United States

Abstract Title: The Role of the Gut Microbiome in the Pathogenesis of Pre-Clinical Autoimmune Disease and the Prevention of Progression

Background: Autoimmune diseases often develop silently over years, with immune dysregulation and autoantibodies occurring long before symptoms appear. This silent period is what is referred to as “pre-clinical” autoimmunity. Pre-autoimmunity has been observed in several autoimmune diseases, including Rheumatoid Arthritis, Systemic Lupus Erythematosus, Sjögren’s Syndrome, Primary Biliary Cirrhosis, Inflammatory Bowel Disease, and Multiple Sclerosis. The gut microbiome has been increasingly implicated in this pre-clinical phase of autoimmune disease. This review examines how alterations in the gut microbiome can lead to the early stages of autoimmunity before symptoms begin and how the microbiome can be targeted to prevent the progression to autoimmune disease. 

Methods: The author carried out a literature review over PubMed using the keywords Gut microbiome pre-clinical autoimmune disease. 17 articles were reviewed. Peer-reviewed articles, systematic reviews, primary sources, clinical trials, case reports, cohort studies, interventional studies with appropriate control groups were included, while studies not related to pre-clinical autoimmunity, lacking control groups, in-vitro/ex vivo studies, and non-English publications were excluded. 

Results: A key factor in autoimmune disease pathology is gut dysbiosis, or alterations in the gut flora. This can cause increased gut permeability which allows bacterial byproducts like LPS to enter circulation triggering a cytokine cascade, altered gut-derived metabolites like changes in short chain fatty acids, tryptophan and bile acid metabolism leading to a pro-inflammatory state, and molecular mimicry which causes the immune system to attack self-tissues. Strategies to prevent the progression of pre-clinical autoimmunity to symptomatic autoimmunity include dietary intervention, supplementation of prebiotics and probiotics, fecal microbiota transplantation, microRNA therapies, targeted antibiotic therapy, autophagy regulation, and use of natural compounds. 

Conclusions: There is increasing recognition of the gut microbiome’s influence on immune system regulation and emerging evidence linking gut dysbiosis to immune dysregulation. Understanding how changes in the gut microbiome influence the expression of autoantibodies before the development of symptoms paves the way for mechanism-targeted, personalized interventions, and more accurate predictive and diagnostic tools, ultimately aiming to improve patient outcomes and potentially prevent the development of autoimmune disease. 


AMA Research Challenge 2025 

The Role of the Gut Microbiome in the Pathogenesis of Pre-Clinical Autoimmune Disease and the Prevention of Progression

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.

If you see this page it means you are not logged in or registered.

If you feel like you see this message by mistake please make sure you are logged in with the email you registered with.

If you are not registered yet, you can do so [**here**](https://www.sebiology.org/events/seb-conference-prague-2024/registration2024.html). 

Please login or register

SEB Conference 2024

SEB’s Annual Conference is celebrated for creating the perfect composition of research, new discoveries and collaborative connections. Be creative and innovative by exchanging ideas and knowledge with masters of biology from all over the world.

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. **

Registration Is Required

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).

Non-small cell lung cancer (NSCLC) presents a complex tumor microenvironment (TME) that modulates immune cell populations, including a distinct subset of SiglecF high neutrophils associated with tumor progression. While previous studies have identified soluble factors like cytokines as key modulators of immune cell behavior, the specific factors driving SiglecF high neutrophil induction in NSCLC remain unclear. We hypothesized that specific soluble factors within the NSCLC TME, particularly granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and transforming growth factor-beta 1 (TGF-β1), synergistically induce the expansion of SiglecF high neutrophils, and that this induction would vary across different cancer models.

Flow cytometry was utilized to identify and quantify SiglecF high neutrophils across different cancer models, including KP NSCLC, B16-F10 melanoma, and 4T1 triple-negative breast cancer (TNBC) lines. Cytokine array analysis was conducted to determine the presence and levels of G-CSF, GM-CSF, and TGF-β1 in the TME of these models. ANOVA was used to assess the statistical significance of SiglecF neutrophil populations between the different models and in response to the presence of these cytokines. Statistical significance was defined as p < 0.05.

The KP NSCLC model demonstrated a significantly higher proportion of SiglecF high neutrophils compared to the B16-F10 and 4T1 models, which also had elevated levels compared to the naive control. Cytokine analysis revealed that GM-CSF and TGF-β1 were significantly elevated in the KP model compared to others, while G-CSF was highest in the 4T1 model. ANOVA analysis of in vitro cytokine induction indicated that GM-CSF and G-CSF synergistically induced SiglecF high neutrophils, with the highest induction observed when combined with TGF-β1.

