United States

As the field of genomics continues to grow, the identification of substantial genes as it relates to rare diseases remains a significant challenge. Traditionally, researchers work backwards starting from the disease to determine associated proteins. Then use expensive and time-consuming methods, such as mass spectroscopy or immunofluorescence staining, to determine genetic codes. While this method may work with single gene disorders, it is lacking when dealing with complex, polygenic diseases. To address this, we aimed to develop a novel method of discovering genes that displayed similar phenotypes when knocked out in mice using data from the International Mouse Phenotyping Consortium (IMPC), streamlining gene identification for complex genetic conditions.
We used the National Organization for Rare Diseases (NORD) to select the Map2K2 gene, associated with cardiofaciocutaneous syndrome, as our starting gene. We then used the IMPC database to identify phenotypes associated with Map2k2 knockout mice. These phenotypes were organized into an Excel spreadsheet. Next, we searched the IMPC database for genes that exhibited the same phenotypes when knocked out in mice. This information was compiled into separate Excel tabs, one for each phenotype. We then applied variations of a conditional format to visualize the frequencies of each gene, assigning specific colors based on the number of hits. 
The IMPC database revealed five phenotypes associated with Map2k2 knockout mice: increased mean corpuscular volume, decreased total body fat, decreased body weight, abnormal gait, and decreased body length. Each phenotype had varying amounts of contributing genes, the most being decreased body weight with 403 associated genes, and the least being abnormal gait with 241 associated genes. Of the 947 unique genes found, 129 genes contributed to two phenotypes, 24 contributed to three phenotypes, and four contributed to four phenotypes. Notably, the gene Slc38a10 was the only gene associated with all five Map2k2 knockout phenotypes.
While the Map2k2 gene’s involvement in the MAPK signaling pathway has been extensively studied, literature regarding interacting genes has been sparse. Our analysis identified Slc38a10 as a potential interacting gene, suggesting interplay between the pathways regulated by Map2k2 and Slc38a10. The Slc38a10 gene encodes an amino acid transporter crucial for metabolism and neurotransmission. Given the role of Map2k2 in the MAPK signaling pathway, it is plausible that these two genes have downstream interactions.
Our gene detection method proved to be inexpensive and quick, offering a valuable tool for future genetic research. We hope that novel approaches like ours will contribute to enhancing the understanding of complex genetic disorders and advancing personalized medicine.


AMA Research Challenge 2024 

Unlocking Gene Interactions: Novel Insights from the IMPC Database

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.
 

[![](https://assets.underline.io/markdown_image/1/image/20e1e77aa954efeac7a15d84c85eb383.jpg)](https://underline.io/events/468/sessions?searchGroup=all)
 
[![](https://assets.underline.io/markdown_image/1/image/068522d81aca0b68a369787c7dfccd1b.jpg)](https://underline.io/events/468/posters?searchGroup=all)
 
[![](https://assets.underline.io/markdown_image/1/image/d3971ece79d5e5575b178bdc2c8bbfae.jpg)](https://underline.io/events/468/help-desk)
 

#### 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. 

### 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/)

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

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

Introduction
Normal weight type-2 diabetes (NWD), characterized by body mass index (BMI) below 25 kg/m2, is more prevalent amongst Asian Americans (AA). While recent studies have shown strength training (ST) was superior to aerobic (AER) or combination training (COMB) in lowering HbA1c levels in individuals with NWD, it is unclear if these exercises have the same effects across AA subgroups, which differentiate due to cultural norms.

Methods
We used data from a 9-month randomized control trial for NWD (conducted in San Francisco Bay Area in 2019-2020) to evaluate the effectiveness of weekly planned ST, AER, and COMB training sessions for Chinese and Asian Indian populations. Our primary outcome was change in diabetic control (HbA1c). Secondary outcomes included changes in body composition (e.g. fat and lean mass) and muscle strength (e.g. leg extension).We performed intention-to-treat (ITT) and per-protocol analysis (PP, restricted to those with 50% minimum adherence) to evaluate changes to HbA1c and body composition using repeated measures ANOVA and paired t-tests.

