Veterinary Medicine 
(4) School of Biomedical Engineering, Dalhousie University
(5) Department of Small Animal Clinical Sciences, Virginia-Maryland College of Veterinary Medicine 

BACKGROUND: Oral tumors are frequently diagnosed in both humans and dogs and have potential for malignancy. Current standard of care treatments like surgery, radiation, and chemotherapy are invasive, often leading to facial disfigurements and other adverse effects that decrease quality of life. There is a pressing need for less invasive treatments that preserve oral function and appearance.

Histotripsy is a non-invasive, non-thermal, and non-ionizing image-guided focused ultrasound technique. It utilizes acoustic cavitation to break down tissue into acellular debris by producing high-pressure ultrasound pulses that converge at the transducer focus to generate a cavitation bubble cloud for mechanical tissue ablation. While current clinical histotripsy systems are large, cart-based devices operating at low frequencies for deep tissue treatments like the liver, this study aims to develop miniaturized high-frequency histotripsy transducers for targeting smaller, superficial structures like oral tumors.

METHODS: CT images from ten canines with oral tumors were segmented and analyzed in MATLAB to identify accessible acoustic windows with histotripsy treatment. This guided the design of 2MHz and 6.3MHz single-element transducers which were tested on excised canine oral tumors, producing bubble cloud diameters of approximately 2mm and 0.25 mm, respectively. Excised tumors were degassed, embedded in 1% agarose gel, treated with a pulse repetition frequency of 1kHz and 2,000 pulses per point. Real-time ultrasound imaging was used for treatment monitoring and post-treatment ablation was confirmed histologically with hematoxylin and eosin staining.

RESULTS: Acoustic window analysis visualized accessible ultrasound paths to the tumor without impedance by bony structures such as the skull. The tumor volumes varied in size and ranged from 2.2 cm3 to 223.8 cm3, informing the need for a flexible and versatile device to target tumors in the oral cavity. Both transducers produced well-confined cavitation bubble clouds during treatment. After treatment, hypoechoic regions observed via ultrasound correlated to tissue ablation, which was confirmed histologically with sharply demarcated, decolorized ablation zones. Notably, the 2MHz transducer achieved ablation approximately 7.4 times faster than the 6.3MHz transducer. 

CONCLUSIONS: Acoustic window analysis informed optimal transducer positioning and device design for targeting oral tumors effectively with a miniaturized transducer system. Both 2MHz and 6.3 MHz transducers successfully ablated excised canine oral tumors, demonstrating feasibility. Future work will include analysis of additional canine and human tumors and development of an electronically steerable transducer to reduce treatment time. Overall, this study is the first to demonstrate the feasibility of histotripsy for ablating oral tumors and offers key insights for optimizing transducer design for oral tumor treatment.

Histotripsy for the Treatment of Oral Tumors: A Canine Ex-Vivo Feasibility Study

Background
Raoultella planticola is a gram-negative, encapsulated bacillus, commonly isolated from environmental sources such as soil and water, as well as the human gastrointestinal tract. Formerly classified under Klebsiella, improved microbiologic identification methods—such as MALDI-TOF MS and VITEK 2—have refined its classification and led to increased recognition in clinical settings. Historically, it is rarely associated with clinical infection, but there has been an increase in the number of case reports attributed to R .planticola infections in the past decade. This calls the need for establishing guidelines to manage infections due to this emerging pathogen.

Case Presentation
A 46-year-old woman with a past medical history of anxiety disorder and depression presented with concerns of recurrent urinary symptoms (dysuria and frequency). She was previously treated empirically with Amoxicillin-Clavulanate without improvement. She has a remote history of vulvovaginal candidiasis resolved with a single Fluconazole dose. Repeat urine culture done on this visit grew Raoultella planticola (50,000–100,000 CFU/mL), resistant to ampicillin but sensitive to multiple other antibiotics including Cephalexin with which the patient was treated for 7 days. The patient has no other comorbidities and neither is immunocompromised.

