United States

With the rise in Technological advances and availability of handheld inexpensive ultrasound devices, we investigated the feasibility of cranial ultrasound image acquisition in a moving ambulance using a handheld device. We concluded that cranial ultrasound image acquisition using handheld ultrasound is feasible in a moving ambulance providing reasonable quality images of visualized landmarks with a very quick turnaround time required for TeleGuidance™ &amp; remote image interpretation by a physician in a 4G network with variable signal strength. This experience is the basis of an ongoing clinical trial for using cranial ultrasound for the detection of intracranial hemorrhage in prehospital settings.

AMA Research Challenge 2022

Cranial ultrasound image acquisition and remote physician interpretation in a moving ambulance-simulated test to assess feasibility for pre-hospital EMS applications 

ICPR 2020 is the premier international conference on Pattern Recognition, gathering leading researchers from around the world to present their latest achievements in Pattern Analysis and Recognition, applied Artificial Intelligence and Machine Learning. The topic of the year is " Putting Artificial intelligence to work on Patterns". 
For the first time, ICPR has employed a two-round review process similar to journal submissions, with new papers allowed to be submitted in either the first or second round and papers submitted in the first round and not accepted allowed to be revised and re-submitted for second round review. 

In this edition of the **ICPR 2020** Conference, there are 1407 Oral/Poster presentations organized in five Tracks: 
- Artificial Intelligence, Machine Learning for Pattern Analysis; 
- Biometrics, Human Analysis and Behavior Understanding; Computer Vision, Robotics and Intelligent Systems; 
- Document and Media Analysis; and 
- Image and Signal Processing. 

The **ICPR 2020** Main Conference program also includes 7 Keynotes from prominent world-renowned scientists, 8 Tutorials, 28 Demos, 8 Challenges, 2 Panels, as well as an Industry/Sponsor Exhibition. In addition, 41 Workshops on a range of important current and emerging research topics complete the ICPR 2020 program. 


ICPR 2020

With the rise in Technological advances and availability of handheld inexpensive ultrasound devices, we investigated the feasibility of cranial ultrasound image acquisition in a moving ambulance using a handheld device. We concluded that cranial ultrasound image acquisition using handheld ultrasound is feasible in a moving ambulance providing reasonable quality images of visualized landmarks with a very quick turnaround time required for TeleGuidance™ & remote image interpretation by a physician in a 4G network with variable signal strength. This experience is the basis of an ongoing clinical trial for using cranial ultrasound for the detection of intracranial hemorrhage in prehospital settings.

poster

Welcome to the largest national, multi-specialty research event **featuring research from 750 medical students, residents and fellows, and international medical graduates.**


#### Event Details
The 2022 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 sponsored by Laurel Road.
 

#### Attend Live Forums
**Oct. 20, 6 — 7 p.m. Central**
 **How to plan a career in academic medicine** 
 Learn how to plan a career in academic medicine and get your questions answered. [Access Here](https://underline.io/events/363/lounge)
 
 **Speakers**
![](https://assets.underline.io/uploads/markdown_image/1/image/d5adb133dfc58eb0a6d469e9cedcd5d4.jpg)
Ricardo Correa, MD, EdD
![](https://assets.underline.io/uploads/markdown_image/1/image/5f9437ab87861538bf02b9ca7c0d926e.jpg)
Stuart M. Greenstein, MD
![](https://assets.underline.io/uploads/markdown_image/1/image/05b1b637c99286d558c371f8e8b185ce.jpg)
Stephanie Strohbeen, MD
 
 **Oct. 22, 11 a.m. — Noon Central**
 **How to get published**
 Learn tips to get your research published and get your questions answered. [Access Here](https://underline.io/events/363/lounge)
 
 **Speakers**
![](https://assets.underline.io/uploads/markdown_image/1/image/8ce6235e957d31cb5e6f9097cc6f6972.jpg)
Jessica Li, MD/PhD Candidate
![](https://assets.underline.io/uploads/markdown_image/1/image/9950ef2e0237f54cb3d8280334d42491.jpg)
Charles Lopresto, DO

### 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 the judging or scoring process to select the winner. The winner will be announced on Dec. 7 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)

**There is no registration fee for this event! **

Register to explore research in a variety of topics. AMA members get exclusive access to score posters in the semifinals.