Results
Of 129 participants, 59 were Chinese (19 ST, 20 AER, 20 COMB), and 70 were Asian Indian (26 ST, 20 AER, 24 COMB). Average baseline HbA1c was 7.5% for both groups (p=0.96). Chinese participants (77.3%) showed higher adherence than Asian Indian (56.6%) participants. Chinese participants showed reduced HbA1c for each intervention group (ITT: ST -0.42% [-0.742, -0.11], AER -0.6% [-0.92 -0.28], and COMB -0.13% [-0.43, 0.17], all p < 0.05), with AER intervention yielding the largest decrease (p<0.01). Results from PP provided a similar conclusion. Asian Indians had no meaningful reductions in HbA1c levels ( ITT: ST -0.36% [-0.83, -0.11], AER 0.26% [-0.27, 0.78], and COMB -0.42% [-0.97, 0.13]. For body composition, Chinese participants showed significantly reduced FMI-Z (ST: -1.03 [-1.94, -0.12], AER: -0.17 [-0.31, -0.04], COMB: -0.17 [-0.23, -0.10]) in all intervention groups. Asian Indian participants only showed significant reduction in FMI-Z (fat mass adjusted for height, age, sex) in the ST group -0.22 [-0.40, -0.03]. Body composition changes were associated with improvement in HbA1c.

Conclusion
Overall, AA persons in over 50% of prescribed exercise improved their diabetic control. In Chinese participants, all interventions lowered HbA1c with higher reductions observed in ST and AER interventions. They also experienced significant reductions in fat mass in all exercise interventions. Asian Indian participants showed no significant reductions in HbA1c in any exercise intervention, but showed significant reductions in fat mass in ST. By recognizing unique cultural and social factors that influence AA behaviors and choices, healthcare professionals can develop tailored interventions that foster cultural understanding, thus empowering individuals to take control of their health and reduce the burden of diabetes.

Is Strength or Cardio Training More Effective at Affecting Diabetes Control & Body Composition for Asians? Subgroup Analysis of a Randomized Control Trial.

Introduction: Blood lead screening has become a critical preventive health element of pediatric well-child appointments, as exposure to lead can lead to developmental delays and many other serious health consequences. However, research shows that clinics face difficulty carrying out lead screening recommendations. 
Purpose: This project evaluated rates of lead screening at the UT St. Francis Family Medicine Residency clinic according to the Tennessee Department of Health (TN DOH) guidelines and proposed quality improvement strategies based on several factors that potentially prevent routine screenings. 
Methods: A retrospective chart review was conducted using a random sample of 353 pediatric patients aged 12 to 72 months who had established UT-St. Francis as their primary care clinic. We utilized a retrospective chart review to determine if patients had 12- and 24-month well-child appointments, if lead screenings were completed at these times, and the blood lead levels reported at each screening. We also investigated whether any acute medical issues were coded in patient charts for these well-child encounters and the rates of “catch-up” screenings among patients who had never been screened for lead. To assess physician awareness of lead screening recommendations, an anonymous questionnaire was distributed to the UT St. Francis Family Medicine residents, attendings, and fellows.​
Results: 47% of the sampled patients were screened for lead at 12-month well-child appointments, and 26% were screened at 24 months. This study identified several likely clinical barriers to the completion of the TN DOH recommended lead screenings, which included clinical workflow and few systems-based reminders but excluded physician awareness. Based on these results, we developed several suggestions for future quality improvement projects. 
Conclusions: Our recommendations directly address the identified causes of missed screenings from this study’s analyses and hold potential to increase future lead screening rates based on the clinic’s specific needs. Future study should investigate the short- and long-term effects of these recommendations on subsequent lead screening rates.