Discussion
R. planticola infection was first reported in 1984. Since then it has been implicated in various infections including urinary tract infections (UTIs), bacteremia, pneumonia, cholangitis, and soft-tissue infections. This pathogen primarily infects patients who are immunocompromised, critically ill, have undergone invasive surgical procedures, or individuals with underlying malignancies. Infections in immunocompetent individuals are exceedingly rare, with fewer than 10 documented cases in the literature. We report an unusual case of R. planticola UTI in an immunocompetent patient which was treated to resolution with Cephalexin 500mg BID for 7 days. The majority of reported R.planticola UTIs occurred in older adults with underlying conditions such as diabetes mellitus, chronic kidney disease, or neurogenic bladder manifesting as asymptomatic bacteriuria to severe urosepsis that have responded well to antibiotic therapy when guided by susceptibility results. Notably, R. planticola is intrinsically resistant to ampicillin but generally remains susceptible to β-lactam/β-lactamase inhibitors, cephalosporins, fluoroquinolones, and aminoglycosides. Rare cases of extended-spectrum beta-lactamase production and carbapenem resistance have been reported, underscoring the need for precise identification and targeted treatment. Given its emerging clinical relevance, particularly in recurrent or treatment-resistant UTIs, recognition of R. planticola as a potential uropathogen is important. This case adds to the limited but growing literature and highlights the importance of accurate microbiological diagnosis in managing atypical urinary pathogens.


Rare case of Raoultella planticola UTI in an immunocompetent patient

Background:
Mycoplasma pneumoniae (MP) is a common respiratory pathogen that may cause extrapulmonary manifestations, including mucocutaneous involvement. Reactive infectious mucocutaneous eruption (RIME) is one such manifestation, with skin involvement occurring in approximately 25% of the pediatric cases. MP has also been associated with immune dysfunction and autoimmune disease development. We present a case of a previously healthy child who developed RIME following MP infection, and was subsequently diagnosed with new-onset type 1 diabetes mellitus (T1D). 

Case Presentation: 
A previously healthy 9-year-old male presented with ten days of decreased oral intake due to painful oral mucosa, along with dysuria, penile blisters, and weight loss. Two days prior to symptoms onset, he developed fever and respiratory symptoms, testing positive for MP by PCR. He was initially treated at a community hospital for eight days with presumptive pneumonia and herpes, but was referred to our tertiary center due to persistent mucocutaneous lesions. 

On arrival, he was afebrile with stable vital signs, but presented marked hyperglycemia (glucose 431 mg/dL). Physical examination revealed oral mucositis with hemorrhagic crusting at the lip commissures, erosions on the glans penis, and erythematous plaques on lower extremities and lateral neck. Nikolsky sign was negative. Laboratory workup revealed elevated HbA1c (7.2%), hyponatremia, hyperkalemia, and glucosuria. 

During prior hospitalization, the patient received methylprednisolone for four days. However, glucosuria was documented before steroid initiation. His family denied diabetes symptoms or abnormal glucose levels, and there was no family history. He was started on azithromycin, mucocutaneous supportive care, and insulin therapy. At three week follow-up, skin lesions had resolved, but glycemic control required ongoing endocrinologic follow-up.

Discussion: 
RIME is a clinical diagnosis that may be underrecognized due to overlap with other mucocutaneous conditions. Diagnostic criteria include evidence of an infectious trigger, mucosal involvement at ≥2 sites, or cutaneous involvement affecting <10% of body surface. Drug-induced etiologies must be excluded. In this case, RIME was diagnosed after referral to a supra-tertiary pediatric center, where dermatologists evaluated the patient.

Although autoimmune diseases associated with MP have been documented within the first two years post-infection, the concurrent presentation of T1D raises the possibility of infection-triggered autoimmune disease. While the elevated HbA1c suggests a more prolonged course, the absence of glucose data limits definitive conclusions. Although the timing of both diagnosis could be coincidental, the possibility of a primary autoimmune process cannot be excluded. Clinicians should remain alert to this rare but notable association.

A Pediatric Case of Mycoplasma Pneumoniae presenting with Reactive Infectious Mucocutaneous Eruption with concurrent Type 1 Diabetes Mellitus

Note: All abstracts must be submitted using this approved template and must not include pictures, graphs, or tables. The abstract must not exceed 400 words, excluding the abstract title, page numbering, or any ancillary entrant information. Citations or references are not a necessity with the abstract. If your abstract is accepted for consideration, you may include citations or references with your poster.

