Aaron Boussina
Graduate Trainee, University of California, San Diego
7 active projects
wearables_analysis
Scientific Questions Being Studied
We'd like to study what features (including social determinants of health and wearables data) contribute to readmission in patients with acute kidney injury (AKI) and sepsis. Understanding these features is important since the ability to proactively identify readmission risk could lead to improved consultation & follow-up for high-risk patients.
Project Purpose(s)
- Disease Focused Research (Acute Kidney Injury & Sepsis)
- Educational
Scientific Approaches
We'll be analyzing subsequent visits and looking at a range of features including vital signs, laboratory measurements, and survey responses. We anticipate we'll be using the PERSON, VISIT_OCCURRENCE, MEASUREMENT, and OBSERVATION tables. From these features we'll apply logistic regression, multi-scale entropy, and deep learning to build a predictive model for readmission risk. We'll identify the corresponding importance of features using odd-ratios or SHAP values.
Anticipated Findings
We anticipate that we'll be able to achieve a performant predictive model for the readmission of patients with AKI and sepsis. We hope this work will result in novel findings and publication.
Demographic Categories of Interest
This study will not center on underrepresented populations.
Data Set Used
Registered TierResearch Team
Owner:
- Arshia Nayebnazar - Other, University of California, San Diego
- Aaron Boussina - Graduate Trainee, University of California, San Diego
Collaborators:
- Hayden Pour - Project Personnel, University of California, San Diego
- Fatemeh Amrollahi - Graduate Trainee, University of California, San Diego
MED264_AKI_Readmission
Scientific Questions Being Studied
We'd like to study what features (including social determinants of health) contribute to readmission in patients with acute kidney injury (AKI). Understanding these features is important since the ability to proactively identify readmission risk could lead to improved consultation & follow-up for high-risk patients.
Project Purpose(s)
- Disease Focused Research (Acute Kidney Injury)
- Educational
Scientific Approaches
We'll be analyzing subsequent visits and looking at a range of features including vital signs, laboratory measurements, and survey responses. We anticipate we'll be using the PERSON, VISIT_OCCURRENCE, MEASUREMENT, and OBSERVATION tables. From these features we'll apply logistic regression and deep learning to build a predictive model for readmission risk. We'll identify the corresponding importance of features using odd-ratios or SHAP values.
Anticipated Findings
We anticipate that we'll be able to achieve a performant predictive model for the readmission of patients with AKI. To our knowledge this is the first such study conducted on All of Us.
Demographic Categories of Interest
This study will not center on underrepresented populations.
Data Set Used
Registered TierResearch Team
Owner:
- Haben Yhdego - Research Fellow, University of California, San Diego
- Arshia Nayebnazar - Other, University of California, San Diego
- Aaron Boussina - Graduate Trainee, University of California, San Diego
Collaborators:
- Fatemeh Amrollahi - Graduate Trainee, University of California, San Diego
- Shamim Nemati - Early Career Tenure-track Researcher, University of California, San Diego
V6 - MED264_AKI_Readmission_Group
Scientific Questions Being Studied
We'd like to study what features (including social determinants of health) contribute to readmission in patients with acute kidney injury (AKI). Understanding these features is important since the ability to proactively identify readmission risk could lead to improved consultation & follow-up for high-risk patients.
Project Purpose(s)
- Disease Focused Research (Acute Kidney Injury)
- Educational
Scientific Approaches
We'll be analyzing subsequent visits and looking at a range of features including vital signs, laboratory measurements, and survey responses. We anticipate we'll be using the PERSON, VISIT_OCCURRENCE, MEASUREMENT, and OBSERVATION tables. From these features we'll apply logistic regression and deep learning to build a predictive model for readmission risk. We'll identify the corresponding importance of features using odd-ratios or SHAP values.
Anticipated Findings
We anticipate that we'll be able to achieve a performant predictive model for the readmission of patients with AKI. To our knowledge this is the first such study conducted on All of Us.
Demographic Categories of Interest
This study will not center on underrepresented populations.
Data Set Used
Registered TierResearch Team
Owner:
- Haben Yhdego - Research Fellow, University of California, San Diego
- Arshia Nayebnazar - Other, University of California, San Diego
- Aaron Boussina - Graduate Trainee, University of California, San Diego
Collaborators:
- Fatemeh Amrollahi - Graduate Trainee, University of California, San Diego
MED299
Scientific Questions Being Studied
We intend to use this dataset in MED299 at UC San Diego in order to teach students about cloud computing and informatics.
Project Purpose(s)
- Disease Focused Research (Sepsis )
- Educational
Scientific Approaches
Sepsis is one of the country’s most urgent systemic health threats. Each year, more than 1.5 million people in the U.S. get sepsis. Sepsis kills a quarter-million Americans each year. Sepsis can occur when a trauma or an infection – often caused by a superbug or drug-resistant bacteria in the skin, lungs or urinary tract – triggers a chain reaction throughout the body. Without timely recognition and treatment, it can rapidly cause tissue damage, organ failure, and death; each hour treatment of sepsis is delayed mortality risk increases by an additional 4%. One in three patients who die in a hospital have sepsis. We intend to use regression analysis to study the factors that contribute to sepsis readmission.