**Please use the link below to register for free if you have not already done so:**

https://www.ama-assn.org/about/events/ama-research-challenge

Registration Is Required

Welcome to the largest national, multi-specialty research event featuring research from 750 medical students, residents and fellows, and international medical graduates.


Background
Dementia and its impact on the aging population is a growing public health concern. Alzheimer’s Disease (AD) is the most common form of neurodegenerative disease and is primarily characterized by a decline in cognitive function that affects a person’s day-to-day activities. The development and diagnosis of AD occurs in stages, beginning with cognitive symptoms in the preclinical stage and progressing to mild cognitive impairment (MCI) then mild, moderate, or severe AD. This study is designed to categorize subjects within the Alzheimer’s Disease Neuroimaging Initiative database to discover alternative indicators for the development and diagnosis of AD. 
Methods
981 AD patients were extracted from the ADNI database and had a cortical SUVr ≥ 1.2, measured using PET imaging utilizing the AV45 biomarker. This indicated a patient as Aβ+, meeting the requirements for AD diagnosis. Subjects were analyzed via one-way MANOVAs examining the impacts of head trauma, smoking, and marital status on the onset of cognitive symptoms, MCI, and AD symptoms. 
Results
In congruence with the literature, results indicate that TBI+ patients report earlier onset of MCI (F (1, 114) = 5.660, p < .05). Additionally, results yield that patients who smoke develop AD symptoms later (F(1,23) = 9.211, p < .01) and patients who are divorced develop cognitive impairment earlier (F (4,813) = 2.750, p <.05). 
Conclusion
Although the stages of AD are well-described, its diagnosis presents a challenge, particularly in early stages, as its symptoms are often attributed to the consequences of aging. These results support previous research describing TBI effects on the development of AD and may provide diagnostic insight regarding the development of AD. Future research should focus on improving the diagnostic criteria for AD, providing guidance to researchers and physicians on the development and implementation of early interventions.

Demographic Categorization of Subjects within the Alzheimer’s Disease Neuroimaging Initiative Database

Background: Automating the process of obtaining data through natural language processing (NLP) can be useful in extracting salient information from unstructured radiology reports. However, to ensure rigorous quality control, scalable gold-standard protocols for manual verification of automatic data are essential. Our aim was to develop a standardized pipeline for the verification of large vessel occlusion (LVO) locations in head CT angiograms (CTAs) determined by an NLP algorithm. Our developed pipeline can be applied to other variables that require manual verification for quality control of retrospective datasets. 

Methods: Our initial dataset included 866 patients with 4184 radiology reports; from this we extracted 489 patients with 513 CTA reports, who presented to two tertiary-care institutions with acute large ischemic stroke (>1/2 middle cerebral artery territory) from 2006-2021. We constructed a rule-based system for feature detection using regular expressions and spaCy, a Python NLP integration with advanced sentence parsing and negation to automatically indicate location of LVO. Thereafter, we developed a pipeline for gold-standard verification including manual review of all reports and documentation of feedback with root-cause analysis to understand automatic errors. Secondary review of selected reports by an attending neurologist was also performed, allowing for correction of errors from initial manual review.

Results: We found that the automated NLP algorithm had an accuracy of 78.8% in terms of LVO designation, after initial manual review of CTA reports. Secondary MD review was performed on 60/513 (11.7%) reports with 7/60 (11.7%) reports found to have errors from the initial manual review. After complete review, the frequencies of LVO locations were found to be 38.4% internal carotid artery (ICA)/tandem, 56.9% M1, 36.3% M2, 4.3% M3/M4, 15.0% anterior cerebral artery (ACA), and 0.4% indeterminate location. The most common reason for NLP error included the inability to interpret radiographic findings in the context of surrounding text. For example, the algorithm would misclassify reports with modifiers including “small” or “incomplete,” indicating the lack of complete occlusion.

Conclusion: While tailored rule-based NLP methods can expedite radiology report review for stroke patients, they may err. Therefore, protocolized manual review of radiology reports is necessary, especially for reports with ambiguous language.

