Pushkar Mutha
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Pushkar Mutha has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Anant Madabhushi
Department of Biomedical Engineering, Emory University, Atlanta, GA
2 shared publications - Bastiaan Driehuys
Medical Physics Graduate Program Duke University Durham USA
1 shared publication Bhavika Kaul
Division of Pulmonary and Critical Care Medicine, Baylor College of Medicine, Houston, TX, USA
1 shared publicationDavid Mummy
Department of Medicine, Duke University, Durham, NC, USA
1 shared publication- George Lucas Silva
Department of Pediatrics, Emory University School of Medicine, Atlanta, GA, USA
1 shared publication - Gourav Modanwal
Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University , Health Sciences Research Building II, 1750 Haygood Drive, Suite N647, Atlanta, GA 30322
1 shared publication
Latest Publications
- Open Access
Radiomics of the Airway (RadAr): Multi-Scale Airway Phenotyping for Disease Characterization on Routine CT Imaging
Background: Airway remodeling is a convergent feature across respiratory diseases, yet current quantitative CT tools quantify a small number of prespecified structural abnormalities of the airway tree. Radiomics of the Airway (RadAr) derives interpretable, multi-scale airway measurements from chest CT to characterize airway deformation and discover quantitative imaging biomarkers from routine chest CT. Methods: RadAr extracts over 400 multi-scale, interpretable airway measurements across lobes or generations capturing luminal dimensions, tapering, architectural distortion, and global morphology. It was evaluated in N=1331 patients across four settings: 63-week mortality prediction in fibrotic interstitial lung disease (fILD), COVID-19 severity prediction, structure-function association in progressive pulmonary fibrosis (PPF) and structure-inflammation markers in pediatric cystic fibrosis (CF). Unsupervised clustering was used to identify airway phenotypes across the fILD and COVID-19 cohorts. Results: In fILD, lower-lobe architectural distortion was associated with mortality (balanced accuracy 0·654). In COVID-19, severe disease was independently associated with luminal dilation (AUC 0·719, odds ratio 2·32, p=0·017). In PPF, airway phenotypes correlated with forced vital capacity (ρ=0·83), mid-expiratory flow (ρ=0·87), and ¹²⁹Xe MRI alveolar gas exchange impairment (ρ=0·70). In pediatric CF, reduced tapering and increased cylindricity were associated with prior exacerbations and bronchoalveolar lavage neutrophilia (ρ=-0·64 to -0·78). Five phenotypes were identified from extensive, tapered airway trees to sparse, dilated, thick-walled, tortuous trees, with increasing COVID-19 severity and fILD mortality across this spectrum. Conclusion: RadAr identified interpretable, disease-specific airway phenotypes associated with function and outcomes across restrictive, obstructive, and mixed lung diseases in adult and pediatric settings. These findings establish a framework for discovery and development of quantitative airway biomarkers for patient stratification, disease monitoring, or imaging endpoint development in pulmonary trials.
Open AccessRadiomics of the Airway (RadAr): Multi-Scale Airway Phenotyping for Disease Characterization on Routine CT Imaging
Background: Airway remodeling is a convergent feature across respiratory diseases, yet current quantitative CT tools quantify a small number of prespecified structural abnormalities of the airway tree. Radiomics of the Airway (RadAr) derives interpretable, multi-scale airway measurements from chest CT to characterize airway deformation and discover quantitative imaging biomarkers from routine chest CT. Methods: RadAr extracts over 400 multi-scale, interpretable airway measurements across lobes or generations capturing luminal dimensions, tapering, architectural distortion, and global morphology. It was evaluated in N=1331 patients across four settings: 63-week mortality prediction in fibrotic interstitial lung disease (fILD), COVID-19 severity prediction, structure-function association in progressive pulmonary fibrosis (PPF) and structure-inflammation markers in pediatric cystic fibrosis (CF). Unsupervised clustering was used to identify airway phenotypes across the fILD and COVID-19 cohorts. Results: In fILD, lower-lobe architectural distortion was associated with mortality (balanced accuracy 0·654). In COVID-19, severe disease was independently associated with luminal dilation (AUC 0·719, odds ratio 2·32, p=0·017). In PPF, airway phenotypes correlated with forced vital capacity (ρ=0·83), mid-expiratory flow (ρ=0·87), and ¹²⁹Xe MRI alveolar gas exchange impairment (ρ=0·70). In pediatric CF, reduced tapering and increased cylindricity were associated with prior exacerbations and bronchoalveolar lavage neutrophilia (ρ=-0·64 to -0·78). Five phenotypes were identified from extensive, tapered airway trees to sparse, dilated, thick-walled, tortuous trees, with increasing COVID-19 severity and fILD mortality across this spectrum. Conclusion: RadAr identified interpretable, disease-specific airway phenotypes associated with function and outcomes across restrictive, obstructive, and mixed lung diseases in adult and pediatric settings. These findings establish a framework for discovery and development of quantitative airway biomarkers for patient stratification, disease monitoring, or imaging endpoint development in pulmonary trials.