The data suggest that the KP NSCLC TME uniquely promotes the induction of SiglecF high neutrophils, primarily through the synergistic effects of GM-CSF, G-CSF, and TGF-β1. These findings implicate these cytokines as key contributors to tumor progression via SiglecF high neutrophil induction, highlighting the potential for therapeutic targeting of these pathways in NSCLC. Future studies will focus on the inhibitory effects of SiglecF neutrophils on T-cell populations and explore strategies to mitigate their pro-tumoral functions. This research was supported by NIH P20GM135004 (Yan/Chesney) and the Summer Scholar Research Program.


Cytokine-Mediated Induction of SiglecF⁺ High Neutrophils in Non-Small Cell Lung Cancer Tumor Microenvironments

Abstract
Background: Arnold-Chiari malformation presents unique considerations during pregnancy and delivery due to its neurological implications. While cesarean delivery is often considered for safety, limited data exist regarding maternal outcomes across different racial groups and delivery methods in Chiari patients.

Objective: To assess maternal and obstetric outcomes by delivery method (vaginal vs. cesarean) in Black and White pregnant patients diagnosed with Chiari malformation.

Methods: A retrospective cohort study was conducted using TriNetX, an electronic health record-based research network. Separate cohorts of Black and White pregnant patients with Chiari malformation (ICD-10 Q07.0) were constructed. Patients were stratified by delivery method using CPT codes for vaginal and cesarean delivery. Maternal outcomes were compared using ICD-10 codes for gestational hypertension (O13), preeclampsia (O14), eclampsia (O15), perineal laceration (O70), and other labor-related complications (O60–O77).

Results: Among White patients (n=1,522), vaginal delivery was associated with a significantly lower risk of pre-eclampsia compared to cesarean delivery (4.0% vs 7.1%, p=0.02). This association was not statistically significant in Black patients (4.1% vs 5.7%, p=0.246). Although eclampsia was higher in vaginal deliveries for both races, it did not reach statistical significance. Perineal lacerations were significantly more common in vaginal deliveries among Black patients (12.7% vs 2.0%, p<0.001). No significant racial differences were found in rates of headache, fetal heart abnormalities, or hypertensive disorders.
Conclusion: 
White patients with AC malformation had significantly increased risk of pre-eclampsia following cesarean delivery, suggesting a potential benefit of vaginal delivery when feasible. Racial differences in maternal outcomes were limited but suggest potential disparities in delivery-related complications such as perineal trauma. Further studies adjusting for socioeconomic and clinical covariates are warranted.
Key words: Maternal outcomes; Chiari malformation; gestational hypertension, preeclampsia; eclampsia;
Declarations of interest: none
Funding: none


Beyond Assumptions: Racial and Delivery Method Outcomes in Pregnant Patients With Chiari Malformation

Introduction: The treatment of laryngeal cancer often results in alterations of speech and swallowing, which may contribute to the development of depression in this patient population and subsequently affect treatment outcomes and survivorship. This study aims to determine the prevalence of depression in patients with laryngeal cancer and identify contributing risk factors. 

Methods: Patients with laryngeal cancer were identified using ICD-10 codes within the TriNetX Research Network. Patients were stratified by treatment type (surgical versus non-surgical), age, socioeconomic (SES) risk factors, and post-treatment side effects. The effect of these factors on the development of depression was analyzed while controlling for cancer stage. 

Results: 93,018 patients with laryngeal cancer were identified. These patients had a significantly higher risk of depression compared to the overall head and neck cancer population (RR=1.15, 95% CI: [1.12,1.18]). Patients under 70 were more likely to develop depression than those over 70 (RR=1.27, 95% CI: [1.22,1.33]). SES risk factors were associated with over double the risk of depression (RR=2.67, 95% CI: [2.42,2.97]). Surgical treatment conferred a higher risk than non-surgical treatment modalities (RR=1.54, 95% CI: [1.41,1.68]). Patients who experienced long-term complications such as dysphagia, xerostomia, or malnutrition had a 7.62% increased risk of depression (p&lt;0.01). 

Conclusions: Patients with laryngeal cancer face a uniquely high risk of developing post-treatment depression. Future studies evaluating optimal screening protocols and supportive care strategies may help improve mental health care in this population.