Manual Validation of CTA Stroke Data Analyzed by a Natural Language Processing (NLP) Method

Background: Non-convulsive seizures (NCS) are increasingly recognized as a cause of persistent encephalopathy in the emergency department (ED) and are diagnosed using electroencephalography (EEG). However, EEG is often delayed or unavailable, especially in the ED, resulting in significant uncertainty in clinically diagnosing NCS and leading to delayed or inappropriate treatment with anti-seizure medications (ASMs), unnecessary intubation, and increased morbidity. Point-of-care EEG (pocEEG) could fill this critical gap in the ED, as it has in inpatient and intensive care unit (ICU) settings. In this study, we describe the impact of pocEEG on the management of ED patients with suspected NCS.
Methods: We retrospectively identified patients who underwent pocEEG monitoring in the ED at Mission Hospital (1/1/2020-12/31/2020). We extracted pocEEG findings (categorized as seizure, highly epileptiform patterns [HEP], and slow/normal activity; excluding repeat studies for the same patient) and ASM treatment information (pre-hospital, pre-pocEEG, post-pocEEG) to determine whether pocEEG led to appropriate ASM escalation or avoidance of inappropriate ASM escalation. Statistical significance was assessed using chi-square tests with Bonferroni correction.
Results: Of 157 ED patients (mean age 57.7±22.4 years, 49% female), pocEEG revealed seizures in 22 (14%), HEP in 33 (21%), and slow/normal activity in 102 (65%). Most studies (53%) were performed after-hours (5p-8a), and most patients were admitted (54% ICU, 41% floor). Five patients (3%) were treated with ASMs prior to ED arrival, but 93 (59%) were treated in the ED prior to pocEEG monitoring and 80 (51%) were treated after pocEEG monitoring. By reviewing each patient’s ED course and the relationship between pocEEG monitoring and ASM treatment, we found a significant association between pocEEG findings and changes in management (p<0.001). Patients with seizures were significantly more likely to be treated than patients with slow/normal activity (59% vs. 25%, p=0.002) and patients with slow/normal activity were significantly more likely to not have treatment escalation (28% vs. 0% for seizure/HEP patterns, p<0.001).
Conclusion: By guiding bedside diagnosis and treatment decision-making, pocEEG can empower ED physicians and consulting neurologists to evaluate and manage patients with suspected NCS and can positively and substantially impact the care of patients with neurological emergencies.

Point-of-Care EEG Enables Rapid Management of Non-Convulsive Seizures in the Emergency Department

Background
Traumatic brain injury (TBI) is a leading cause of death and disability in children worldwide. Hyperoxia has been suggested as a mechanism of secondary injury following adult TBI, but its effect on pediatric patients has not been well-described. 
Our goal is to determine if hyperoxia, measured by the highest PaO2 and AUC PaO2 in the first 24 hours after TBI, is correlated with the neurological outcome. 

Methods
We conducted a review of all pediatric (18yo) TBI patients in a prospective institutional registry from October 2008 to January 2021 at UC Davis Health in Sacramento, California. PaO2 was extracted from the arterial blood gas report for the first 24 hours after admission to the ICU. The first PaO2, highest PaO2, and the AUC calculation of PaO2 were collected and calculated for each patient. Outcomes 6 months after injury were measured using the modified Rankin Scale (mRS) and dichotomized as favorable (1-2) or unfavorable (3-6). A multivariable logistic regression model was used to determine if minimum PaO2, maximum PaO2, or AUC PaO2, predicted an unfavorable outcome.

Results
We identified 122 pediatric patients with severe TBI during the study period. The incidence of hyperoxia (PaO2 > 300 mmHg) was 25%. Elevated PaO2 values, determined by both maximum and AUC PaO2 within the first 24 hours of hospitalization, was an independent predictor of worse neurological outcome at 6 months. 

Conclusion
Using an institutional database, hyperoxia was an independent predictor of poor neurological outcome in pediatric patients. Further studies are needed to determine if efforts to prevent and/or correct hyperoxia are associated with improved outcomes.


The Association between Hyperoxia after TBI and Degree of Disability in Pediatric Patients

Background: The number of fertility treatments and consequently, the number of births conceived through fertility treatments, has continued to increase in the United States. Treatments, including assisted reproductive technology (ART) and non-ART treatments such as fertility-enhancing drugs, make conceiving a child possible for those who face infertility. However, fertility treatments are associated with a higher rate of multifetal gestations which are known to come with greater adverse outcomes. However, there is more to understand on if multifetal gestations conceived with fertility treatments pose an even greater risk than singleton gestations. In our study, we examined neonatal outcomes among twin gestations conceived through spontaneously, non-ART treatments, or ART.

Methods: We utilized linked vital statistics-patient discharge data from California (2007-2011) and included twins with gestational ages of 23-42 weeks. We categorized mode of conception into three categories- spontaneous or non-ART or ART. We compared demographics and outcomes (NICU admission >24 hours, respiratory distress syndrome, macrosomia, small for gestational age, congenital anomalies, intrauterine fetal demise (IUFD), and infant death) using chi-square tests. Multivariable logistic regression models then assessed the association of mode of conception with the outcomes.

Results: Among 64,500 neonates, 88.0% were spontaneous gestations, 9.0% were ART gestations, and 2.93% were non-ART treatment gestations. Pregnant patients who used fertility enhancing drugs or ART were more likely to be white, older, highly educated, on private insurance and have normal BMI (p<0.001 for all). Multivariable logistic regression showed that pregnancies conceived through fertility enhancing drugs had significantly higher odds of NICU admission [aOR=1.41 (1.28,1.56)] and respiratory distress syndrome [aOR=1.27 (1.15,1.42)]. Similar results were found in pregnancies conceived through ART, but ART also had higher odds of infant death [aOR=1.35 (1.03,1.78)] and congenital anomalies [aOR=1.11 (1.02,1.21)].

Conclusion: Although patients who underwent ART or non-ART treatments were more likely to have demographics that are associated with favorable neonatal outcomes, both methods of non- spontaneous twin gestations were more likely to have adverse neonatal outcomes in NICU admission and respiratory distress syndrome. The higher risk nature of pregnancies conceived through ART and fertility-enhancing drugs may necessitate closer follow ups throughout gestation to avoid adverse neonatal outcomes.

Adverse Neonatal Outcomes of Twin Gestations: Do They Differ By Mode of Conception?

Background: Women ≧35 years old are categorized as “advanced maternal age” (AMA) and are at an increased risk for a decline in ovarian reserve and oocyte competence. Therefore, they may seek fertility assistance through IVF. Pre-implantation genetic testing for aneuploidy (PGT-A) is recommended to AMA women undergoing IVF to determine an embryo’s ploidy status. However, PGT-A findings are currently a point of contention due to mosaicism. This study seeks to identify gene expression patterns found in blastocoel fluid from IVF-embryos from mothers of AMA that resulted in successful pregnancy outcomes.

Methods: Blastocoel fluid-conditioned media was collected from day-5 IVF-embryos that underwent PGT-A biopsy. RNA was purified from the fluid and cDNA was synthesized. Genes identified from a previous RNASeq study were selected for further analysis using RT-qPCR. GAPDH, BCL2L12, CUL2, SHARPIN and DRAP1 gene expression were assessed using RT-qPCR in blastocoel fluid samples from AMA mothers and mothers 35.

Results: Preliminary analysis on 170 blastocoel fluid samples (n= 48 for younger individuals and n=122 individuals of AMA), indicates that BCL2L12, SHARPIN, and DRAP1 are upregulated and expressed to a higher extent in AMA samples.

Conclusion: Initial findings suggest that BCL2L12, SHARPIN, and DRAP1 genes might be associated with AMA. BCL2L12 is an anti-apoptotic gene involved in regulated cell death. SHARPIN stimulates the formation of linear ubiquitin which regulates NF-KB signaling and inhibits extrinsic apoptosis. DRAP1 regulates gene transcription, while CUL2 supports ubiquitin ligase activities within cells. Ongoing studies aim to identify gene expression patterns associated with AMA, which could serve as an additional marker along with PGT-A to aid in selecting the most viable embryo for transfer.


Analysis of apoptotic gene expression in blastocoel fluid of IVF-embryos from individuals of advanced maternal age

Introduction:
In vitro fertilization (IVF) cycles are monitored for premature progesterone rise, which contribute to implantation failure. Progesterone thresholds prompting recommendation for freeze-all cycles vary between institutions, depending on laboratory assay and patient population. To our knowledge, this has not been evaluated at a military treatment facility.

Methods:
A retrospective chart review was performed; all IVF cycles with fresh embryo transfers between 2016 and 2019 at Brooke Army Medical Center (BAMC) were eligible. Serum progesterone levels were obtained via electrochemiluminescence assay on Roche Cobas 80036 hours or less prior to ovulation trigger. The number of clinical pregnancies (CP, n=190) and live births (LB, n=159) were stratified by progesterone levels.

Results:
345 women who underwent 393 fresh embryo transfers were included in the analyses. Progesterone was significantly lower in patients with a CP (0.693ng/mL ± 0.300) versus those without a CP (0.758ng/mL ± 0.438). Patients with a progesterone <1.3ng/mL are significantly more likely to achieve a CP (49.7% v 28%, p=0.035). On linear regression, when controlling for age and parity, a progesterone level ≥1.3ng/mL is significantly negatively associated with CP (p=0.045). When controlling for age and parity, a progesterone level <1.3ng/mL trends toward a significant association with LB (p=0.1)

Conclusion:
A progesterone <1.3ng/mL was established as a significant threshold for CP, and should be considered in future fresh IVF cycles at BAMC. However, there was lack of statistical significance for LB given this progesterone level, possibly impacted by the low number of patients with higher progesterone (n=25), which resulted in only 6 live births.

Determining a Progesterone Threshold for Fresh IVF Cycles in a Military Facility

Posterior reversible encephalopathy syndrome (PRES) is a neurotoxic condition that can present with seizures, headaches, and radiologic findings of vasogenic edema. Eclampsia and pre-eclampsia can predispose patients to developing PRES in the setting of endothelial dysfunction that could be caused by COVID-19, or another vasogenic mechanism. It is not well understood how COVID-19 can affect the brain but there is evidence that it can cause neurological symptoms. COVID-19 has been implicated in the pathogenesis of eclampsia, PRES and subarachnoid hemorrhage (SAH). PRES has been implicated in the pathogenesis of SAH and eclampsia. However, eclampsia is not well connected to the presence of SAH. In our case, a 33-year-old multiparous woman presented to the hospital and was admitted for supervision of labor complicated by COVID-19 infection and anemia. She delivered a healthy infant. She proceeded to have severe range blood pressures after delivery, and experienced multiple tonic-clonic seizures that were controlled with magnesium sulfate, labetalol, and nifedipine. At this time, she had radiological evidence of SAH and PRES. Our patient had a SAH in the context of PRES, eclampsia and a COVID-19 infection. Her SAH could be due to the vasogenic edema from PRES that could have been triggered by endothelial damage from COVID-19, and vasogenic dysfunction from eclamptic seizures.

Eclampsia complicated by subarachnoid hemorrhage in the context of posterior reversible encephalopathy and COVID-19 infection

Background: Ovarian cysts frequently resolve on their own, but little information on this resolution is available for use in clinical practice. The objective of this study was to characterize the resolution of incident ovarian cysts in relation to cyst diameter, structure, age, body habitus, and menopausal status. These categorizations are important for decisions on whether to continue monitoring the cyst or intervene surgically.

Methods: A total of 2,638 women with incident cysts were selected who had undergone 51,356 TVUS examinations in the University of Kentucky Ovarian Cancer Screening Program. The variables that were collected in this study were: age, BMI, cyst diameter, cyst structure, family history of ovarian cancer, and use of hormone replacement therapy. All images were reviewed by a physician. Methods of analysis include t-tests, chi-square, and Cox regression analysis of cyst resolution times.

Results: Women with a history of 1 or more cysts were included so that 2465 unilocular and 1420 septated cysts were available for analysis. Septated cysts (mean + SEM, 3.8+0.03), had a larger diameter than unilocular cysts (3.4+0.03), P<0.05. The resolution of incident septated cysts was faster than unilocular cysts. Unilocular cysts resolved the quickest in women 70+ years of age. Resolution was the most prolonged in women <40 years. No difference in resolution was observed between pre- and post-menopausal status nor BMI in the resolution of unilocular or septated cysts.

Conclusion: Septated cysts were significantly more likely to resolve than unilocular cysts in univariate and multivariate analyses. As the cyst diameter increased, cysts were found to decrease in time to resolution in univariate and multivariate analyses. With HRT use, cysts were significantly less likely to resolve in univariate analyses but were not significantly affected in multivariate analysis. As the patients aged, it was determined they were more likely to have cyst resolution in univariate and multivariate analyses. Menopausal status and body mass index were not seen to have an influence on the time to resolution of unilocular and septated cysts. For the clinician, this study establishes expectations for cyst resolution based on a primary estimation of cyst diameter in a population where no malignancy was observed.

Findings on ovarian cyst resolution for the practicing physician based on cyst diameter

Abstract Title: Hypoplastic Left Heart Syndrome: Prevalence of Anatomic Subtypes and Coexisting Genetic and Noncardiac Anatomic Anomalies Among Neonates at the University of Florida

Background
Hypoplastic left heart syndrome (HLHS) is a form of congenital heart disease (CHD) characterized by the underdevelopment of the left ventricle. It affects 2-3 of 10,000 live births and accounts for 2.7% of CHD. The severity of HLHS depends on the degree of left ventricular inlet and/or outlet obstruction arising from patterns of structural malformations including stenosis and atresia of the mitral and aortic valves. The objective of this study is to determine the prevalence of anatomic subtypes of HLHS and to assess fetal co-morbidities among fetuses with HLHS at the University of Florida (UF). 

Methods
Retrospective chart review of neonates born with HLHS at UF between 1/1/2006 and 12/31/2019. Data pertinent to maternal history, fetal workups, prenatal sonography, labor and delivery, neonatal hospital course, and pre- and postnatal echocardiograms were collected. Descriptive statistics were used.

Results
A total of 51 cases were included. Of all 51 cases, there were various combinations of valvular abnormalities as follows: mitral atresia and aortic atresia (MA/AA) in 19/51 (37.3%), mitral stenosis and aortic atresia (MS/AA) in 14/51 (27.4%), mitral stenosis and aortic stenosis (MS/AS) in 5/51 (9.8%), and variant subtypes in 13/51 (25.5%) cases. Noncardiac congenital anatomic abnormalities, fetal growth abnormalities, and genetic abnormalities were present in 11/51 (21.6%), 2/51 (3.9%), and 4/51 (7.8%), respectively. Genetic testing was performed in a majority of patients 49/51 (96.1%) with prenatal genetic testing performed in 20/51 (39.2%) patients and postnatal genetic testing in 46/51 (90.2%). Genetic testing revealed 4/49 (8.2%) identifiable genetic abnormalities and 11/49 (22.4%) variances of unknown significance. 

Conclusion
The most common structural subtype of HLHS was MA/AA. Approximately 1/5 of the fetuses had major extracardiac congenital anatomic abnormalities. Genetic anomalies were present in 8.2% of the fetuses although only 39.2% received prenatal genetic testing. This highlights the importance of offering prenatal genetic testing to pregnancies in which HLHS is suspected. A thorough prenatal evaluation including assessment of anatomic subtype and identification of anatomic and genetic comorbidities is essential because it enables extensive patient counseling and ensures the involvement of a multidisciplinary team to plan perinatal management for optimal neonatal outcome.

Hypoplastic Left Heart Syndrome: Prevalence of Anatomic Subtypes and Coexisting Genetic and Noncardiac Anatomic Anomalies Among Neonates at the University of Florida

Premium content

Next from AMA Research Challenge 2022

Demographic Categorization of Subjects within the Alzheimer’s Disease Neuroimaging Initiative Database

Similar lecture

Edge-guided CNN for Denoising Images from Portable Ultrasound Devices