KGKevin Grudzinski

Kevin Grudzinski

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kevin Grudzinski has a total of 109 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Treatment trajectories by radiologic pattern of lung injury in interstitial lung disease with myositis-associated antibodies: a retrospective cohort study

    by Kevin Grudzinski, Thomas C Bolig, Malek Shawabkeh, Kavitha C Selvan, Raymond J Goodwin, Emily Olson, Bradford C Bemiss, Nishant Parekh, Hatice Savas, Jane E Dematte, et al.

    Objectives: Lung involvement in interstitial lung disease (ILD) with myositis-associated antibodies (myositis-ILD) predominantly has two radiologic patterns—nonspecific interstitial pneumonia (NSIP) and organizing pneumonia (OP)—that often coexist, yet clinical outcomes of these patterns remain unclear. We aimed to describe functional and radiologic trajectories during immunomodulatory therapy by predominant radiologic pattern in myositis-ILD. Methods: In this retrospective, single-centre cohort study, we studied participants from the Northwestern University ILD Registry with a circulating myositis-associated antibody, ILD, and at least 6 months of immunomodulatory therapy during a 24-month observation period. Two thoracic radiologists assigned the predominant radiologic pattern (NSIP or OP). The primary outcome was the change in forced vital capacity (FVC); secondary outcomes were the change in diffusing capacity of the lungs for carbon monoxide (DLCO) and qualitative and quantitative (Kazerooni score) radiologic change. Groups were compared with Fisher exact and Wilcoxon rank-sum tests. Results: Forty-one participants were characterized (29 [71%] OP-predominant, 12 [29%] NSIP-predominant). Both subgroups had improved FVC on immunomodulatory therapy (OP +0.36 l [P = 0.006]; NSIP +0.25 l [P = 0.09]). DLCO did not change significantly in either subgroup (OP +1.22 [P = 0.11]; NSIP -0.75 [P = 0.37]). The OP cohort improved on qualitative imaging analysis (P < 0.05), driven by reduced groundglass/consolidative opacity (P = 0.006). Participants with NSIP/OP overlap had greater baseline radiologic severity. Conclusions: In myositis-ILD, both radiologic subgroups showed measurable functional and radiologic change during immunomodulatory therapy, with larger improvements in the OP-predominant subgroup; the magnitude of change did not differ significantly between subgroups. These descriptive, hypothesis-generating findings require multicentre validation.

  • Open Access

    Adult Life Course Trajectories of Lung Function and the Development of Interstitial Lung Abnormalities: The CARDIA Lung Study

    by Kevin M. Grudzinski, Gabrielle Y. Liu, Laura A. Colangelo, Kavitha C. Selvan, Rachel Putman, Gary M. Hunninghake, Raul San Jose Estepar, George Washko, Ravi Kalhan, Anthony J. Esposito

    Background: Interstitial lung abnormalities (ILA) are radiologic findings of increased lung density or fibrosis in individuals without clinical interstitial lung disease (ILD) and are associated with increased mortality and progression to ILD. Understanding physiologic trajectories of lung function preceding ILA diagnosis may illuminate early mechanisms of lung injury. Methods: We recruited participants from the Coronary Artery Risk Development in Young Adults (CARDIA) Lung Study, a prospective cohort of adults enrolled at ages 18–30 years and followed longitudinally for 25 years. Percent predicted forced vital capacity (ppFVC) was measured at five study visits over 20 years. Individual ppFVC trajectories were estimated using random coefficient models. Person-specific slopes were incorporated into logistic regression models to examine associations with visually detected ILA on chest CT at exam year 25. Models were adjusted for age, sex, race, body mass index, pack-years of smoking, and study center. Results: Among 3,136 participants with complete data, 57 (1.8%) had ILA at mean age 51 years. In univariable and multivariable models, individuals with ILA had greater cumulative decline in ppFVC over the 20 years preceding diagnosis. Each 10% absolute decline in ppFVC was associated with more than twice the odds of ILA (adjusted OR 2.21, 95% confidence interval 1.47–3.31, p = 0.0001). Conclusions: Greater longitudinal decline in FVC from early adulthood was strongly associated with the presence of ILA at midlife. These findings suggest that physiologic impairments precede radiologic evidence of subclinical parenchymal lung abnormalities, underscoring the potential of life course lung function trajectories to identify individuals at risk for developing ILD.