Depression Among Patients With Laryngeal Cancer: A Retrospective TriNetX Study

Background
Developing novel, safe treatments for adolescent depression is imperative, as rates of depression are rising and conventional pharmacotherapies are often limited by modest efficacy and considerable side effects. Stanford Accelerated Intelligent Neuromodulation Therapy (SAINTTM), an optimized precision-targeted form of transcranial magnetic stimulation, is FDA-cleared for treatment-resistant depression in adults but remains understudied in adolescents. This pilot study investigates if SAINT, a rapid-acting noninvasive intervention with minimal side effects, can address this urgent need. We evaluate the safety, tolerability, and efficacy of SAINT in treatment-naïve adolescents experiencing their first episode of depression who remain antidepressant-free throughout treatment. By studying this demographic, we provide compelling evidence for SAINT as a first-line treatment. SAINT, however, requires a substantial time commitment—5 consecutive 10-hour days. This study introduces Truncated SAINT: a half-dose protocol of five consecutive 5-hour days supported by SAINT trials showing depression remission in highly treatment-resistant subjects after an average of 26 sessions. We hypothesize Truncated SAINT will yield equivalent antidepressant effects to Standard SAINT, with greater acceptability, preference, convenience, and overall satisfaction.

Methods
Two study arms will recruit 20 participants each to receive Standard or Truncated SAINT in a single-blind open-label design. The primary outcome is change in Children’s Depression Rating Scale-Revised (CDRS-R) scores one month post-treatment. Pre- and post-treatment interviews with adolescents and their parents provide measures of acceptability, preference, convenience, and satisfaction. All participants receive a brain resting-state functional magnetic resonance image (rs-fMRI) before treatment to identify the personalized treatment target. Two additional brain rs-fMRIs are obtained at the immediate and one-month follow-up visits to investigate changes in functional connectivity from baseline. Participants remain antidepressant-free for one month post-treatment, after which they resume standard psychiatric care. Participants are followed for 6 months to assess treatment durability.

Results
As of this writing, four participants have completed treatment for the study with three completing the primary outcome measure CDRS-R at the one-month follow-up. No adverse events have been reported. CDRS-R decreased from 60 to 22, 36 to 17, and 49 to 21 from baseline to one-month post-treatment in these subjects. 

Conclusion
Preliminary results suggest that both Standard and Truncated SAINT are safe and efficacious with no adverse events reported. Both participants who completed their one-month follow-up achieved depression remission. We expect SAINT to continue to demonstrate safety and efficacy and ultimately be considered as a first-line treatment for adolescent depression.


Investigating a Truncated Version of SAINT in Treatment-Naive Adolescents with Depression: An Open-Label Acceptability Trial.

Background
Cardiac myxomas are rare, benign tumors and the most common primary cardiac neoplasms. They are typically found in the left atrium and may present with nonspecific symptoms related to obstruction, embolization, or systemic effects. These tumors can be life-threatening if not detected and treated promptly. Because of their subtle and varied presentation, they can be overlooked until symptoms become severe. Cardiac imaging, such as transthoracic or transesophageal echocardiogram (TTE/TEE) is essential to accurate diagnosis. While traditional management involves open thoracotomy, video-assisted thoracoscopic surgery (VATS) has emerged as a less invasive alternative in select cases. Close follow-up with cardiology is important for management of supraventricular arrythmias that may occur after surgery. 

Case Presentation
A 49-year-old woman with no significant past medical history began experiencing dizziness, fatigue, chest discomfort, and near-syncopal episodes over a two-month period, worse with her typical exercise and running patterns. Initial neurologic evaluation, including brain MRI, showed a small colloid cyst and nonspecific left parietal findings but no clear explanation for her symptoms. An initial cardiac evaluation, which did not include imaging, was unremarkable and reassurance was given. After persistent symptoms, her primary care provider ordered a TTE, which revealed a large, mobile mass measuring 6.3 x 3.6 cm nearly filling the left atrium and prolapsing into the left ventricle, suspicious for a myxoma. She was urgently referred and underwent coronary angiography, which showed normal vessels and incidental blood flow to the mass. The patient quickly proceeded to surgical resection of the mass via thoracotomy. Pathology confirmed a left atrial myxoma measuring 6.2 x 5.5 x 2.5 cm. Postoperatively, she developed symptomatic atrial flutter with rates >200 bpm, thought to have stemmed from the right atriotomy during surgery. It was initially managed with TEE-guided cardioversion and ultimately resolved with catheter ablation. She has since returned to sinus rhythm and normal activity.
Discussion
This case highlights how cardiac myxomas can present with vague, nonspecific symptoms and mimic neurologic conditions, leading to delays in diagnosis. In this patient, echocardiography was critical in identifying the underlying cause of her symptoms. Prompt surgical resection prevented potential complications such as embolism or sudden cardiac death. The patient’s postoperative atrial arrhythmias underscore the importance of ongoing cardiac monitoring and follow-up. Clinicians should consider cardiac etiologies in patients with unexplained presyncope or exertional symptoms, especially when neurologic workup is inconclusive. Early detection and multidisciplinary care can lead to excellent outcomes in cases like this.


