Sourav Mukherjee

ORCID: 0000-0003-1647-7324
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Blood groups and transfusion
  • Erythrocyte Function and Pathophysiology
  • Blood disorders and treatments
  • Statistical Methods and Inference
  • Tensor decomposition and applications
  • Statistical Methods and Bayesian Inference
  • Bayesian Methods and Mixture Models
  • Cancer Immunotherapy and Biomarkers
  • Markov Chains and Monte Carlo Methods
  • Melanoma and MAPK Pathways
  • Neurosurgical Procedures and Complications
  • Advanced Bandit Algorithms Research
  • Stochastic processes and statistical mechanics
  • Complex Network Analysis Techniques
  • Herpesvirus Infections and Treatments
  • Takotsubo Cardiomyopathy and Associated Phenomena
  • Neonatal Health and Biochemistry
  • Pancreatic and Hepatic Oncology Research
  • CAR-T cell therapy research
  • Gastric Cancer Management and Outcomes
  • Consumer Market Behavior and Pricing
  • Cardiac Arrest and Resuscitation
  • Auction Theory and Applications

US Forest Service
2025

Bristol-Myers Squibb (Germany)
2024

Bristol-Myers Squibb (United States)
2024

University of Florida
2021-2023

Open-label phase II study (RELATIVITY-060) to investigate the efficacy and safety of first-line nivolumab, a PD-1-blocking antibody, plus relatlimab, lymphocyte-activation gene 3 (LAG-3)-blocking chemotherapy in patients with previously untreated advanced gastric cancer (GC) or gastroesophageal junction (GEJC).

10.1200/jco.23.01636 article EN Journal of Clinical Oncology 2024-05-09

9504 Background: NIVO (anti–PD-1 antibody) is approved as monotherapy and in combination with either RELA (anti–LAG-3 or IPI (anti–CTLA-4 for the treatment of patients (pts) advanced melanoma. RELATIVITY-048 (NCT03459222) a phase 1/2, nonrandomized trial evaluating immuno-oncology (I-O) triplets, including + IPI, pts select solid tumors. This analysis from focused on use melanoma cohort first disclosure study. Methods: Pts were treated first-line 480 mg Q4W 160 1 mg/kg Q8W until progression...

10.1200/jco.2024.42.16_suppl.9504 article EN Journal of Clinical Oncology 2024-06-01

Introduction Takotsubo Cardiomyopathy (TCM) is a rare but well recognized complication of subarachnoid hemorrhage associated with increased morbidity and poor clinical outcomes. There scarcity literature describing the prevalence risk factors this complication. The aim study was to identify patients who are at developing TCM in non-traumatic hemorrhage. Methods 2016 2021 National Inpatient Sample (NIS) used adult inpatients primary diagnosis Univariate multivariable analyses adjusting for...

10.1177/08850666251313759 article EN Journal of Intensive Care Medicine 2025-01-19

The tensor Ising model is a discrete exponential family used for modeling binary data on networks with not just pairwise, but higher-order dependencies. A particularly important class of models are the Curie-Weiss models, where all tuples nodes particular order interact same intensity. computationally efficient alternative to intractible maximum likelihood estimator (MLE) in this model, pseudolikelihood (MPLE). In paper, we show that MPLE fact as MLE (in Bahadur sense) 2-spin and values null...

10.1214/24-ejs2255 article EN cc-by Electronic Journal of Statistics 2024-01-01

ABSTRACT Background and Objectives: “Block D” in hemolytic disease of the fetus newborn (HDFN) is quite rare should be evaluated meticulously. We analyzed phenomenon neonates with HDFN due to rhesus (Rh) incompatibilities. The objective this study was identify blocking by simple, sensitive methods so that they can implemented routine practices. Methods: conducted over 10 years, written informed consent taken from each participating parent/guardian. Detailed immunohematological workup for...

10.4103/gjtm.gjtm_28_24 article EN cc-by-nc-sa Global Journal of Transfusion Medicine 2024-07-01

The logistic and probit link functions are the most common choices for regression models with a binary response. However, these not robust to presence of outliers/unexpected observations. robit function, which is equal inverse CDF Student’s t-distribution, provides alternative functions. A multivariate normal prior coefficients standard choice Bayesian inference in models. resulting posterior density intractable Data Augmentation (DA) Markov chain used generate approximate samples from...

10.1214/22-ejs2098 article EN cc-by Electronic Journal of Statistics 2023-01-01

We focus on online second price auctions, where bids are made sequentially, and the winning bidder pays maximum of second-highest bid a seller specified reserve price. For many such does not see all or total number bidders accessing auction, only observes current selling prices throughout course auction. develop novel non-parametric approach to estimate underlying consumer valuation distribution based this data. Previous approaches in literature use final assume knowledge bidders. The...

10.48550/arxiv.2312.07882 preprint EN cc-by arXiv (Cornell University) 2023-01-01

Examples of group B red cells that react weakly or not at all with anti-B have been described. Subgroups such as B3, Bx, Bm, and Bel are rare less frequently reported. We studied the frequency subgroups in our healthy blood donor population serologically characterized differentiated these subgroups. The 9-year prospective study included 84,534 donors. Initial grouping antibody screening samples were performed using automated solid-phase assay. Any sample showing discrepancy weaker...

10.4103/ajts.ajts_36_23 article EN cc-by-nc-sa Asian Journal of Transfusion Science 2023-11-07

The logistic and probit link functions are the most common choices for regression models with a binary response. However, these not robust to presence of outliers/unexpected observations. robit function, which is equal inverse CDF Student's $t$-distribution, provides alternative functions. A multivariate normal prior coefficients standard choice Bayesian inference in models. resulting posterior density intractable Data Augmentation (DA) Markov chain used generate approximate samples from...

10.48550/arxiv.2112.10349 preprint EN cc-by-nc-sa arXiv (Cornell University) 2021-01-01

The tensor Ising model is a discrete exponential family used for modeling binary data on networks with not just pairwise, but higher-order dependencies. A particularly important class of models are the Curie-Weiss models, where all tuples nodes particular order interact same intensity. maximum likelihood estimator (MLE) explicit in this model, due to presence an intractable normalizing constant likelihood, and computationally efficient alternative use pseudolikelihood (MPLE). In paper, we...

10.48550/arxiv.2109.11764 preprint EN cc-by-nc-sa arXiv (Cornell University) 2021-01-01
Coming Soon ...