  • Open Access

    Therapeutic Response by Radiologic Pattern of Lung Injury in Myositis-Associated Interstitial Lung Disease: A Retrospective Cohort Study

    by Thomas C. Bolig, Kevin Grudzinski, Malek Shawabkeh, Kavitha C. Selvan, Raymond J. Goodwin, Emily Olson, Bradford C. Bemiss, Nishant Parekh, Hatice Savas, Jane E. Dematte, et al.

    Objective: Myositis-associated interstitial lung disease (myositis-ILD) consists of two predominant radiologic patterns of lung injury—nonspecific interstitial pneumonia (NSIP) and organizing pneumonia (OP)—that oftentimes coexist. However, it remains unclear whether either is associated with clinical outcomes. We aimed to assess the therapeutic response in patients with NSIP-compared to those with OP-predominant myositis-ILD. Methods: This retrospective, single-center cohort study recruited participants from the Northwestern University ILD Registry with a circulating myositis-associated antibody, ILD, and at least 6 months of follow-up while on immunomodulatory therapy during a 24-month observation period after diagnosis. Two thoracic radiologists determined the predominant radiologic pattern (NSIP or OP). The primary outcome was the absolute change in forced vital capacity (FVC) at 24 months post-diagnosis. Secondary outcomes included changes in the diffusing capacity of the lung for carbon monoxide (DLCO) and radiologic qualitative and quantitative measures of lung injury. Results: Forty-one participants were included in analyses. 71% had an OP-predominant while 29% had an NSIP-predominant radiologic pattern of lung injury. Both exposure cohorts had improvement in mean absolute FVC (OP cohort = +0.18L [p=0.005], NSIP cohort = +0.24L [p=0.07]) over the 24-month observation period. The OP (pRadiologic phenotyping may predict therapeutic response in myositis-ILD . This study demonstrates that an OP-predominant computed tomography (CT) pattern of lung injury is associated with greater improvement in lung function and radiologic signs of inflammation over 24 months on at least 6 months of immunomodulatory therapy compared with an NSIP-predominant pattern, suggesting that CT pattern may provide clinically meaningful prognostic information. First study to integrate blinded qualitative radiologic adjudication with quantitative CT scoring in myositis-ILD . By combining dual-radiologist review with Kazerooni quantitative scoring and longitudinal pulmonary function testing, this study offers a rigorous and multidimensional assessment of treatment response. Expands risk stratification beyond antibody-based toward imaging-based phenotyping strategies . In a heterogeneous population defined by diverse myositis-associated antibodies, this work introduces radiologic pattern as a practical and accessible framework for anticipating treatment responsiveness. Provides hypothesis-generating data for precision management in myositis-ILD . The findings support the concept that imaging-defined subgroups may exhibit differential therapeutic trajectories, laying groundwork for future multicenter studies integrating CT phenotype, antibody profile, and treatment strategy.


  • REN: Anatomically-Informed Mixture-of-Experts for Interstitial Lung Disease Diagnosis

    by Alec K. Peltekian, Halil Ertugrul Aktas, Gorkem Durak, Kevin Grudzinski, Bradford C. Bemiss, Carrie Richardson, Jane E. Dematte, G. R. Scott Budinger, Anthony J. Esposito, Alexander Misharin, et al.

    Mixture-of-Experts (MoE) architectures achieve scalable learning by routing inputs to specialized subnetworks through conditional computation. However, conventional MoE designs assume homogeneous expert capability and domain-agnostic routing-assumptions that are fundamentally misaligned with medical imaging, where anatomical structure and regional disease heterogeneity govern pathological patterns. We introduce Regional Expert Networks (REN), the first anatomically-informed MoE framework for medical image classification. REN encodes anatomical priors by training seven specialized experts, each dedicated to a distinct lung lobe or bilateral lung combination, enabling precise modeling of region-specific pathological variation. Multi-modal gating mechanisms dynamically integrate radiomics biomarkers with deep learning (DL) features extracted by convolutional (CNN), Transformer (ViT), and state-space (Mamba) architectures to weight expert contributions at inference. Applied to interstitial lung disease (ILD) classification on a 597-patient, 1,898-scan longitudinal cohort, REN achieves consistently superior performance: the radiomics-guided ensemble attains an average AUC of 0.8646 ± 0.0467, a +12.5% improvement over the SwinUNETR single-model base-line (AUC 0.7685, p = 0.031). Lower-lobe experts reach AUCs of 0.88-0.90, outperforming DL baselines (CNN: 0.76-0.79) and mirroring known patterns of basal ILD progression. Evaluated under rigorous patient-level cross-validation, REN demonstrates strong generalizability and clinical interpretability, establishing a scalable, anatomically-guided framework potentially extensible to other structured medical imaging tasks. Code is available on our GitHub https://github.com/NUBagciLab/MoE-REN.

