United States

AMA Research Challenge 2022

Assessing Time in Therapeutic Range (TTR): Is remote anticoagulation management as effective as in-person care?

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

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-26 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to AAAI 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://aaai.org/conference/aaai/aaai-26/registration/) first.

**Online Registration Form**: https://aaai.getregistered.net/conference-2026 

Registration Required

AAAI 2026

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

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.


Abstract Title: Catheter ablation for atrial fibrillation: a national study of demographics & socioeconomic factors.

Background: Atrial fibrillation(AF) is the most commonly diagnosed cardiac arrhythmia. With a worldwide prevalence of >33 million, it poses a significant economic burden and serious cardiovascular morbidity and mortality. Catheter ablation(CA) has proven to be an important curative treatment for paroxysmal, persistent, and permanent AF.

Methods: This retrospective study acquired data from the NIS between January,2007 and December,2017. Primary and secondary diagnosis of non-valvular atrial fibrillation were identified using ICD-9 and 10-CM codes. Chi square and unpaired t test were used to evaluate socio-demographic disparities.

Results: From 35,906,946 patients with NVAF, 343641(0.96%) underwent catheter ablation. Patients who underwent CA were mostly in the age group of 50-75 years(63.67%) vs <50 years(8.63%) vs >75 years(27.71%). The majority of patients undergoing CA were men(61.6%) and was more common amongst White race(86.17%) vs Native American(0.58%) vs Black(7.31%). Patients with a median household income falling in 76th-100th percentile(28%) were the majority to have the procedure. Most of the patients had medicare insurance(59.5%) vs Medicaid(4.06%) vs private insurance(32.71%). CA was performed more often during weekdays(89.54%) vs weekends(10.46%) and frequently done as a non-elective procedure(51.41%) vs elective(48.59%). It was more frequent in urban teaching hospitals(73.14%) vs urban non teaching(24.1%) and was more common in the South region(39.57%) vs West region(19.33%). Co-morbidity comparison between the cohorts with vs without CA include hypertension(67.4% vs 75.52%), diabetes(24.53% vs 33.8%), CHF(33.4%
vs 42.45%), alcohol use(2.12% vs 3.17%), tobacco use(7.38% vs 7.72%), and drug abuse(1.02% vs 1.37%). p value was noted to be <0.0001 for above comparisons. In regression analysis, age>50 years(OR-1.59,1.52-1.66,P<0.0001), native-Americans (OR1.39,1.24-1.55,p<0.0001), patients with private insurance(OR-1.307,1.28-1.33,p<0.0001) and high income group(OR-1.154,1.12-1.18,p<0.0001) were associated with higher odds of utilization of catheter ablation.

Conclusion: There is a significant racial and sex disparity in patients undergoing CA for NVAF that mandates further investigation. These results are important when interpreting the characteristics of the patients, so future trials can be made to decrease the high burden of NVAF through CA. Study’s limitation is insufficient data on disease severity and cardiovascular
diagnostic test variables for accurately predicting the factors associated with under utilization.

Catheter ablation for atrial fibrillation: a nationwide study of demographics and socioeconomic factors.

Clinical Characteristics and Outcome of Patients with Left Ventricular Thrombus After ST-segment Elevation Myocardial Infarction: A Single-Center Retrospective Study

1Ngoda Manongi, MD MS., 2Emmanuel Moustakakis, MD., 2Sijun Kim, DO. 
1 Department of Internal Medicine, New York Presbyterian Queens
2 Division of Cardiology, New York Presbyterian Queens

Abstract

Background: Left ventricular thrombus (LVT) can occur in the setting of an ST-elevation myocardial infarction (STEMI). This is a single-center retrospective study to evaluate the clinical characteristics and outcome of patients who develop LVT as a complication of STEMI to help identify at-risk patients for early diagnosis and treatment.