Abstract Title
Prototype for a Predictive Healthcare Inventory Management System

Background
Inefficient medical inventory management significantly burdens healthcare, exacerbating rising hospital expenses, with supplies (22%) and labor (56%) being major drivers. Challenges include high-value items, varied stock, product expiration, manual tracking inefficiencies, and financial implications leading to waste and potential delays in care. Hospitals predominantly use real-time or retrospective systems rather than anticipatory systems. While structured approaches like frequency-based or mathematical modeling within these existing systems can reduce management time, their implementation can be challenging. Existing solutions are not cost-effective, with some systems costing hundreds of thousands of dollars. Other industries leverage machine learning (ML) for large-scale inventory applications, but this approach is largely unapplied in healthcare. We aim to create an ML model for the implementation of a predictive system to streamline the process of inventory management in the hospital setting.

Methods
We created an ML-based forecasting tool to predict monthly and quarterly order quantities for medical supplies using historical data from an academic tertiary care center. Our dataset included information on dates, categories, quantities received, and costs for all item orders from January 2022 to December 2024 for the interventional radiology department. Items ordered fewer than 10 times per year were excluded. For each item, order history and quantities were analyzed, and an ML model was created using Python 3.9 using Random Forest, Gradient Boosting, and XGBoost algorithms, tuning each with GridSearchCV. Models were trained with 10 iterations per item to average predictions and reduce variance. Performance was analyzed with median absolute error (median AE), mean absolute error (MAE), and explained variance score.

Results
Our training data was trained on 26 months of order data, and the testing was done on the subsequent 10 months. Over the 42 items, the Random Forest model showed the best performance with monthly data, resulting in a mean MAE of 5.69 and SD of 4.54. In contrast, Gradient Boosting outperformed other models with quarterly data, resulting in a mean MAE of 6.00 and SD of 4.7.

Conclusion
Our results suggest that an ML model can be utilized in the healthcare inventory setting to optimize item procurement and storage, relieving the burden of inventory from staff and improving efficiency. With our model, we seek to minimize procedure cancellations from a lack of requisite supplies and avoid overstocking. Further iterations will modify model inputs and parameters to improve performance and validate the model using another dataset from other departments and institution-wide orders.


Low-Cost Prototype for a Predictive Healthcare Inventory Management System

Background: 
Effective patient education in ophthalmology is critical for promoting adherence and outcomes, yet existing resources are often limited by cost, readability, and language accessibility. The American Academy of Ophthalmology (AAO) provides expert-developed handouts, but these are paywalled and may exceed recommended reading levels. ChatGPT-4 offers potential to generate low-cost, accessible materials that are easier to scale. Prior work has evaluated AI-generated medical responses, but its ability to create comprehensive, multilingual ophthalmic handouts comparable to trusted educational materials remains underexplored. This study compares the readability, content quality, and clinician preference of ChatGPT-4–generated handouts with AAO materials in English and Spanish.

Methods:
Ten AAO handouts on common ocular conditions were obtained in English and Spanish. ChatGPT 4.0 was prompted to generate matched handouts at an 8th-grade reading level and similar word count. 

Readability was assessed with Simple Measure of Gobbledygook (SMOG) and Flesch Kincaid Reading Ease (FKRE) in English; and Fernández Huerta Index (FHI) and Szigriszt-Pazos Perspicuity Index (SPPI) in Spanish. 
Three ophthalmologists per language independently evaluated each handout using the validated Quality of Generated Language Outputs for Patients (QGLOP) rubric, which scores accuracy, comprehensiveness, bias, currency, and tone on a four-point scale. Graders were blinded to authorship and asked to indicate their preferred version and guess the source.

Results:
In English, ChatGPT and AAO handouts had comparable readability (SMOG 7.86 vs. 7.49, p=0.35; FKRE 62.37 vs. 65.70, p=0.24) and content quality (QGLOP 14.73 vs. 14.87, p=0.97).

In Spanish, ChatGPT handouts were slightly less readable (SPPI 60.26 vs. 64.82, p=0.04), but still fell within accepted ranges and significantly outperformed AAO materials in QGLOP scores across all domains (15.83 vs. 14.23, p<0.001).

Clinicians preferred ChatGPT handouts in 76.7% of English and 96.7% of Spanish comparisons. ChatGPT authorship was correctly identified in 56.7% of English and 10% of Spanish cases.