Anticipated Findings
Successful prevention and management of sepsis, septic shock, and organ injury rely on the ability of clinicians to anticipate and estimate the risk, and administer the right life-saving treatments (e.g., antibiotics, fluids and vasopressors) at the right time.
Demographic Categories of Interest
This study will not center on underrepresented populations.
Data Set Used
Registered TierResearch Team
Owner:
- Zaid Yousif - Research Fellow, University of California, San Diego
- Supreeth Shashikumar - Research Fellow, University of California, San Diego
- Shamim Nemati - Early Career Tenure-track Researcher, University of California, San Diego
- Fatemeh Amrollahi - Graduate Trainee, University of California, San Diego
- Aaron Boussina - Graduate Trainee, University of California, San Diego
Collaborators:
- Arshia Nayebnazar - Other, University of California, San Diego
MED299-Spring2022
Scientific Questions Being Studied
The goal is this research is to study the effect of social determinants of health on sepsis related mortality.
Project Purpose(s)
- Methods Development
Scientific Approaches
Social determinants of health will be extracted from the survey questions among hospitalized patients.
Anticipated Findings
Findings from the study may help uncover social determinants of health associated with sepsis related mortality.
Demographic Categories of Interest
This study will not center on underrepresented populations.
Data Set Used
Registered TierResearch Team
Owner:
- Supreeth Shashikumar - Research Fellow, University of California, San Diego
- Shamim Nemati - Early Career Tenure-track Researcher, University of California, San Diego
- Fatemeh Amrollahi - Graduate Trainee, University of California, San Diego
- Aaron Boussina - Graduate Trainee, University of California, San Diego
Collaborators:
- Rahul M Patil - Graduate Trainee, University of California, San Diego
- Nancy Yuan - Graduate Trainee, University of California, San Diego
- Haben Yhdego - Research Fellow, University of California, San Diego
- Arshia Nayebnazar - Other, University of California, San Diego
- Archil Srivastava - Graduate Trainee, University of California, San Diego
MED264_AKI_Readmission_Group
Scientific Questions Being Studied
We'd like to study what features (including social determinants of health) contribute to readmission in patients with acute kidney injury (AKI). Understanding these features is important since the ability to proactively identify readmission risk could lead to improved consultation & follow-up for high-risk patients.
Project Purpose(s)
- Disease Focused Research (Acute Kidney Injury)
- Educational
Scientific Approaches
We'll be analyzing subsequent visits and looking at a range of features including vital signs, laboratory measurements, and survey responses. We anticipate we'll be using the PERSON, VISIT_OCCURRENCE, MEASUREMENT, and OBSERVATION tables. From these features we'll apply logistic regression and deep learning to build a predictive model for readmission risk. We'll identify the corresponding importance of features using odd-ratios or SHAP values.
Anticipated Findings
We anticipate that we'll be able to achieve a performant predictive model for the readmission of patients with AKI. To our knowledge this is the first such study conducted on All of Us.
Demographic Categories of Interest
This study will not center on underrepresented populations.
Data Set Used
Registered TierResearch Team
Owner:
- Aaron Boussina - Graduate Trainee, University of California, San Diego
Collaborators:
- Fatemeh Amrollahi - Graduate Trainee, University of California, San Diego
- Arshia Nayebnazar - Other, University of California, San Diego
MED264_AKI_Readmission
Scientific Questions Being Studied
We'd like to study what features (including social determinants of health) contribute to readmission in patients with acute kidney injury (AKI). Understanding these features is important since the ability to proactively identify readmission risk could lead to improved consultation & follow-up for high-risk patients.
Project Purpose(s)
- Disease Focused Research (Acute Kidney Injury)
- Educational
Scientific Approaches
We'll be analyzing subsequent visits and looking at a range of features including vital signs, laboratory measurements, and survey responses. We anticipate we'll be using the PERSON, VISIT_OCCURRENCE, MEASUREMENT, and OBSERVATION tables. From these features we'll apply logistic regression and deep learning to build a predictive model for readmission risk. We'll identify the corresponding importance of features using odd-ratios or SHAP values.
Anticipated Findings
We anticipate that we'll be able to achieve a performant predictive model for the readmission of patients with AKI. To our knowledge this is the first such study conducted on All of Us.
Demographic Categories of Interest
This study will not center on underrepresented populations.
Data Set Used
Registered TierResearch Team
Owner:
- Aaron Boussina - Graduate Trainee, University of California, San Diego
Collaborators:
- Fatemeh Amrollahi - Graduate Trainee, University of California, San Diego
Request a Review of this Research Project
You can request that the All of Us Resource Access Board (RAB) review a research purpose description if you have concerns that this research project may stigmatize All of Us participants or violate the Data User Code of Conduct in some other way. To request a review, you must fill in a form, which you can access by selecting ‘request a review’ below.