Methods: In this retrospective single center study, we identified patients diagnosed with STEMI admitted to NewYork-Presbyterian Queens between December 2015 and December 2020 using electronic medical record. Transthoracic echocardiogram (TTE) was used to assess the presence of LV thrombus. Demographic data, choice of anticoagulation (AC), use of concomitant antiplatelet agents were collected and reviewed. Clinical outcomes following initiation of AC included bleeding events, stroke, systemic embolism, echocardiographic resolution of LV thrombus, and all-cause mortality. 
Results: Among the 545 patients admitted with STEMI, 14 patients (2.6%) had LVT. The left anterior descending artery was the culprit vessel in all of these cases. The median time from admission to diagnosis using echocardiography was 48 hours (range: 3-74) and the mean LV ejection fraction was 34% (range: 20-45%). Ultrasound enhancing agent was used in all patients. Ten patients (71%) had symptom onset more than 24 hours prior to presentation. Therapeutic anticoagulation was started with heparin and bridged to VKA in 10 patients (71%) and DOAC in 3 patients (21%). One patient did not receive anticoagulation due to bleeding concerns. One patient suffered a stroke shortly after initiating DOAC. Major bleeding event occurred in 1 patient. Follow-up echo was available at six months in 7 patients and showed resolution of LVT in all patients with a mean LV ejection fraction of 42% (range: 25-60%). The median length of hospital stay was 7 days (range 4-21 days). 
Conclusions: The incidence of LVT following STEMI between December 2015 and December 2020 was 2.6%. A high index of suspicion for LVT is needed in patients who have a delayed presentation of STEMI and involve the left anterior descending artery. TTE using ultrasound enhancing agent is important to accurately make the diagnosis.

Clinical Characteristics and Outcome of Patients with Left Ventricular Thrombus After ST-segment Elevation Myocardial Infarction: A Single-Center Retrospective Study

Background: Asymptomatic aortic stenosis management is not only a daily practice challenge but also an area of clinical research controversy. As evidence on conservative versus early aortic valve replacement (AVR) is debatable, we conducted a systematic review and meta-analysis to compare the efficacy of conservative versus AVR in asymptomatic aortic stenosis (AS).

Methods: Electronic databases were systematically searched to identify any observational or interventional studies that compared the effects of conservative management versus AVR on all causes, cardiovascular, non-cardiovascular mortality. Pooled risk ratios (RR) and their 95% confidence intervals (CI) were calculated using random-effects models.

Results: A total of 6216 patients were included from 12 studies (observational=10, and randomized clinical trial=2) with a mean follow-up duration of 5 years. When compared to early AVR, conservative management has statistically significant higher risk of all cause (RR: 0.36, 95%CI [0.32 to 0.46], p<0.00001), cardiovascular (RR: 0.32, 95%CI [0.24 to 0.41], p<0.00001), and non-cardiovascular mortality with (RR: 0.30, 95%CI [0.12 to 0.73], p=0.008). Regarding other complications, conservative treatment showed significant higher risk in sudden cardiac death (RR: 0.46, 95%CI [0.30-0.71], p=0.0005), yet no difference over AVR in stroke, myocardial infarction (RR: 0.86, 95% CI [0.35 to 2.13], p=0.75) and (RR: 0.54, 95% CI [0.19 to 1.53], p=0.25), respectively. 

Conclusion: Patients with asymptomatic aortic valve stenosis, managed with early AVR, have a lower risk of all-cause, cardiovascular, and non-cardiovascular mortality when compared to conservative management.

Conservative management versus early aortic valve replacement for treatment of asymptomatic aortic stenosis: a systematic review & meta-analysis

Background: Atrial fibrillation (AF) is a common postoperative complication after 
coronary artery bypass surgery (CABG). Studies have shown a higher incidence 
of postoperative AF in White patients than in Black patients. This retrospective 
cross-sectional study aims to characterize the incidence and risk factors for 
postoperative AF in Middle Eastern and Black patients compared to White 
patients. 

Methods: Adult patients were included if they had undergone CABG between 
January 2014 and December 2021. Baseline demographics and perioperative 
data were collected. Patients’ race designations were based on self-identification 
in the electronic health records. Univariate and multivariable analyses were 
conducted to identify predictors of postoperative AF. 

Results: A total of 1686 patients were included (median age 67 years, 27% were 
females). There was a total of 1169 White, 440 Black, and 77 Middle Eastern patients. 
The overall incidence of postoperative AF was 40%. We observed a lower incidence 
of postoperative AF in Middle Eastern (32%, OR 0.59, 95% CI 0.36-0.97, P = 0.034) 
and Black patients (28%, OR 0.47, 95% CI 0.37-0.60, P = <0.001) as compared to 
White patients (45%). Other variables associated with postoperative AF on the 
univariate analysis included female gender, pre-operative AF, age, chronic kidney 
disease, left atrial size, obstructive sleep apnea, pre-operative AV nodal blockade, 
concomitant valvular surgery, and moderative-severe valvular disease. On 
multivariable analysis, the Middle Eastern race was associated with a lower incidence 
of postoperative AF with a trend towards significance (OR 0.60 95% CI 0.34-1.07, P 
0.081), whereas the Black race continued to be significantly associated with a lower 
incidence of postoperative AF (OR 0.59 95% CI 0.44-0.79, P <0.001). Other variables 
that continued to be significantly associated with postoperative AF included age ≥65 
years (OR 1.44 P 0.007), female gender (OR 0.63 P 0.002), left atrial size index >32 
ml/kg (OR 1.64 P 0.005), moderate-severe aortic stenosis (OR 1.88 P 0.005), 
preoperative AF (OR 4.55 P 0.005), and AV nodal blockade (OR 0.632 P 0.019). 

Conclusion: We report a lower prevalence of postoperative AF in Middle Eastern 
and Black patients. Further studies are needed to investigate the underlying 
associations between racial and ethnic categories and AF post CABG.

Effect of Race on the incidence of atrial fibrillation post-coronaryartery bypass surgery in a Middle Eastern Community in the United States

Introduction: Heart failure is a growing public health problem with an estimated prevalence of 4-5 million American population. Despite significant advances in heart failure management, the mortality remains high. This study aspires to investigate the risk factors that are associated with mortality in hospitalized heart failure patients.
Methods: We utilized the 2019 National Inpatient Sample (NIS) dataset sponsored by the Agency for Healthcare Research and Quality (AHRQ) under the Healthcare Cost and Utilization Project (HCUP). Patients with Heart Failure (HF) were identified using Clinical Classifications Software Refined (CCSR) for International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). Variable of interest include hypertension, diabetes, dyslipidemia, peripheral vascular disease, nicotine use and sleep apnea. Clinical outcomes include HF hospitalizations and effect of different variable on HF mortality. 
Results: A weighted total of 5,675,784 HF related hospitalizations were recorded in the year 2019. A total of 3907455(68.9%) had hypertension, 2080710(36%) had diabetes and 3081445 (54%) had dyslipidemia . 651035(11%) had a comorbidity of peripheral arterial disease, 1542990 (27.1%) were nicotine dependent, and 991150 (17.4%) had sleep apnea. The mortality rate per 10,000 HF hospitalizations were analyzed. There were 7047 deaths due to hypertension as compared to 2953 in non-hypertensive population, 2368 deaths due to diabetes as compared to 2122 deaths in patients without diabetes. The Mortality rate was 4502 in patients with dyslipidemia as compared to 4234 in patients without dyslipidemia. Peripheral arterial disease (PAD) lead to 2434 death as compared to 2183 death in patients without PAD. The mortality rate 4649 deaths due to nicotine use as compared to 4278 in patients who do not use nicotine and 4580 was due to sleep apnea as compared to 4336 death in patients with no sleep apnea. 
Conclusion: Heart failure lead to significant number of hospitalization in the United States in year 2019. Among all studied factors, hypertension was found to be associated with highest mortality risk. This study guides the importance of strict blood pressure control in significantly decreasing mortality associated with heart failure. 





Heart Failure World: Risk factors for heart failure mortality in hospitalized patients

Impact of the type of Heart failure and Social determinants of health on Health Related Quality of Life (HRQOL) among patients in the South Bronx

BACKGROUND: Coronary microvascular dysfunction (CMD) and left atrial (LA) strain abnormality are more common in patients with Heart Failure with Preserved Ejection Fraction (HFpEF), but the relationship between CMD and LA strain is not well characterized. HFpEF is a major health burden in the United States, representing the number one cause of hospitalization for Americans over 65
years old. (1) Understanding how coronary microvascular dysfunction impacts HFpEF is important because CMD may be an important prognostic sign of deteriorating heart function, particularly in the HFpEF subset of heart failure. Despite CMD’s prognostic potential, it is currently difficult to measure; only invasive techniques can measure CMD.

OBJECTIVE: We undertook a study to see if non-invasive measurements of LA strain could be correlated with invasive measurements of CMD. We additionally looked at differences between HFpEF and control patients to see if LASr and CMD correlations were generalizable across disease states.