Unexplained Presyncope Unveils a Cardiac Myxoma: A Case of Delayed Diagnosis

Budd-Chiari Syndrome (BCS) is a rare condition characterized by hepatic venous outflow tract obstruction not caused by cardiac disease, pericardial disease, or sinusoidal obstruction syndrome. We present a case of BCS initially mistaken for metastatic liver cancer, highlighting the diagnostic challenge posed by its variable radiologic appearance.
A 36-year-old woman with a past medical history of irritable bowel syndrome (IBS) presented to the emergency department with abdominal pain and distension, nausea, bilious vomiting, and bilateral lower extremity edema. She had been diagnosed with IBS by her primary care provider after months of abdominal discomfort and diarrhea. She denied alcohol, tobacco, or illicit drug use, and her only medication was an oral contraceptive pill. On presentation, she was hemodynamically stable. Laboratory studies were notable for WBC 12.09 × 10⁹/L, total bilirubin 1.9 mg/dL, and alkaline phosphatase 172 U/L; AST and ALT were within normal limits. A CT scan of the abdomen and pelvis showed hepatic parenchymal heterogeneity with nodular, enhancing foci and moderate ascites. A viral hepatitis panel was negative, and CA-125 was elevated at 237 U/mL. MRI revealed an abnormal, heterogeneous liver with morphologic distortion, and compression, but no occlusion, of the hepatic veins. Persistent ascites and omental abnormalities raised concern for peritoneal carcinomatosis. Differential considerations included multifocal hepatic adenomas, focal nodular hyperplasia, or metastatic disease. The caliber of the hepatic veins suggested possible BCS. A percutaneous liver biopsy and peritoneal fluid cytology were obtained. Liver histology showed cirrhotic changes consistent with BCS, without evidence of malignancy. Cytology of ascitic fluid was negative for malignant cells. Additional workup, including alpha-fetoprotein (2.8 ng/mL), autoimmune liver markers, and ceruloplasmin, was unremarkable. The patient was diagnosed with cirrhosis secondary to BCS with features of portal hypertension. She underwent Transjugular Intrahepatic Portosystemic Shunt (TIPS) placement, which reduced the portosystemic gradient from 15 mmHg to 8 mmHg. The patient’s symptoms resolved and she was discharged home on lactulose and apixaban.
This case illustrates how Budd-Chiari syndrome can mimic hepatic malignancy on imaging due to liver heterogeneity and nodularity, vascular distortion, hepatomegaly and ascites. Early recognition and appropriate intervention, such as TIPS, can lead to clinical improvement and avoidance of unnecessary oncologic treatments. Clinicians should maintain a high index of suspicion for BCS in patients presenting with abnormal liver imaging, ascites, and risk factors for venous thrombosis. Timely biopsy and hepatic venous assessment are crucial for accurate diagnosis.

Budd-Chiari Syndrome Masquerading as Hepatocellular Carcinoma

Background: Pediatric dog bites frequently result in imaging evaluations, but geographic disparities influencing these practices are poorly understood. This study investigates whether rurality of patient residence, defined by Rural-Urban Commuting Area (RUCA) codes, impacts imaging use in pediatric dog-bite injuries.

Methods: A cross-sectional analysis was conducted using data from the Pediatric Health Information System (PHIS), which included 13,901 pediatric dog-bite encounters across 45 U.S. children's hospitals from 2021 to 2022. Patient residences were categorized by RUCA codes, ranging from 1 (most urban) to 10 (most rural). Multivariable logistic regression adjusted for demographics, insurance status, hospital disposition, and length of hospital stay, assessing associations between RUCA codes and the likelihood of diagnostic imaging (e.g., radiographs, CT scans, ultrasounds).