Lead Screening in Pediatric Patients at the UT St. Francis Family Medicine Clinic​

Background
Since the COVID-19 pandemic’s onset, utilizing virtual visits has increased significantly. Although a beneficial aspect of implementing virtual visits is to increase access, there have been few studies with a focus on the opinions of African Americans or uninsured patients, populations where distrust in physicians and healthcare is more prevalent. It has been reported that patient-physician trust leads to better self-reported health outcomes, patient satisfaction, health behaviors, quality of life, and decreased symptom severity. Therefore, identifying ways to
increase trust is essential to improving healthcare, especially among these groups.
Our study aims to evaluate the knowledge, attitudes, and experiences of the uninsured community living in Detroit regarding virtual care and trust in physicians and the healthcare system. We hypothesize that factors such as virtual care and free clinic use will be correlated with increased trust.
Methods
This project involves three student-run free clinics in a large Midwestern metropolitan city. We are conducting a survey-based research study evaluating the knowledge, attitudes, and experiences of uninsured people with virtual healthcare services, physician trust, and trust in the overall healthcare system using combined, previously published, survey instruments. A report of the demographics and average responses will analyze the data. Then, using t-tests and ANOVA via the SPSS platform, the effect of virtual care and free clinic use on trust will be evaluated.
Results
Data collection is currently in progress. At this point in time, we have a sample size of 48 uninsured participants above the age of 18 years and approximately an 89% response rate. We aim to have over 50 participants by the conference deadline in order to extrapolate more meaningful interpretations from data analysis. Currently, 71.8% of the participants are African American and 75% earn less than $40,000 per year. Our initial findings show that 84.4% of participants have utilized free clinics before, with 96.4% of those reporting an extremely or mostly positive experience. Furthermore, preliminary analysis finds that 90.6% of participants trust their physician’s judgment about their medical care. In regard to telemedicine, 43.8% have previously accessed health care via phone or video chat, although only 23.8% of users report obtaining “better access to healthcare services by use of telemedicine”.
Conclusion
Further research is necessary to illuminate perceptions of access and trust in free clinics and telehealth services among underserved populations and to determine mechanisms to implement to improve these views.

Perception of Trust, Virtual Care, and Access Among the Uninsured Population: A Multi-Site Study at Student-Run Free Clinics

Background: Trauma, characterized by persistent emotional distress following exposure to a distressing event, is linked to adverse physical and mental health outcomes. Both childhood and adulthood trauma have been associated with increased risk of chronic diseases and psychiatric disorders in adulthood. According to literature, African Americans, compared to their White counterparts, experience higher levels of trauma and greater prevalence of chronic diseases. This study investigates whether African American adults with a history of trauma, regardless of
timing, exhibit a higher chronic disease burden. Secondarily, the study explores the relationship between traumatic experiences and emotional/psychiatric issues within this population.

Methods: This secondary data analysis utilized data from the Health and Retirement Study to examine the association between chronic disease burden and lifetime trauma, childhood trauma, and adulthood trauma. Multiple linear regression analyses were conducted to assess the relationship between each trauma category and chronic disease burden. Multiple logistic regression models were also implemented to examine associations between trauma categories and emotional/psychiatric problems, adjusting for sociodemographic and health-related variables.

Results: Trauma, regardless of timing, was significantly associated with increased chronic disease burden (childhood: p = .008, adulthood: p < .001, lifetime: p < .001) and higher likelihood of emotional/psychiatric problems (childhood: OR = 1.368, 95% CI = 1.220, 1.534; adulthood: OR = 1.375, 95% CI = 1.185, 1.596; lifetime: OR = 1.284, 95% CI = 1.192, 1.382) in adulthood among African American adults. These associations remained robust after adjusting for age, sex, education, BMI, and smoking status.

Conclusion: This study highlights the profound impact of trauma on both physical and mental health outcomes in African American adults. Beyond the well-established biological mechanisms linking trauma to adult health, these findings demonstrate the critical need for healthcare providers to address trauma-related concerns in clinical practice. Encouraging patients to adopt healthy coping strategies and facilitating access to professional support would be essential steps in mitigating the adverse health effects of trauma in this population.

The Association Between Trauma and Chronic Disease Burden in African American Adults

Background: Hairstyles common in Black communities, braids, locs, twists, (BLTs) can present as artifacts, posing unique diagnostic challenges in interpreting radiological imaging, though limited research has been conducted on the current climate of their interpretation.

Purpose: To present a review of hair artifacts in literature and investigate the comfortability and familiarity of physicians and radiology technicians in assessing braids, twists, and locs (BTLs) as hair artifacts on radiological imaging.