METHODS: We performed a retrospective, observational study of 90 patients at the University of Chicago Medical Center who had an indication for coronary angiography with coronary physiology testing from 2010-2021. CMD was defined as having an invasively-determined coronary flow reserve (CFR) ≤ 2.0. Non-invasive echocardiographic auto-strain analysis was used to obtain LA reservoir strain (LASr), LA conduit strain, and LA contractile strain. Univariable and multivariable logistic
regression was performed for HFpEF patients with various degrees of CMD.

RESULTS: For both HFpEF and control patients, lower LASr was more commonly found in patients with abnormal CFR (HFpEF: 22.5% ± 12.1 vs. 30.7% ± 10.7, P = 0.018; control: 26.0% ± 16.4 vs. 31.0% ± 9.8 , P = 0.185). Both a univariable and multivariable logistic regression of HFpEF patients found LASr to be a significant predictor of abnormal CFR (OR: 0.94, P =0.041).

CONCLUSIONS: There were substantially more patients with CFR ≤ 2.0 in the HFpEF group (50%) compared to control (28%). In HFpEF patients, non-invasively measured LASr showed significant correlation with invasively measured CFR, thus emerging as a potential predictive variable for CFR ≤ 2.0 in patients with this form of heart failure.

Left Atrial Strain Echocardiography May Be Correlated with Coronary Microvascular Dysfunction

Background: Accumulating evidence suggests hospitalized patients with heart failure (HF) are especially susceptible to COVID-19-related complications and mortality approaches 30%. We hypothesized that Boosted Tree (BT) mortality risk variance explained (VE%) by cardiac injury and putative COVID-19 thrombo-inflammatory biomarkers would be different between survivors and non-survivors in HF patients undergoing index hospitalization for SARS-CoV-2.

Methods: Presentation demographics, anthropometrics, laboratory results (within 8h), and ICD-10 codes were collected. Discrete data summarized as proportions were compared with chi-squared test. Continuous data summarized with median [IQR] were compared using Kruskal-Wallis test. COVID-19 therapy across 6-pandemic surges, demographics, and comorbidities were statistically balanced. Univariate logistic regression tested association (p<.05) of COVID-19 clinical traits putatively linked with mortality risk and provided a receiver operating characteristic (ROC) curve while computing Youden’s cut point. BT algorithm optimized area under ROC curve (AUROC) representing prediction accuracy while providing VE% contributed by each cut point variable.

Results: Among 5,841 consecutive patients discharged from index COVID-19 hospitalization between March 18, 2020, and April 30, 2022, 1039 harbored extant HF. Those with at least 1 high sensitivity troponin (hs-cTnI) assay, 483, were studied. Hospital mortality (n=62, 13%) and survival (n=421, 87%) patient demographics were statically similar with 63% males and 37% females aged 75[64-83] years distributed across Whites (82%), Blacks (11%), and other races (7%). Subsequent results are similarly sequenced. Most prevalent comorbidities were hypertension (81% vs. 92%, p=.007), diabetes (48% vs. 42%, p=.41), obesity (50% vs. 34%, p<.02), renal failure (34% vs. 38%, p=.58), iron deficiency anemia (35% vs. 35%, p=1.0), and chronic coagulation disorder (39% vs 17%, p=.0002). Biomarker cut points were hs-cTnI ≥26 ng/L (71% vs. 53%, p=.006), CRP ≥6.0 mg/dL (63% vs. 22%, p<.0001), d-dimer ≥2.98 ug/mL (45% vs. 26%, p=.015), LDH ≥444 U/L (61% vs. 20%, p<.0001), and ferritin ≥668 ng/mL (68% vs. 26%, p<.0001). BT model demonstrated 38% R2, AUROC=0.86 with VE% contributed by hs-cTnI (41%), CRP (27%), d-dimer (14%), LDH (10%), and ferritin (8%). 

Conclusion: Our training model demonstrates a machine learning strategy to integrate hs-cTnI with putative biomarkers of COVID-19 severity to enhance understanding of host response contributing to mortality risk at presentation in hospitalized HF patients.

Mortality Risk Variance Explained by High Sensitivity Troponin I Exceeds Putative COVID-19 Severity Biomarkers in Heart Failure Patients Hospitalized with Acute SARS-CoV-2 Infection

Cardiac catheter ablation is a useful tool in the treatment of Atrial Fibrillation (AFib). A subset of patients with AFib may also present with low Ejection Fraction (EF). This study seeks to quantify and analyze the impact of ablation on the EF of AFib patients who presented with a pre-operative EF below 50%. Patient data was retrieved from a private practice group, Florida Cardiology. Patients who underwent an ablation from Jan 2019 through Nov 2021, had a pre-operative EF less than 50%, and had a post-operative EF on record were included in this study. Diagnoses of Heart Failure as well as status of AFib recurrence at follow-up were also noted. Following data collection and statistical analysis, cardiac catheter ablation was found to lead to a statistically significant improvement in the EF of AFib patients that presented with low pre-operative EF < 50%. The average change was +13.15% (Std Dev of 10.11%). Ablation also led to resolution of AFib symptoms in 88.24% of patients. Ablation remains an effective and successful treatment modality for the management of Atrial Fibrillation in patients with a decreased EF (< 50%).

Not Too Late To Ablate: Effect of Cardiac Catheter Ablation on Ejection Fraction in Patients with Atrial Fibrillation and Pre-Op Ejection Fraction under 50%

Hypertension is one of the most prevalent chronic diseases in the United States, affecting approximately 1 in every 2 adults. Barriers to control blood pressures in the outpatient setting from a patient’s perspective include lack of understanding of disease, fear of medication adverse effects or dependence, medication affordability, healthy food choices and time required to follow up with medical providers. In HTN, there has been limited work investigating the role of how blood pressure (BP) monitors affect patients taking an active role in disease monitoring, management and risk modification. In this project, we hypothesize that patients will actively use provided BP monitors and will feel more empowered for controlling their hypertension. This in turn will improve medication adherence and positively reinforce patients to make lifestyle modifications. This study was conducted in the Internal Medicine Clinic at Prisma Health, Greenville, SC. 34 patients with a diagnosis of uncontrolled hypertension were enrolled. Patients were provided brachial cuff BP monitors and followed for 4 months while receiving medical therapy. We systemically contacted patients via phone interviews to determine their adherence (using adherence estimator survey), patients’ perspective on the BP monitors and how to use them. These phone interviews were coded independently by two members of the research team and cross-checked for confirmation, reconciling differences by discussion. Codes were grouped according to overarching themes and subthemes. For mixed analysis, the qualitative and quantitative data was merged after it had been independently analyzed to provide a mixed methods data analysis. Our results suggest that when BP monitors are provided to patients, and patients are educated on proper use of them, they are effective tools to help patients take an active role in managing their disease. Of the 34 patients, 23 patients responded to phone interviews. Assessment of the patients’ blood pressure (BP) was done in the clinic at the beginning of the study and then again at the conclusion of the study. Using a paired t-test, Systolic and diastolic blood pressure showed significant difference between the two time points dropping from a mean of 147.2/864 ± 23.3/16.4 mmHg to 133.7/78.7 ± 15.9/11.2 mmHg (± STD, p<0.005). Of the 23 patients, seventeen (77%) patients had high adherence, three (13%) patients had medium and two (9%) had low adherence according to the adherence estimator used. The low adherence was due to lack of conviction of the importance of blood pressure medication to their health. When questioned further about their opinions on home BP monitors, 3 themes arose from talking to patients: helpful, not-helpful, and somewhat-helpful. The subthemes identified within “helpful” category were Reinforcement, Lifestyle modification, and medical care. These themes were reflected in how often participants used their BP monitors. 16 (70%) patients demonstrated regular usage of BP monitors, while 4 (17%) patients demonstrated no usage whatsoever, and 3 (13%) patients demonstrated intermittent usage which matches the thematic analysis that was apparent in their responses of helpful, not-helpful, and somewhat-helpful, respectively (p<0.005). This study explores a unique perspective of hypertension management in the outpatient setting by providing patients with a tangible way to monitor their blood pressures that in turn augmented adherence. By empowering patients to understand the effects of lifestyle modifications and medications on their blood pressures on a regular basis via providing a BP monitor, there was improvement in the BP control. Although a small sample size, this study substantiates the important role the BP monitors can play in empowering patients for their HTN management. By positively impacting patients’ engagement in disease monitoring and risk modification, BP monitor provision has the potential to improve HTN control.

Premium content

Next from AMA Research Challenge 2022

Catheter ablation for atrial fibrillation: a nationwide study of demographics and socioeconomic factors.

Similar lecture

Think Then Rewrite: Reasoning Enhanced Query Rewriting for Domain Specific Retrieval