Results: Imaging was performed in 21.5% (2,992/13,901) of pediatric dog-bite encounters. Stratification by RUCA classification revealed an increasing trend of imaging use in more rural areas, with the lowest imaging rates in the most urban groups and the highest in the most rural groups. Linear regression analysis revealed a significant positive correlation between RUCA codes and imaging utilization (p = .002), accounting for approximately 72% of the variance in imaging rates (R² = 0.72). Logistic regression further indicated that each one-point RUCA increase was independently associated with higher odds of imaging (β = 0.042 per RUCA point; OR ≈ 1.04 per point; p = .0016). Additionally, longer hospital stays were significantly linked to increased imaging utilization (β = 0.341 per day; OR ≈ 1.41 per day; p < .001).

Conclusion: Pediatric patients from rural areas are significantly more likely to undergo diagnostic imaging for dog-bite injuries than their urban counterparts. This disparity likely reflects differences in available resources, clinical practice patterns, and provider experience rather than differences in injury severity. Although imaging may provide reassurance in rural settings, unnecessary utilization raises concerns about excess radiation exposure and healthcare costs. These findings emphasize the importance of establishing standardized, evidence-based imaging guidelines and enhancing pediatric-specific resources in rural trauma centers to ensure equitable and appropriate pediatric trauma care.

Impact of Rural-Urban Commuting Area Codes on Imaging Utilization in Pediatric Dog-Bite Injuries

Background:
Psoriasis affects 3% of US adults, with treatment costs exceeding $35 billion annually. While biologic therapies have transformed treatment and management, current risk stratification tools fail to predict disease severity, limiting opportunities for early intervention. Traditional single-biomarker approaches show poor predictive accuracy, while evidence-based non-pharmacologic interventions remain underutilized due to inability to identify high-risk patients. Recent machine learning advances in chronic disease phenotyping suggest multi-dimensional clustering may outperform conventional approaches. This study applied unsupervised machine learning to identify clinically meaningful psoriasis phenotypes using routinely available inflammatory, nutritional, and sociodemographic data, aiming to transform risk stratification and enable precision-targeted interventions.

Methods:
We analyzed nationally representative NHANES data (2003-2006, 2011-2014) including 380 adults ≥20 years with complete psoriasis, inflammatory marker, dietary, and sociodemographic data. Psoriasis severity was assessed by self-reported body surface area (BSA) using standardized palm-size references (<1, 1-2, 3-10, >10 palms). Mixed-data clustering utilized Gower's distance metrics and hierarchical clustering algorithms optimized for heterogeneous clinical data. Cluster stability was rigorously validated via bootstrapped Jaccard coefficients (2000 iterations). Survey-weighted analyses accounted for complex sampling design. Multivariable logistic regression models, adjusted for age, sex, race/ethnicity, and alcohol intake, quantified associations between cluster membership and disease severity.

Results:
Two distinct and clinically meaningful phenotypes emerged with high reproducibility (mean Jaccard=0.77). Cluster 2 ("High-Risk Phenotype") demonstrated more severe psoriasis (greater BSA involvement), vitamin deficiencies (D, B2, B12), poor dietary quality, elevated systemic immune-inflammation indices (SII, SIRI residuals), and distinct inflammatory profiles. In adjusted analyses, High-Risk Phenotype membership conferred >2-fold increased odds of severe psoriasis (BSA >10 palms vs. <1 palm: OR=2.24, 95%CI: 1.04-4.80, p=0.038). Binary severity prediction showed even stronger associations (BSA ≥3 vs. <3 palms: OR=2.65, 95%CI: 1.42-4.94, p=0.003), demonstrating superior discriminative ability compared to individual risk factors.


Conclusion:
This machine learning approach identified two clinically distinct psoriasis phenotypes that significantly outperform traditional single-marker prediction models. The High-Risk Phenotype, characterized by systemic inflammation and nutritional deficiencies, enables early identification of patients requiring intensive intervention. These findings support implementation of precision medicine approaches including targeted nutritional supplementation, anti-inflammatory dietary interventions, and earlier biologic therapy consideration. Lastly, these findings underscore potential for health system improvement through EHR-integrated phenotyping using routinely collected laboratory and survey data, potentially reducing healthcare costs while improving patient outcomes through personalized risk-stratified care delivery.