Conclusion:
ChatGPT-4 can generate ophthalmic handouts with content quality comparable to AAO materials in English and superior in Spanish, despite slightly lower Spanish readability within acceptable ranges. Strong clinician preference suggests potential to expand multilingual patient education access. However, the discrepancy between preference and source recognition indicates possible clinician bias toward AI-generated content, underscoring the need to better understand how such perceptions influence clinical adoption. Future work will also explore patient perspectives to ensure that materials are not only accurate but also accepted by the intended audience.

Utility of ChatGPT 4.0 in Creating Patient Handouts in Ophthalmology: A Comparison with the American Academy of Ophthalmology Educational Materials

Background: University students often encounter significant health challenges while simultaneously managing their health independently for the first time. The aim of this study was to evaluate the perceived impact of community-based educational events on participants’ health perceptions across five health domains.
Methods. This was a cross-sectional study conducted at the annual Women’s Health Day (WHD) annual event held on the Florida State University campus. Self-reported perceptions about the impact of the event and future health behaviors were assessed with the post-event survey using the Likert scale model.
Results. Of the 327 respondents (mean age 20.14 years, 82.0% women), out of 512 event participants, 92.4% were first-time attendees at WHD. Participants strongly agreed with the event's educational impact on their future health (M = 1.49, SD = 0.79), the importance of interactive and community-based events on campuses (M = 1.29, SD = 0.69), and their likelihood of attending similar events in the future (M = 1.25, SD = 0.53). Sexual health was reported to have the greatest perceived impact, followed by mental and physical health, with significant differences across domains (p < .001). Gender identity influenced perceived impact, with cisgender participants reporting greater impact in mental, clinical, and physical health compared to gender minority participants (p < .05).
Conclusions. WHD demonstrated significant benefits in promoting health literacy and preventative care resources, particularly in sexual and mental health. Future studies should explore the longitudinal impacts, enhance engagement across all health domains, increase involvement among underrepresented groups, and tailor educational content to diverse health needs.

Impact of a Community-Based Health Education Event on Participant Perceptions and Behaviors

Background
The impacts of climate change are experienced in the community and in clinical practice. This study combines interdisciplinary climate change research with advocacy strategies to address the evolving challenges imposed by climate change upon patients and practitioners. A systematic review of the neurological literature related to climate change was conducted and findings were compiled into examples of patient health impacts. In addition, examples are provided of ways practitioners can adapt their practices to address the impacts of climate change on patients and to reduce environmental harm in the clinic and operating room.

Methods 
Literature pertaining to the neurological effects of climate change on patients and practices was reviewed. Following PRISMA guidelines, search parameters for the systematic review were identified and relevant studies were sorted for inclusion based on selection criteria. Studies sought discussed neurological disease variance due to climate change that may require neurological consultation, or directly identified information relevant to sustainable practices. Two tables were constructed presenting (1) specific patient health implications of climate change on neurological illness, and (2) strategies for clinical practices to prepare for and address these emerging sequelae.

Results
A total of 421 studies were initially obtained via the search criteria and 84 were removed as duplicates. After abstract screening of 337 studies, the full text of 95 studies were reviewed. 70 were relevant to neurological disease impacted by climate change, and 20 studies were deemed relevant to sustainable practices. The neurological impacts of climate change on patients and practices include cerebrovascular disease (increased stroke, aneurysmal subarachnoid hemorrhage, and heatstroke), neurodegeneration (increased hospitalizations, symptom exacerbation, and mortality), CNS infectious diseases (viral, bacterial, fungal, parasitic, prion, and amoeba), trauma (extreme weather events and violence), and environmental exposures (birth defects, contamination, brain tumorigenesis, and seizure trigger).

Conclusion
Through sustainable clinical practices and education/training the healthcare field can prepare patients and practitioners for these impacts and rise to address the challenges facing our communities in the 21st century.

Premium content

Next from AMA Research Challenge 2025

Topological Characterization of the Glioblastoma Microenvironment Reveals Perivascular Niche Interactions

Similar lecture

A Rare Anterior Choroidal Artery Pseudoaneurysm Secondary to Moyamoya Syndrome in a Patient with Sickle Cell Anemia