Rajeev Dhupar

Dr. Rajeev Dhupar

1. Department of Cardiothoracic Surgery, University of Pittsburgh School of Medicine, Suite 715, Shadyside Medical Building, 5200 Centre Avenue, Pittsburgh, PA 15232, USA
+1
orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
1 April 2022 in The Journal of Thoracic and Cardiovascular Surgery
Verified
Journal Article
Open access
SASyed Faaz Ashraf
Syed Faaz Ashraf
KYKe Yin
Ke Yin
CMCindy X. Meng
Cindy X. Meng
QWQi Wang
Qi Wang
QWQiong Wang
Qiong Wang
RDRajeev Dhupar
Rajeev Dhupar
The study objective was to investigate if machine learning algorithms can predict whether a lung nodule is benign, adenocarcinoma, or its preinvasive subtype from computed tomography images alone. A dataset of chest computed tomography scans containing lung nodules was collected with their pathologic diagnosis from several sources. The dataset was split randomly into training (70%), internal validation (15%), and independent test sets (15%) at the patient level. Two machine learning algorithms were developed, trained, and validated. The first algorithm used the support vector machine model, and the second used deep learning technology: a convolutional neural network. Receiver operating characteristic analysis was used to evaluate the performance of the classification on the test dataset. The support vector machine/convolutional neural network-based models classified nodules into 6 categories resulting in an area under the curve of 0.59/0.65 when differentiating atypical adenomatous hyperplasia versus adenocarcinoma in situ, 0.87/0.86 with minimally invasive adenocarcinoma versus invasive adenocarcinoma, 0.76/0.72 atypical adenomatous hyperplasia + adenocarcinoma in situ versus minimally invasive adenocarcinoma, 0.89/0.87 atypical adenomatous hyperplasia + adenocarcinoma in situ versus minimally invasive adenocarcinoma + invasive adenocarcinoma, and 0.93/0.92 atypical adenomatous hyperplasia + adenocarcinoma in situ + minimally invasive adenocarcinoma versus invasive adenocarcinoma. Classifying benign versus atypical adenomatous hyperplasia + adenocarcinoma in situ + minimally invasive adenocarcinoma versus invasive adenocarcinoma resulted in a micro-average area under the curve of 0.93/0.94 for the support vector machine/convolutional neural network models, respectively. The convolutional neural network-based methods had higher sensitivities than the support vector machine-based methods but lower specificities and accuracies. The machine learning algorithms demonstrated reasonable performance in differentiating benign versus preinvasive versus invasive adenocarcinoma from computed tomography images alone. However, the prediction accuracy varies across its subtypes. This holds the potential for improved diagnostic capabilities with less-invasive means.

1 April 2017 in The Annals of Thoracic Surgery
Verified
Journal Article
Closed access
RDRajeev Dhupar
Rajeev Dhupar
LKLauren Van Der Kraak
Lauren Van Der Kraak
APArjun Pennathur
Arjun Pennathur
MSMatthew J. Schuchert
Matthew J. Schuchert
KNKatie S. Nason
Katie S. Nason
MLMichael T. Lotze
Michael T. Lotze

11 July 2022 in Medical Physics
Verified
Journal Article
Open access
LPLucy Pu
Lucy Pu
SASyed F. Ashraf
Syed F. Ashraf
NGNaciye S. Gezer
Naciye S. Gezer
IOIclal Ocak
Iclal Ocak
DDDaniel E. Dresser
Daniel E. Dresser
RDRajeev Dhupar
Rajeev Dhupar
Background: Estimating whole‐body composition from limited region‐computed tomography (CT) scans has many potential applications in clinical medicine; however, it is challenging.Purpose: To investigate if whole‐body composition based on several tissue types (visceral adipose tissue [VAT], subcutaneous adipose tissue [SAT], intermuscular adipose tissue [IMAT], skeletal muscle [SM], and bone) can be reliably estimated from a chest CT scan only.Methods: A cohort of 97 lung cancer subjects who underwent both chest CT scans and whole‐body positron emission tomography‐CT scans at our institution were collected. We used our in‐house software to automatically segment and quantify VAT, SAT, IMAT, SM, and bone on the CT images. The field‐of‐views of the chest CT scans and the whole‐body CT scans were standardized, namely, from vertebra T1 to L1 and from C1 to the bottom of the pelvis, respectively. Multivariate linear regression was used to develop the computer models for estimating the volumes of whole‐body tissues from chest CT scans. Subject demographics (e.g., gender and age) and lung volume were included in the modeling analysis. Ten‐fold cross‐validation was used to validate the performance of the prediction models. Mean absolute difference (MAD) and R‐squared (R2) were used as the performance metrics to assess the model performance.Results: The R2 values when estimating volumes of whole‐body SAT, VAT, IMAT, total fat, SM, and bone from the regular chest CT scans were 0.901, 0.929, 0.900, 0.933, 0.928, and 0.918, respectively. The corresponding MADs (percentage difference) were 1.44 ± 1.21 L (12.21% ± 11.70%), 0.63 ± 0.49 L (29.68% ± 61.99%), 0.12 ± 0.09 L (16.20% ± 18.42%), 1.65 ± 1.40 L (10.43% ± 10.79%), 0.71 ± 0.68 L (5.14% ± 4.75%), and 0.17 ± 0.15 L (4.32% ± 3.38%), respectively.Conclusion: Our algorithm shows promise in its ability to estimate whole‐body compositions from chest CT scans. Body composition measures based on chest CT scans are more accurate than those based on vertebra third lumbar.

2 September 2022 in Medical Physics
Verified
Journal Article
Closed access
LPLucy Pu
Lucy Pu
NGNaciye S Gezer
Naciye S Gezer
SASyed Faaz Ashraf
Syed Faaz Ashraf
IOIclal Ocak
Iclal Ocak
DDDaniel E. Dresser
Daniel E. Dresser
RDRajeev Dhupar
Rajeev Dhupar

19 August 2021 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
CEChigozirim N. Ekeke
Chigozirim N. Ekeke
MVMatthew Vercauteren
Matthew Vercauteren
SISmiljana Istvaniczdravkovic
Smiljana Istvaniczdravkovic
RSRoy W. Semaan
Roy W. Semaan
RDRajeev Dhupar
Rajeev Dhupar
The incidence of lung nodules has increased with improved diagnostic imaging and screening protocols. Despite improvements for diagnosing pulmonary nodules with technologies such as electromagnetic navigational bronchoscopy (ENB), several limitations still exist including adequate visualization, localization, and diagnostic yield. Robotic-assisted bronchoscopy with ENB has been introduced as a method to overcome these shortcomings. We describe our initial experience in evaluating lung nodules with robotic assisted bronchoscopy. We retrospectively reviewed data on the first 25 patients that underwent robotic-assisted bronchoscopy and biopsy. We analyzed success with localization, diagnostic yield, and post procedural morbidity. Diagnostic yield was 96% (24/25) with no periprocedural morbidity. The majority of nodules were malignant or atypical (76%) and were located in the right upper lobe. Diameter ranged between 0.8–6.9 cm (median size 1–2 cm). Seventy-five percent of patients underwent subsequent treatment for cancer based on these results, with 25% having continued surveillance. Robotic assisted bronchoscopy is safe and accurate. Studies with larger numbers will allow better understanding of the diagnostic yield and clinical utility of this approach in comparison to other diagnostic tools for lung nodules.

1 April 2022 in The Journal of Thoracic and Cardiovascular Surgery
Verified
Journal Article
Open access
CEChigozirim N. Ekeke
Chigozirim N. Ekeke
KRKira L. Russell
Kira L. Russell
PMPranav Murthy
Pranav Murthy
Zong Sheng GuoZong Sheng Guo
Zong Sheng Guo
ASAdam C. Soloff
Adam C. Soloff
RDRajeev Dhupar
Rajeev Dhupar
The mainstay of treatment for patients with malignant pleural disease is fluid drainage and systemic therapy. A tumor-specific oncolytic virus or T-cell-activating interleukin-2 immunotherapy may provide an opportunity for local control. We previously developed a vaccinia virus-expressing interleukin-2, an oncolytic virus that mediated tumor regression in preclinical peritoneal tumor models with expansion of tumor-infiltrating lymphocytes. We evaluated the antitumor efficacy and immune modulatory effects of vaccinia virus-expressing interleukin-2 in malignant pleural disease. A murine model of malignant pleural disease was established with percutaneous intrapleural deposition of the Lewis lung carcinoma cell line and monitored with bioluminescent imaging. After intrapleural or systemic administration of vaccinia viruses (vaccinia virus yellow fluorescent protein control, vaccinia virus-expressing interleukin-2), systemic anti-programmed cell death-1 antibody, or combination therapy (vaccinia virus-expressing interleukin-2 and anti-programmed cell death-1), tumor mass, immune cell infiltration, T-cell receptor diversity, and survival were assessed. Intrapleural vaccinia virus resulted in significant tumor regression compared with phosphate-buffered saline control (P < .05). Inclusion of the interleukin-2 transgene further increased intratumoral CD8+ T cells (P < .01) and programmed cell death-1 expression on CD8+ tumor-infiltrating lymphocytes (P < .001). Intrapleural vaccinia virus-expressing interleukin-2 was superior to systemic vaccinia virus-expressing interleukin-2, with reduced tumor burden (P < .0001) and improved survival (P < .05). Intrapleural vaccinia virus-expressing interleukin-2 alone or combined treatment with systemic anti-programmed cell death-1 reduced tumor burden (P < .01), improved survival (P < .01), and increased intratumoral αβ T-cell receptor diversity (P < .05) compared with systemic anti-programmed cell death-1 monotherapy. Intrapleural vaccinia virus-expressing interleukin-2 reduced tumor burden and enhanced survival in a murine malignant pleural disease model. Increased CD8+ tumor-infiltrating lymphocytes and αβ T-cell receptor diversity are associated with enhanced response. Clinical trials will enable assessment of intrapleural vaccinia virus-expressing interleukin-2 therapy in patients with malignant pleural disease.