Predictors of High-Risk Psoriasis Phenotypes Identified by Machine Learning in NHANES

Background
Resident-led code status discussions are common in teaching hospitals, yet trainees often lack formal training to navigate these complex conversations effectively. Prior studies show that patients frequently hold unrealistic expectations about resuscitation and are more likely to opt for a do-not-resuscitate (DNR) status when provided with education that supports informed decision-making. Moreover, DNR orders have been associated with the inappropriate withholding of peri-arrest interventions. This study aims to assess internal medicine (IM) residents’ comfort, self-reported practices, and observations related to code status conversations to identify educational gaps and inform targeted training.
Methods
We designed and administered a survey to all residents in our IM program, collecting demographic information, including training level. Residents rated their comfort and knowledge of code status, resuscitation components, and expected outcomes on a 5-point Likert scale. Additional items addressed the frequency, depth, and barriers to discussions; patient-related factors, documentation practices, observations regarding peri-arrest interventions for DNR patients, and conflicting code statuses.
Results
Of 115 residents, 51 responded; 49% were interns. Over 90% expressed comfort discussing code status, but only 72% felt knowledgeable about the expected outcomes of in-hospital cardiac arrest. Discussions occurred most frequently on hospital (74.4%) and ICU admissions (60.5%), but less often during ICU transfers. Major barriers included limited time and patient anxiety. Only 44% reported consistently documenting their conversations. Although over 90% of residents felt familiar with code status components, fewer than half routinely discussed all interventions in the full code bundle. Patient acuity strongly influenced the level of detail of conversations. Of note, 40% of residents reported that ICU transfer, intubation, or vasopressors were ‘always’ or ‘often’ withheld for DNR patients. Additionally, 65% had encountered conflicting code status orders on multiple occasions.
Conclusion
These findings reveal significant knowledge gaps and inconsistencies in how code status conversations and decisions are approached. While residents are having code status conversations more frequently than documented, there is considerable variation based on the stage of care and interventions discussed. Documentation was lacking, and DNR orders were frequently misinterpreted. Moreover, patients’ requests for incompatible interventions underscore persistent misunderstandings. Targeted education is critical to ensure that residents engage in informed, comprehensive, and patient-centered code status conversations. Improving training in this area will enhance decision-making, align care with both clinical standards and patient values, and ultimately improve patient outcomes.

Resident Approaches to Code Status Discussions – Perspectives and Practices

Background: The rise of ChatGPT and similar large language models (LLMs) transformed how medical and dental students accessed information and engaged in learning. This technological shift presented a challenge for educators, necessitating curriculum and policy adjustments. While existing literature has focused on how educators should integrate artificial intelligence tools into curricula, less attention has been placed on the perspectives and actual practices of students themselves. This study aimed to bridge that gap by investigating the usage of and attitudes towards LLMs, centering primarily on OpenAI’s GPT, among medical and dental students at a US allopathic medical school, thereby informing more effective educational strategies.
Methods: A cross-sectional survey was distributed via email to all medical and dental students at a single university, with responses collected through Qualtrics. The survey included multiple-choice, Likert scale, and free-text questions, and remained open for two months. Quantitative data were analyzed descriptively, and inductive thematic analysis was performed on free-text responses.
Results: A total of 160 students completed the survey. Of these, 88% reported using large language models, with 89% preferring OpenAI’s free version of ChatGPT. Overall, 73% of students used these models to supplement their medical/dental education or research. Common uses included quick information retrieval (82%), assistance with school assignments (56%), tutoring (30%), and research assistance/literature reviews (26%). Most students (81%) felt that integrating AI technologies is necessary to keep pace with medical advances but emphasized the need for proper training on the use of artificial intelligence. Thematic analysis revealed a strong student preference for open discussions that promote LLMs as beneficial tools when used appropriately, rather than restrictive bans or negative framings, which some felt could foster irresponsible or covert use. Concerning usage patterns were also identified: 16% of respondents reported not fact-checking LLM-generated information, and most students indicated they do not cite AI tools as sources.
Conclusion: Medical and dental students have integrated ChatGPT and similar large language models into their education. Their widespread adoption, driven by perceived usefulness and efficiency, underscores a critical need for structured educational interventions. The usage patterns identified, including inconsistent fact-checking and lack of citation, highlight significant risks to academic integrity and patient safety as students transition into clinical roles. Students believe educators must develop and implement curricula focused on responsible and effective AI usage, advocating for transparent guidance that embraces AI’s potential rather than discouraging its use. Equipping future clinicians with tools and knowledge will be essential for navigating AI responsibly in both academics and clinical practice.

Premium content

Next from AMA Research Challenge 2025

Cytokine-Mediated Induction of SiglecF⁺ High Neutrophils in Non-Small Cell Lung Cancer Tumor Microenvironments

Similar lecture

The gut microbiome as a driver of foraging behaviour in predator-prey interactions.

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES