Dipankar Gupta

Dipankar Gupta

Auto-generated profile, based on publicly available information

Dipankar Gupta

Congenital Heart Center, Department of Pediatrics, University of Florida College of Medicine, Gainesville, USA

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2021 in JAMIA Open
Verified
Journal Article
Open access
PDF file
MKMichael O Killian
Michael O Killian
SPSeyedeh Neelufar Payrovnaziri
Seyedeh Neelufar Payrovnaziri
DGDipankar Gupta
Dipankar Gupta
DDDev Desai
Dev Desai
ZHZhe He
Zhe He
Objectives: Prediction of post-transplant health outcomes and identification of key factors remain important issues for pediatric transplant teams and researchers. Outcomes research has generally relied on general linear modeling or similar techniques offering limited predictive validity. Thus far, data-driven modeling and machine learning (ML) approaches have had limited application and success in pediatric transplant outcomes research. The purpose of the current study was to examine ML models predicting post-transplant hospitalization in a sample of pediatric kidney, liver, and heart transplant recipients from a large solid organ transplant program. Materials and Methods: Various logistic regression, naive Bayes, support vector machine, and deep learning (DL) methods were used to predict 1-, 3-, and 5-year post-transplant hospitalization using patient and administrative data from a large pediatric organ transplant center. Results: DL models did not outperform traditional ML models across organ types and prediction windows with area under the receiver operating characteristic curve values ranging from 0.50 to 0.593. Shapley additive explanations (SHAP) were used to increase the interpretability of DL model results. Various medical, patient, and social variables were identified as salient predictors across organ types. Discussion: Results showed that deep learning models did not yield superior performance in comparison to models using traditional machine learning methods. However, the potential utility of deep learning modeling for health outcome prediction with pediatric patients in the presence of large number of samples warrants further examination. Conclusion: Results point to DL models as potentially useful tools in decision-support systems assisting physicians and transplant teams in identifying patients at a greater risk for poor post-transplant outcomes.

1 December 2021 in Pharmacological Research
Verified
Journal Article
Closed access
DGDipankar Gupta
Dipankar Gupta
AKAshok Kumar
Ashok Kumar
AMAvinash Mandloi
Avinash Mandloi
VSVinayak Shenoy
Vinayak Shenoy

27 May 2022 in The Journal of Heart and Lung Transplantation
Verified
Journal Article
Closed access
DGDipankar Gupta
Dipankar Gupta
NBNeha Bansal
Neha Bansal
BJByron C. Jaeger
Byron C. Jaeger
RCRyan C. Cantor
Ryan C. Cantor
DKDevin Koehl
Devin Koehl
DSDavid L. Sutcliffe
David L. Sutcliffe

20 June 2023 in JMIR Cardio
Verified
Journal Article
Open access
MKMichael O Killian
Michael O Killian
STShubo Tian
Shubo Tian
AXAiwen Xing
Aiwen Xing
DHDana Hughes
Dana Hughes
DGDipankar Gupta
Dipankar Gupta
ZHZhe He
Zhe He
Background: The prediction of posttransplant health outcomes for pediatric heart transplantation is critical for risk stratification and high-quality posttransplant care. Objective: The purpose of this study was to examine the use of machine learning (ML) models to predict rejection and mortality for pediatric heart transplant recipients. Methods: Various ML models were used to predict rejection and mortality at 1, 3, and 5 years after transplantation in pediatric heart transplant recipients using United Network for Organ Sharing data from 1987 to 2019. The variables used for predicting posttransplant outcomes included donor and recipient as well as medical and social factors. We evaluated 7 ML models—extreme gradient boosting (XGBoost), logistic regression, support vector machine, random forest (RF), stochastic gradient descent, multilayer perceptron, and adaptive boosting (AdaBoost)—as well as a deep learning model with 2 hidden layers with 100 neurons and a rectified linear unit (ReLU) activation function followed by batch normalization for each and a classification head with a softmax activation function. We used 10-fold cross-validation to evaluate model performance. Shapley additive explanations (SHAP) values were calculated to estimate the importance of each variable for prediction. Results: RF and AdaBoost models were the best-performing algorithms for different prediction windows across outcomes. RF outperformed other ML algorithms in predicting 5 of the 6 outcomes (area under the receiver operating characteristic curve [AUROC] 0.664 and 0.706 for 1-year and 3-year rejection, respectively, and AUROC 0.697, 0.758, and 0.763 for 1-year, 3-year, and 5-year mortality, respectively). AdaBoost achieved the best performance for prediction of 5-year rejection (AUROC 0.705). Conclusions: This study demonstrates the comparative utility of ML approaches for modeling posttransplant health outcomes using registry data. ML approaches can identify unique risk factors and their complex relationship with outcomes, thereby identifying patients considered to be at risk and informing the transplant community about the potential of these innovative approaches to improve pediatric care after heart transplantation. Future studies are required to translate the information derived from prediction models to optimize counseling, clinical care, and decision-making within pediatric organ transplant centers.