- Open Access
- 18 August 2025
- Published by Wolters Kluwer Health in Kidney360
- Vol. 7 (1) , 81-93
Fundus Photograph-Derived Computational Features Predict Risk of Cardiovascular Events in the Chronic Renal Insufficiency Cohort Clinical Observational Study
Key Points: Retinal imaging reveals microvascular changes associated with cardiovascular (CV) risk in patients with CKD. A retinal vascular model improves risk stratification compared with established clinical risk scores for CV events. Integrating retinal analysis with clinical markers enhances CV risk assessment in patients with CKD. Background: Patients with CKD face an elevated but variable risk of cardiovascular (CV) disease. Retinal imaging in CKD provides a noninvasive opportunity for CV risk stratification through microvascular analysis. The objective of this study was to evaluate retinal vascular features extracted via computer vision and machine learning approaches for CV risk and their added value over established risk calculators in patients with CKD. Methods: Retinal scans from 1333 participants of the multicenter clinical observational study Chronic Renal Insufficiency Cohort (NCT00304148) were analyzed. A deep learning pipeline segmented vessels and then identified arterioles and venules from them. Segmented vessel, arteriole, and venule masks were used to extract 384 vascular features. An elastic-net model—Cardiovascular Assessment through Retinal Evaluation in CKD (CARE-CKD; M CARE )—was trained, using the top eight features, on 567 participants (101 major adverse cardiovascular events [MACE]: composite of myocardial infarction, stroke, heart failure) and validated on 244 participants (44 MACE). A nomogram integrating M CARE with clinical markers (age, sex, blood pressure, smoking, eGFR, albuminuria, cholesterol, body mass index, and diabetes status) was developed. Results: M CARE demonstrated strong prognostic performance for predicting MACE, (C-index=0.70, hazard ratio [HR] =3.95, above versus below median; 95% confidence interval [CI], 2.36 to 6.63; P P < 0.01) and Predicting Risk of cardiovascular disease EVENTs (PREVENT; C-index=0.65; HR=1.84; LRT P < 0.001) calculators. M CARE improved risk stratification within Framingham Risk Score–based high-risk (HR=3.73; P < 0.001) and Predicting Risk of cardiovascular disease EVENT–based high-risk (HR=4.73; P < 0.001) categories. Nomogram enhanced risk stratification (C-index=0.77; HR=3.81; P P < 0.01). Conclusions: CARE-CKD provides a novel, opportunistic approach to CV risk assessment in CKD, outperforming the established risk calculators and refining stratification within high-risk categories. By enabling earlier identification, close monitoring, and targeted management of high-risk patients, CARE-CKD addresses gaps left by traditional calculators, maximizing the benefits of emerging therapies and potentially improving long-term outcomes.
- 18 August 2025
- Published by Wolters Kluwer Health in Kidney360
- Vol. 7 (1) , 81-93
Open AccessFundus Photograph-Derived Computational Features Predict Risk of Cardiovascular Events in the Chronic Renal Insufficiency Cohort Clinical Observational Study
Key Points: Retinal imaging reveals microvascular changes associated with cardiovascular (CV) risk in patients with CKD. A retinal vascular model improves risk stratification compared with established clinical risk scores for CV events. Integrating retinal analysis with clinical markers enhances CV risk assessment in patients with CKD. Background: Patients with CKD face an elevated but variable risk of cardiovascular (CV) disease. Retinal imaging in CKD provides a noninvasive opportunity for CV risk stratification through microvascular analysis. The objective of this study was to evaluate retinal vascular features extracted via computer vision and machine learning approaches for CV risk and their added value over established risk calculators in patients with CKD. Methods: Retinal scans from 1333 participants of the multicenter clinical observational study Chronic Renal Insufficiency Cohort (NCT00304148) were analyzed. A deep learning pipeline segmented vessels and then identified arterioles and venules from them. Segmented vessel, arteriole, and venule masks were used to extract 384 vascular features. An elastic-net model—Cardiovascular Assessment through Retinal Evaluation in CKD (CARE-CKD; M CARE )—was trained, using the top eight features, on 567 participants (101 major adverse cardiovascular events [MACE]: composite of myocardial infarction, stroke, heart failure) and validated on 244 participants (44 MACE). A nomogram integrating M CARE with clinical markers (age, sex, blood pressure, smoking, eGFR, albuminuria, cholesterol, body mass index, and diabetes status) was developed. Results: M CARE demonstrated strong prognostic performance for predicting MACE, (C-index=0.70, hazard ratio [HR] =3.95, above versus below median; 95% confidence interval [CI], 2.36 to 6.63; P P < 0.01) and Predicting Risk of cardiovascular disease EVENTs (PREVENT; C-index=0.65; HR=1.84; LRT P < 0.001) calculators. M CARE improved risk stratification within Framingham Risk Score–based high-risk (HR=3.73; P < 0.001) and Predicting Risk of cardiovascular disease EVENT–based high-risk (HR=4.73; P < 0.001) categories. Nomogram enhanced risk stratification (C-index=0.77; HR=3.81; P P < 0.01). Conclusions: CARE-CKD provides a novel, opportunistic approach to CV risk assessment in CKD, outperforming the established risk calculators and refining stratification within high-risk categories. By enabling earlier identification, close monitoring, and targeted management of high-risk patients, CARE-CKD addresses gaps left by traditional calculators, maximizing the benefits of emerging therapies and potentially improving long-term outcomes.