Methods: A web-based survey on the experiences and awareness of BTLs on imaging was distributed to physicians in multiple departments within a nationally respected academic medical center. Survey responses were collected between April 2023 through August 2023 and included demographic information, career length, how often they came across hair artifacts in imaging and a quiz to distinguish pathology from hair artifact on chest x-ray images. Respondents also provided reports of their comfortability and familiarity in identifying BTL’s in plain film images.

Results: 46 surveys were collected from various departments, including Temple Radiology (41.3%), Temple Emergency Medicine (21.7%), Temple Jeanes Campus (26.1%), and St. Lukes Bethlehem Radiology Group (2.2%). Demographic data in Table 1 reveal the Radiologist cohort had an average experience of 9.40 years (SD=12.41), Emergency Medicine cohort had 7.11 years (SD=7.08), and Internist cohort had 0.5 years (SD=0.71). Career length (p < 0.0001) and patient load (p = 0.029) significantly affected self-reported scores (p < 0.0001), with longer careers and larger loads associated with higher score of comfort and familiarity among study participants.

Conclusions: Our findings demonstrated that the number of years in practice and the number of patients per day improved physicians’ and technicians' awareness of how Black hairstyles can affect medical imaging. This study presented a radiological phenomenon lacking in literature and provided valuable insights into the impact of experience on physicians' ability to identify and interpret radiographic images affected by styles that can mimic pathology. The conclusions conveyed a need for early education of BLTs in medical training. Further applications of this research may be directed towards comparing pathology localized in lung apices to hairstyles popular in the Black community, assessing the validity of an educational module in increasing awareness, and expanding on patient populations to explore further associations.

The Imperative for Early Education of Common Black Hairstyles in Radiological Imaging: A Pilot Study


Background
This study investigates potential cost reduction by replacing erythromycin with azithromycin for managing PPROM. Effective strategies are vital due to the high risks associated with PPROM. Targeting Hispanic PPROM patients at Centro Medico Episcopal San Lucas in Ponce, Puerto Rico, this research focuses on comparing the cost-effectiveness of azithromycin and erythromycin in extending pregnancy latency.

Methods
Using a retrospective study design with predictive model elements, patient data from 2020 to 2021 at Centro Medico Episcopal San Lucas was analyzed, projecting the expected number of PPROM diagnoses in 2022. Detailed cost data for azithromycin and erythromycin were obtained from hospital’s pharmacy. Costs for erythromycin encompassed the 250mg IV bag and vial, 100 mL saline bag, primary line for initial dosing, and erythromycin 500 mg tablets for treatment continuation. Azithromycin costs were evaluated for each dosing tablet (1g, 250mg, and 500mg). Average total cost of treatment per year was computed based on the projected number of PPROM patients in 2022. Statistical analysis, including a 95% confidence interval, guided the estimation of the sample size.

Results
The expected number of patients with PPROM in 2022 at Centro Médico Episcopal San Lucas is 136 patients. The combined IV and PO erythromycin regimen, incurs a per- patient cost of $449.30, resulting in a total cost of $60,880 for 136 expected PPROM cases in 2022. In contrast, azithromycin costs per patient varies from $1.16 (1g = 4 tablets of 250mg) and $0.98 (1g = 2 tablets of 500mg) with an estimated total cost of $163.50 and $132.80, respectively, for 136 average expected patients in 2022. A population size of 114 measurements are needed for the confidence level to be at 95%. A 99% cost reduction is projected if azithromycin is employed as standard of care for PPROM compared to erythromycin. The economic advantage of azithromycin lies in its lower cost and single-dose regimen, which is effective for managing PPROM. This regimen benefits from improved pharmacokinetics, slower elimination, and rapid intracellular uptake and concentration.

Conclusion
In conclusion, our study provides insights into the potential cost- reduction with the substitution of azithromycin for erythromycin as the standard treatment for PPROM, without compromising clinical efficacy. This study lays the groundwork for conducting a larger clinical trial to further evaluate the safety, efficacy and economic benefits of this treatment approach. The findings of such a trial could have significant implications for optimizing PPROM management protocols and improving healthcare resource utilization.


The Journey for Optimizing Care: Cost-Informed Perspectives on Erythromycin Vs. Azithromycin for PPROM Management in Southern Puerto Rico

There is currently a gap in knowledge regarding the knowledge, trust, and barriers of the Hispanic community regarding the influenza vaccine. A descriptive, mixed quantitative survey was administered to 100 participants at flu clinics as well as various locations throughout Metro Detroit between September 2023 and September 2024 to identify their levels of knowledge, trust, and barriers regarding influenza vaccination. There were a significant proportion of individuals with misconceptions regarding the vaccine, with 59% of respondents unware that it was safe for pregnant women, and 67% under the belief that it could interact with medications. Reported barriers to vaccination included lack of time (31%) and other/unspecified (47%), while only 12% reported a lack of insurance. Contrary to our hypothesis, our study shows that Hispanic patients of Metro Detroit may have sufficient levels of knowledge and trust regarding the influenza vaccine. Barriers may primarily be time-related, as opposed to access-related.

Understanding Vaccine Hesistancy and Barriers to Access in Hispanic Individuals of Metro Detroit

Abstract Title
Unemployment in Trauma Patients with Substance Use Disorder

Background
People with substance use disorders (SUD) experience disproportionately high rates of unemployment. Challenges in attaining and maintaining employment are exacerbated in trauma patients with SUD. Employment has been associated with positive outcomes for individuals with SUD, though the effect of employment on health service utilization is unknown. This study aimed to explore the relationship between employment, emergency department (ED) utilization, and hospital readmission in trauma patients with SUD. 

Methods
This was a retrospective cohort study of adult (age≥ 18) trauma patients with SUD at a University Level 1 trauma center from January 2021 to December 2022 (n=692). Data was collected through chart review in REDCap. Univariate analyses were calculated using Chi-square, two-tailed t, and Wilcoxon rank sum tests. Multivariable logistic regressions were used to evaluate the primary outcomes: ED utilization (≥1 visit), frequent ED utilization (≥3 visits), and hospital readmission within one year of index encounter. A p-value of ≤0.05 was considered significant.

Results
Unemployed patients (UP) were more often female, older, unhoused, and admitted with a lower injury severity score than employed patients. UP were more likely to experience multiple SUDs and a co-occurring mental health disorder. On multivariable regression, employment status was not an independent predictor of ED utilization (OR 0.943 (0.632, 1.40), p=0.770), frequent ED utilization (OR 1.22 (0.776, 1.91), p=0.392), or hospital readmission (OR 1.31 (0.881, 1.95), p=0.181) within one year. However, having a co-occurring mental health diagnosis and experiencing homelessness were independent positive predictors of ED utilization (OR 2.76 (1.86, 4.15), p<0.001; OR 3.15 (1.94, 5.22), p<0.001), frequent ED utilization (OR 2.88 (1.93, 4.31), p<0.001; OR 3.96 (2.52, 6.28), p<0.001), and hospital readmission (OR 2.09 (1.44, 3.04), p<0.001; OR 1.76 (1.14, 2.72), p=0.011). 

Conclusion
While we did not find an independent association between employment and health care utilization, we identified that risk factors for ED and hospital readmission like housing status and mental health history were more common amongst UP, affirming that employment, housing, and mental health are deeply interconnected in determining healthcare outcomes. These results add to a growing body of literature demonstrating that comprehensive social support is paramount in improving SUD outcomes. Medical advocacy around SUD treatment should prioritize policy and programming that will increase the socioeconomic support provided in standard SUD treatment.