21 May 2021 in Cardiology in the Young
Verified
Journal Article
Open access
PDF file
LWLaura Wilson
Laura Wilson
ACArun Chandran
Arun Chandran
JFJames C. Fudge
James C. Fudge
DMDiego Moguillansky
Diego Moguillansky
ATAkaluck Thatayatikom
Akaluck Thatayatikom
DGDipankar Gupta
Dipankar Gupta
This series describes three adolescent females who presented with chest pain and ventricular dysfunction related to acute coronary ischemia secondary to Takayasu’s arteritis with varied courses of disease progression leading to a diverse range of therapies including cardiac transplantation. While Takayasu’s arteritis is rare in childhood, it should be strongly considered in any adolescent female presenting with systemic inflammation and chest pain consistent with myocardial infarction. A high index of suspicion can lead to early detection and aggressive management of the underlying vasculitis reducing associated morbidity and mortality. The purpose of this report is to describe the challenges in the clinical diagnosis and management of Takayasu’s arteritis with myocardial infarction. We also seek to enhance awareness about unique presentations of Takayasu’s arteritis within the paediatric community.

18 April 2022 in Pediatric Transplantation
Verified
Journal Article
Open access
MKMichael O. Killian
Michael O. Killian
SCStephanie Clifford
Stephanie Clifford
MLMia Liza A. Lustria
Mia Liza A. Lustria
GSGage L. Skivington
Gage L. Skivington
DGDipankar Gupta
Dipankar Gupta
Purpose: HT recipients experience high levels of medication non‐adherence during adolescence. This pilot study examined the acceptability and feasibility of an asynchronous DOT mHealth application among adolescent HT recipients. The app facilitates tracking of patients’ dose‐by‐dose adherence and enables transplant team members to engage patients. The DOT application allows patients to self‐record videos while taking their medication and submit for review. Transplant staff review the videos and communicate with patients to engage and encourage medication adherence.Methods: Ten adolescent HT recipients with poor adherence were enrolled into a single‐group, 12‐week pilot study examining the impact of DOT on adherence. Secondary outcomes included self‐report measures from patients and parents concerning HRQOL and adherence barriers. Long‐term health outcomes assessed included AR and hospitalization 6 months following DOT.Findings: Among 14 adolescent HT patients approached, 10 initiated the DOT intervention. Of these, 8 completed the 12‐week intervention. Patients and caregivers reported high perceptions of acceptability and accessibility. Patients submitted 90.1% of possible videos demonstrating medication doses taken. MLVI values for the 10 patients initiating DOT decreased from 6 months prior to the intervention (2.86 ± 1.83) to 6 months following their involvement (2.08 ± 0.87) representing a 21.7% decrease in non‐adherence, though not statistically significant given the small sample size.Conclusions: Result of this pilot study provides promising insights regarding the feasibility, acceptability, and potential impact of DOT for adolescent HT recipients. Further randomized studies are required to confirm these observations.