United States

Background
University-sponsored mobile health vans are a popular method of supporting individuals with significant barriers to receiving healthcare by tailoring services to the community’s needs. Their main challenges are longevity and sustainability. Departure of an established van program from a community without other programming in place can have negative effects on the neighborhood’s health. These weaknesses must be addressed prior to developing this program.

Methods
Eight qualitative interviews were conducted with key informants to provide a deeper understanding of mobile health financing and operations sustainability. These interviews focused on the programs at five reputed university-affiliated mobile health vans in different stages of sustainability. Participants were identified through snowball sampling. All interviews were recorded using Zoom. Transcripts were coded and analyzed using NVIVO. 

Results
Four emerging themes of funding, affiliation, sustainability, and community focus were identified. A series of recommendation questions were developed from the specific findings of each theme and compiled into a guide. The main downfalls of mobile health vans stem from inconsistent funding and universities not being fully committed to communities. Vans can sidestep most limiting factors they run into by having preemptive plans in place and a clear mission statement that the university can stay grounded to. 

Conclusion
University-sponsored mobile health vans have potential if they are created with the community at the center of their intentions. The recommendations developed from this qualitative analysis will help to create a sustainability and implementation guide for university-sponsored mobile health vans and secure support for communities in need.

AMA Research Challenge 2024 

Creating a Sustainability and Implementation Guide for University-Sponsored Mobile Health Vans

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-23 is the Thirty-Seventh AAAI Conference on Artificial Intelligence. The theme of this conference is to create collaborative bridges within and beyond AI. Like previous AAAI conferences, AAAI-23 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and two new activities: a Bridge Program and a Lab Program. Many of these activities are tailored to the theme of bridges and all are selected according to the highest standards, with additional programs for students and young researchers. 
AAAI is providing you with a conference planner, which you can use to help organize your itinerary of activities. This includes talks to attend in person, talks to attend remotely, breaks with colleagues and your site seeing activities. To access this conference planner, please go to [https://aaai-2023.takemobi.io](https://aaai-2023.takemobi.io).

In order to access this site, you need to register. If you haven't already, please register [here](https://aaai.org/Conferences/AAAI-23/registration/).


AAAI 2023

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines.

poster

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).

ChatGPT-3.5, or the "Generative Pre-trained Transformer 3.5," is an artificial intelligence (AI) chatbot that is pre-trained on a diverse range of internet text to understand and generate human-like text. This internet text may contain inherent biases that perpetuate debunked medical misconceptions, potentially worsening health disparities among patients. It is hypothesized that ChatGPT-3.5 correctly identifies a disease/clinical diagnosis as having a trend towards one gender over another, but mistakenly over-represents this prevalence ratio. The model was interrogated 20 times in a row with the following prompt: “Compose a brief presentation of a patient presenting with [CONDITON]. Please include complete demographic information.” It was presented with four medical conditions that are more commonly diagnosed in females (anxiety, domestic abuse, osteoporosis, and UTI) and four medical conditions more commonly diagnosed in males (autism, HIV/AIDS, myocardial infarction, and COPD). This study showed that ChatGPT-3.5 exhibited gender bias when asked to create a patient presentation for osteoporosis, domestic abuse, and autism. These results shed light on the detrimental effects of excessive reliance on artificial intelligence (AI) in patient care, where patients may inaccurately self-diagnose based on outdated information.

Decoding Bias in ChatGPT-3.5: Does Artificial Intelligence Truly Know Best?

Abstract Title: Development of a new questionnaire to assess the social impact of Non-Communicable Diseases.

Background: Non-communicable diseases (NCDs) have a considerable impact on an
individual's social life. Life-long medication, fear of death, dietary restrictions, hospital admissions, dependency on caregivers, and strained interpersonal relationships play a role. The available questionnaires measure only individual aspects of social impact.

Objective: To develop and validate a comprehensive questionnaire to assess the social impact of NCDs.

Methods: A cross-sectional study was conducted among patients and caregivers
seeking care at a tertiary care hospital for diabetes, hypertension, chronic obstructive pulmonary disease (COPD), stroke, coronary artery disease (CAD), and cancer. Thirty patients, each with the above diseases, were administered the Cousineau Self-Perceived Burden Scale to assess perceived burden among patients, the Caregiver Burden Scale to assess caregiver burden, EuroQol-5D-3L to assess the quality of life, and face validated questions of sociodemographic details, cost of healthcare, and stress due to disease.
Rasch Analysis was performed to identify items that lie within the traditional bounds of mean-square item fit. The Delphi method with ten subject experts was used to identify the questions that would be included in the final tool.

Results: The data consists of 150 observations, with 86 categorical questions
subdivided into various domains. From this, a total of 34 questions were extracted by Rasch analysis. The new questionnaire had good internal consistency (Cronbach’s
alpha: 0.969, 95% CI:0.96-0.98). The questionnaire was finalized after reaching
a consensus among subject experts.

Conclusion: The social impact questionnaire developed consisted of 30 questions and has good internal consistency and can be used to measure the social impact of NCDs.

Development of a new tool to assess the Social Impact scale of Non-Communicable Diseases.

Abstract Title
“Ecological Stressors and Mental Health of Adolescent Migrants in Athens, Greece.”

Background
In Europe, political and cultural reactions have "securitized" refugees, depicting them as threats. Media emphasizes their "otherness" and neglects their perspectives, obscuring their daily struggles and exacerbating psychological distress. Children and youth are particularly affected, as adverse experiences during pre-migration, transit, and post-migration correlate with long-term physiological and psychological issues.

Methods
The participants (ages 10-25), originating from Afghanistan, Iran, Iraq, the Congo, Cameroon, or Uganda, were recruited from refugee community centers in Athens. A 47-question survey included: 1) Refugee Health Screener (RHS) and 2) Demographics and Ecological Risk. The RHS had fourteen 0-4 scale questions and a 0-10 distress thermometer. Scores ≥12 on questions 1-14, or ≥5 on the thermometer, indicated likely emotional distress. The second subsection covered demographics and post-migration ecological risk factors, converted into binary variables with "1" indicating present risk. The seven risk factors were parental death, presence of a close friend, residential mobility, water security, food security, health care security, and access to electricity. Each participant received a cumulative risk score from 0 (no risk) to 8 (all risks present).

Results
All but 2 participants reached the threshold for emotional distress. Significant positive Point Biserial correlations were each found between lack of access to clean water and electricity and RHS scores. A positive correlation also existed between perceived negative attitudes from their host society and RHS scores. The mean cumulative risk score was 2.20. There was a positive correlation between cumulative risk score and the distress thermometer. Ecological cumulative risk was positively correlated with the number of physical (but not psychological) symptoms reported on the RHS.

Conclusion
The study highlights the impact of cumulative risk on youth refugee well-being. While no significant correlation was found between overall Refugee Health Screener (RHS) scores and cumulative risk, there was a significant correlation between cumulative risk and RHS physical symptom scores, suggesting somatization as a defense mechanism. Lack of access to clean water and electricity correlated with overall RHS scores, indicating physical and psychological symptoms like anxiety, depression, and PTSD. These factors are significant due to their foundational role in physiological and safety needs. Pre-migration trauma predicts mental disorders and PTSD, but post-migration factors are equally crucial for mental health. Post-migration stressors can influence recovery, highlighting the need for therapeutic interventions addressing broader conditions affecting refugees and asylum seekers.


Ecological Stressors and Mental Health of Adolescent Migrants in Athens, Greece

Background
The United States Department of Health and Human Services reports that overdose deaths due to synthetic opioids, such as fentanyl, has increased 103-fold since 1999, with the largest increase starting in 2019. Legalized in Georgia in July 2023, fentanyl test strips (FTS) are an underutilized resource in the fight against the opioid epidemic. As a novel harm-reduction tool, it is crucial to educate future physicians on the importance and uses of FTS. Near-peer teaching is a technique in medical school education that provides advantages to both the trainee and instructor. 

Methods
Pre-clerkship students engaged in a 1-hour training session, incorporated into the harm-reduction curriculum, focusing on FTS and Narcan. The session involved a knowledge-based lecture on the “Hows and Whys” of FTS and hands-on demonstration using an FTS. Students received a survey before and after the workshop, which consisted of 5 questions, separated by knowledge and attitudes. These results were analyzed using t-tests to identify significant differences in both knowledge and attitudes.

Results
Overall, 144 pre-clerkship medical students completed the pre- and post-surveys, which showed a significant growth of knowledge regarding legality, possible applications, and usages (p < 0.001). Additionally, students indicated improved perceptions towards the use of FTS, especially the use not only among high-risk patient populations but also as a broad harm-reduction measure in a diverse patient population, including non-chronic or recreational drug users (p < 0.001). 

Conclusion
Through near-peer teaching and demonstration, pre-clerkship medical students improved their knowledge of and attitudes towards both FTS generally and the importance of FTS in addressing the opioid crisis and overdose deaths. As a preventative, inexpensive, and relatively new strategy to fight the opioid epidemic, it is vital to educate medical students on FTS across the United States. Through a 1-hour training workshop, other healthcare schools and residency programs can easily practice preventative medicine and contribute to saving lives against the addictive forces of opioids. This would greatly expand the number of students and healthcare workers trained to administer harm reduction tactics, such as FTS. Furthermore, a decreased knowledge gap and positive improvement in attitude towards FTS suggests that near-peer teaching is an effective instrument that should be utilized in medical education more often. Additionally, we recognize that its use among multiple graduate schools would further the integration of and partnership between many healthcare professionals.

Efficacy of Fentanyl Test Strip Education and Training of Medical Students Using Near-Peer Teaching

Background: Telehealth has the potential to increase healthcare access for people who use drugs (PWUD), a typically medically underserved population, by overcoming geographical and temporal barriers. However, PWUD’s access to telehealth and necessary technological infrastructure, such as the internet and computers, remains understudied. We studied the access to digital technology and telehealth utilization among PWUD to inform the planning of telehealth interventions tailored to their needs.
Methods: We employed respondent-driven sampling to recruit 162 PWUD to participate in a 109-item structured questionnaire. Eligibility criteria required participants to be at least 18 years old, have a history of injection drug use (IDU), and Greek verbal fluency. The main study outcomes included current internet and computer access, and telehealth use. We used logistic regression to evaluate the impact of sociodemographic characteristics and IDU on these outcomes.
Results: The participants' mean (standard deviation) age was 45.9 (8.8) years, with 84.0% (136/162) being male and 90.1% (146/162) Greek. Among the participants, 77.8% (126/162) reported IDU within the past year, 85.2% (138/162) were not currently linked to opioid agonist treatment, and 50.0% (81/162) were currently experiencing homelessness. Internet and computer access were reported by 66.0% (107/162) and 31.5% (51/162) of participants, respectively. Compared to participants with secure housing, those currently experiencing homelessness reported decreased internet (50.6% vs 81.5%, p<0.001) and computer access (11.1% vs 51.9%, p<0.001). Only 1.9% (3/162) had ever used telemedicine, and 24.1% (39/162) reported awareness of telemedicine as a healthcare delivery modality. Multivariable analyses revealed that increasing age (per 1-year increase: odds ratio [OR]=0.94, 95% confidence interval [CI], [0.89, 0.99], p=0.03), IDU within the past year (0.29 [0.10, 0.88], p=0.03), and homelessness (0.29 [0.13, 0.65], p=0.003) were associated with lower odds of internet access. Unemployment (0.25 [0.07, 0.98], p=0.05) was also associated with lower odds of computer access.
Conclusion: Homelessness, older age, unemployment, and IDU within the past year were associated with limited access to digital technology among PWUD. These factors, along with the need to enhance patient education on telehealth, should be considered in policy development to ensure equitable access to digital healthcare for underserved populations.

Exploring digital healthcare access among people who use drugs

Purpose: In California, nearly half of all adults have prediabetes or diabetes. In Fresno County, an even higher percentage is present in minority communities. Due to the lack of research on the culturally specific and socioeconomic barriers to care for these patients, this study focused on furthering research in these regards at United Healthcare Center Fowler, a community healthcare center in Fowler, CA.
Methods: A mixed-methods approach to identify barriers to care in Fowler, featuring both surveys and interviews. Surveys were based on the Diabetes Self-Management Questionnaire-Revised (DSMQ-R) and self-reported barriers the patients were facing. Surveys were offered in English and Spanish. Barriers listed included diet, knowledge regarding diabetes management, and issues related to doctor's appointments, among others. Interviews were conducted voluntarily and limited to those with HbA1c levels greater than 5.7%. They focused on three topics: a personal story related to their life with diabetes, their self-management of diabetes, and their biggest struggle to self-manage diabetes.
Results: The most notable barriers to the management of diabetes in Fowler were lack of access to affordable, healthy dietary options and maintaining a diet that was both healthy and culturally specific, both of which greater than 50% of patients reported as barriers.
Conclusion: By identifying that barriers to dietary options are a vital barrier in managing diabetes in the Fowler community, we can anticipate future interventions to ease this burden, including CalFresh pamphlets regarding healthy food options, as well as patient-specific nutritional counseling that factor in the cultural aspects of their diets.

Identifying Barriers to Diabetes Management Among Patients at UHC Fowler

Background
The American Cancer Society (ACS) is a national non-profit that aims to eliminate cancer via research and lobbying, the latter of which is done through the Cancer Action Network (ACS CAN). ACS CAN hosts Cancer Action Days for survivors and other people involved with ACS to lobby at their state government offices. The bills lobbied for this year were on access to lung cancer screening, creating an adult vaccination record, and protecting genetic information from insurance calculation. It would be important to know if the process of lobbying increases the lobbyists’ agreement with the bills. 

Methods
A questionnaire was drafted using the yearly questions associated with Cancer Action Day that captured demographic data as well as bill opinions graded by 7 point Likert score. The questionnaire was duplicated into a pre- and post- survey with conglomerate data collection. The surveys were administered at the New York State Capitol on 5/7/24. 

Results
42 respondents filled out the pre-survey, 30 respondents filled out the post-survey, and of these there were 20 people who filled out both surveys allowing us to compare these groups. The people who were at Cancer Action Day were cancer survivors (those who completed cancer treatment, 37%), cancer thrivers (those undergoing treatment, 6%), caregivers (those who take care of people with cancer, 20%), lobbyists (11%), and other (26%). Ninety six percent of attendees knew someone with cancer, including themselves (46%), a family member (21%), a work colleague (4%), and more than one person (25%). Most people who attended Cancer Action Day were in the age range of 55-65 (31%), followed by 45-55 (24%), with the median age of participants being in the range of 45-55. The average Likert score of satisfaction on access to lung cancer screening increased from 3 to 4 in pre to post survey (T-test unequal variances, 0.064>0.05). The average Likert score of agreement on increasing vaccinations for adult cancers remained at 6 (0.262>0.05). The average Likert score of agreement on protecting genetic information decreased from 7 to 6(0.274>0.05).

Conclusion
The increased Likert score for lung cancer, and the decrease for genetic information may indicate that through lobbying individuals learn more about the disparities in access to these levels or care. In conclusion, the ACS CAN lobbying changes people’s opinions on current political issues in cancer without statistical significance. For future research, it would be important to expand the sample size of the study by incorporating multiple lobbying events.

Impact of Lobbying Efforts of the American Cancer Society at New York Cancer Action Day

Background
The National Institute of Drug Abuse (NIDA) asserts that effective treatment for substance use disorder (SUD) must address a diverse range of individual needs. Recovery-oriented systems of care (ROSC) have been established to respond to the long-term needs of individuals with SUD. Their mechanism of change includes community-based networks that prevent, intervene in, and treat SUD. In Illinois, ROSC networks include a diverse range of stakeholders at the county level (i.e., healthcare providers, law enforcement, local businesses, and individuals with a history of substance use) who work together to build effective care coordination systems. Understanding how members view their network’s effectiveness is important in determining their utility in the coordination of SUD treatment and prevention. The objective of this study is to assess the capacity of ROSC networks to improve care coordination and Narcan (naloxone) distribution. 

Methods
ROSC council members in the state of Illinois were identified using publicly available membership rosters and contacted via email with a link to an anonymous survey focused on the participant's involvement with their ROSC council and their perception of its quality and effectiveness (n=61). 

Results
ROSC networks were composed of stakeholders representing recovery programs (30%), social services (28%), substance treatment (25%), and behavioral health (25%) among others. Members participated in their ROSC network for an average of 23 months (16.9) and most attended more than 10 meetings (58.3%). Over half of respondents believed the network improved care coordination to a very great extent (48.8%) or to a great extent (26.8%). A little over half also believed the network improved Narcan distribution by a great or very great extent (51.2%). Perception of better care coordination was associated with better perception of Narcan distribution (X2(15)=29.39, p<.01).

Conclusion
The majority of respondents, representing a diverse range of stakeholders, indicated that ROSC networks improve care coordination and Narcan distribution. There is a statistically significant positive relationship between perceived care coordination and Narcan distribution, suggesting that improved care coordination between community stakeholders through ROSC networks may improve Narcan distribution. ROSC networks are perceived to be effective in coordinating SUD care by those who work within them and therefore may be an effective model to improve SUD treatment and prevention within communities.

Impact of Recovery-Oriented Systems of Care on Substance Use Disorder Coordination and Narcan Distribution in Illinois

Objective: Idiopathic inflammatory myopathies (IIMs) have heterogeneous presentations and require multiple lines of testing, some of which are invasive, to diagnose. Additionally, treatment is chronic and can be costly. Delays in diagnosis are known to occur and can lead to significant morbidity. This study aims to investigate the influence of socioeconomic status (SES) on time to diagnosis, treatment, and improvement in disease status of IIMs. 

Methods: This is a retrospective study patients from 2011-2021 from a university hospital and a county safety-net hospital. Salient clinicopathologic and demographic data including race, ethnicity, sex, zip code, median household income, and insurance data were abstracted. Patients with ICD codes for polymyositis, dermatomyositis, inclusion body myositis, other inflammatory or immune-mediated myositis were included. The modeling of time-to-event and binary outcomes incorporated covariates such as median household income, race, ethnicity, sex, insurance status (Charity vs. Medicare/Medicaid vs. Other), and inpatient status (Inpatient = 1, Outpatient = 0). Time-to-event data were analyzed using Cox proportional hazards models through the survival package in R.

Results: Our study included 2,258 subjects, of which 1,734 were from the academic hospital and 524 were from the safety-net hospital. After filtering for patients with the proper diagnosis, there were 46 from the academic and 143 from the safety-net hospital. There is no significant difference in time from first abnormal muscle enzyme level (creatine kinase, CK) to diagnosis between race, median household income, ethnicity, and initial presentation in the inpatient vs. outpatient setting. However, those receiving charity had a 4.58-fold increase in the instantaneous probability of being diagnosed after first positive CK after 230 days (p = 0.009). Those on Medicare/Medicaid had a 62% decrease in the instantaneous probability of being diagnosed after first positive CK within the first 94 days (p = 0.02) but a 4.60-fold increase in the instantaneous probability of being diagnosed after first positive CK after 230 days (p = 0.006). There is no significant difference in the probability of a muscle biopsy being order or the time to disease improvement between races, ethnicities, sexes, income, or insurance status.

Conclusions: Our findings showed that there could be delays in diagnosis for those with Medicare/Medicaid or those receiving charity to finance their care. There was no impact detected from race, ethnicities, or sexes. While income did not clearly correlate with insurance status, income is limited by zip-code level data rather than individual data. Our study sheds light on populations that may have delays in diagnosis and treatment, which could lead to increased disease morbidity. Unfortunately, increased morbidity in the population identified will only perpetuate the cycle of poverty and add greater burden to the healthcare system. Future studies will seek to examine impact of SES and disease outcomes.


Impact of socioeconomic factors on time to myositis diagnosis and treatment


Background
Universal helmeting has been shown to decrease fatalities associated with motorcycle accidents, and helmeting may reduce the incidence of traumatic brain injury, facial fractures, and injury severity for the head and neck. Enhancing rider knowledge of safe riding practices and effective helmeting are approaches that may improve motorcycle safety.

Methods
The dissemination of a motorcycle safety toolkit in areas frequented by riders, or in proximity to vulnerable populations could provide a means for teaching about educational offerings and safety resources. Our team utilized the graphic design platform Canva to create a concise handout highlighting New Jersey helmeting laws, tiers of helmet safety, and crash statistics, among other topics. We utilized quick-response (QR) codes so viewers can interact with the toolkit from their phones. Leaders in roadway safety from across New Jersey were consulted in developing these resources. Collaborators included rider safety educators, injury prevention professionals, and authorities in public safety and policy.

Results
Strong positive feedback was received by the New Jersey Trauma Injury Prevention Council. Several organizations stated they plan to use the toolkit in their community outreach efforts, including:

· New Jersey's Governor's Office
· New Jersey Division of Highway Traffic Safety
· New Jersey Brain Injury Alliance Motorcycle Safety Committee
· AAA - New Jersey Division
· Monmouth and Ocean Traffic Safety Officers Committee.

The toolkit has been received favorably by a cohort of patients hospitalized with motorcycle-related injuries. These individuals received the toolkit together with motivational interviewing emphasizing safe riding.

Conclusion
Efforts may be made to share the toolkit in a variety of settings and to better understand factors contributing to motorcycle-related injury. Our collaborators can broaden the reach of the toolkit, incorporating it into safety courses, sharing it on social media, or using it to guide conversation with survivors. We can track use of the toolkit through analytical tools measuring viewership. A retrospective cohort study investigating the relationship between helmeting and injury severity scores and health outcomes among victims of motor-vehicle collisions locally could identify areas of focus for further education. It remains to be seen where this toolkit will reach the most motorcyclists and whether education is a limiting factor in the prevention of motorcycle-related injury and fatality. A better understanding of the financial and social factors that complicate the use of safe and effective helmets is imperative in reducing fatalities.

Premium content

Next from AMA Research Challenge 2024

Decoding Bias in ChatGPT-3.5: Does Artificial Intelligence Truly Know Best?

Similar lecture

Increasing Impact of Mobile Health Programs: SAHELI for Maternal and Child Care