Unemployment in Trauma Patients with Substance Use Disorder

Abstract Title:
Exploring the Link Between Smoking and Short-Term Mortality in Hospitalized COVID-19 Patients: Insights from Population Data
Background
Smoking cigarettes represents a prevalent form of tobacco consumption, with cigarette smoke containing harmful substances like nicotine, tar, and various carcinogens, which contribute to a spectrum of health issues. This study extends the discussion of smoking in the context of the COVID-19 pandemic, where data suggests that non-smokers hospitalized with the COVID-19 virus are more vulnerable to complications.
Methods
To investigate the relationship between smoking history and short-term mortality among COVID-19 patients, publicly accessible deidentified statewide hospitalization records of individuals aged ≥18 from Q2 2020 to Q1 2022 in Texas were utilized. Identification of smokers was performed using ICD-10-CM codes F17200, Z8789, and U071. Three models were employed to assess this link: multilevel logistic regression with propensity adjustment, supported by alternative analyses including propensity score matching and multivariable logistic regression.
Results
Nicotine administration has been observed to increase nAChR7 expression and inhibit pro-inflammatory cytokines like TNF-α, IL-1, and IL-6, while stimulating anti-inflammatory cytokines such as IL-10. Moreover, nicotine's potential dual effect on ACE2 receptors—increasing expression while reducing available binding sites for the COVID-19 virus—has been highlighted. Additionally, nicotine has been shown to inhibit NF-κB, a critical transcription factor in the activation of inflammation in various immune cells, and as an agonist of α7nAChRs, it can reduce cytokine storm mediators such as HMGB1. Notably, smoking has been associated with reduced furin (PCSK3) levels.
Conclusion
In conclusion, nicotine, widely consumed as tobacco smoking and recognized as an addictive substance, may harbor therapeutic potential that necessitates further exploration. Analogous pharmacological agents with high addictive potential are presently utilized for therapeutic purposes with appropriate pharmaceutical protocols. Further research into nicotine's standard processing, therapeutic index, dosing, and other aspects is imperative. Despite the inherent limitations of retrospective studies, nicotine has demonstrated mortality benefits, underscoring the importance of further investigation through prospective studies.

Exploring the Paradoxical Relationship Between Smoking and Short-Term Mortality in Hospitalized COVID-19 Patients: Insights from a Population-Based Cohort Study

Background:

Vaccines are a crucial tool in combating infectious diseases, yet not all vaccines are equally effective for every individual. Variations in vaccine response can be influenced by demographic factors, including age, race, gender, and ethnicity, which are sometimes overlooked in different studies. With the advancements in Artificial Intelligence (AI) and Machine Learning (ML), there is now an opportunity to predict vaccine responses more accurately. While most current models use genetic data, our study focuses on cytokine data—small proteins that affect cell interactions. Cytokine profiling measures these proteins to better understand the immune response. We initiated our investigation with two primary influenza vaccines as a pilot: the Live Attenuated Influenza Vaccine (LAIV) and the Trivalent Influenza Vaccine (TIV). This study marks the first-ever use of ML models to predict vaccine response using cytokine profiling in combination with demographic data, aiming to develop more inclusive models in healthcare.

Methods:

We analyzed cytokine profiles from over 300 participants across six influenza studies using the ImmPort database. Data were collected on days 0 (pre-vaccination), 7, and 28. Statistical analyses, including t-tests and ANOVAs, identified cytokine differences by demographics. Participants were classified as responders or non-responders using Hemagglutination Inhibition Assay (HAI) data. ML models were created using pre-vaccination cytokine profiles and demographic data to predict responses to the LAIV and the TIV.
Results

Cytokine profiles significantly differed (P < 0.05) by demographics, with significant differences found in 45/55 cytokines measured. Cytokines when compared by race had the most significant differences, especially between African American and White participants, followed by age, gender, and ethnicity. For TIV, the K-Nearest Neighbors model was the most accurate for predicting responders, with accuracies of 80%, 67%, and 74% for the two A strains and one B strain. Logistic regression was most successful for LAIV with an accuracy of about 80% per strain.

Conclusion:

Our study demonstrates the value of cytokine profiling and the impact of demographics on vaccine response predictions. The integration of cytokine data with demographic factors in ML models can lead to more personalized healthcare and help address health disparities in vaccine efficacy, especially those facing African Americans. Future work will involve validating these models on larger datasets and exploring additional data types, such as flow cytometry, to enhance global health decision-making.

Premium content

Downloads

Next from AMA Research Challenge 2024

Is Strength or Cardio Training More Effective at Affecting Diabetes Control & Body Composition for Asians? Subgroup Analysis of a Randomized Control Trial.

Similar lecture

Investigating Atl3-Related Genes Utilizing Data from the International Mouse Phenotyping Consortium