  • Imaging-Based Mortality Prediction in Patients with Systemic Sclerosis

    by Alec K. Peltekian, Karolina Senkow, Gorkem Durak, Kevin M. Grudzinski, Bradford C. Bemiss, Jane E. Dematte, Carrie Richardson, Nikolay S. Markov, Mary Carns, Kathleen Aren, et al.


  • Open Access

    Profibrotic monocyte-derived alveolar macrophages as a biomarker and therapeutic target in systemic sclerosis-associated interstitial lung disease

    by Nikolay S. Markov, Anthony J. Esposito, Karolina J. Senkow, Maxwell Schleck, Luisa Cusick, Zhan Yu, Yuliana V. Sokolenko, Estefani Diaz, Emmy Jonasson, Suchitra Swaminathan, et al.

    Interstitial lung disease (ILD) is present in over 60% of patients with systemic sclerosis (SSc) and is the leading cause of SSc-related deaths. Profibrotic monocyte-derived alveolar macrophages (MoAM) play a causal role in the pathogenesis of pulmonary fibrosis in animal models where their persistence in the niche requires signaling through Colony Stimulating Factor 1 Receptor (CSF1R). We hypothesized that the presence and proportion of MoAM in bronchoalveolar lavage (BAL) fluid from patients with SSc-ILD may be a biomarker of ILD severity. To test this hypothesis, we analyzed BAL fluid from 9 prospectively enrolled patients with SSc-ILD and 13 healthy controls using flow cytometry and single-cell RNA sequencing. Patients with SSc-ILD had more MoAM and interstitial macrophages in BAL fluid than healthy controls, and their abundance was associated with lung fibrosis severity. We identified changes in the MoAM transcriptome as a function of treatment with mycophenolate, an effective therapy for SSc-ILD. In SSc-ILD lung explants, spatial transcriptomics identified an expanded population of interstitial macrophages spilling into the alveolar space. Our findings suggest that the proportion of profibrotic MoAM and interstitial macrophages in BAL fluid may serve as a biomarker of SSc-ILD and credential them as possible targets for therapy.


  • Open Access

    Machine Learning Analysis of Electronic Health Records Identifies Interstitial Lung Disease and Predicts Mortality in Patients with Systemic Sclerosis

    by Alec Peltekian, Kevin Grudzinski, Bradford Bemiss, Jane Dematte, Carrie Richardson, Nikolay Markov, Mary Carns, Kathleen Aren, Natania Field, Mengou Zhu, et al.

    Background: Interstitial lung disease (ILD) affects > 40% of patients with systemic sclerosis (SSc/scleroderma) and is the leading cause of disease-related mortality. Although therapies may slow progression, outcomes remain poor, partly because ILD is often detected after irreversible lung injury has occurred. Although chest computed tomography (CT) is a sensitive tool for ILD detection and is recommended at SSc diagnosis, it is oftentimes not performed and even less often performed serially. We sought to develop tools to predict ILD and mortality in patients with SSc using data routinely available in the electronic health record (EHR) to inform medical decision-making. Methods: We analyzed longitudinal EHR data from two SSc cohorts: Northwestern University (1,169 participants; derivation cohort) and Yale University (376 participants; validation cohort). We identified clinical features from existing cohort-linked EHR queries composing a convenience sample of data from participants spanning decades rather than employing a single unified data collection effort. Three ILD experts independently reviewed CT reports and classified each as having or lacking ILD. To explore derivation cohort data structure, patients with > =3 forced vital capacity (FVC) results available were identified and stratified according to prevalent or absent ILD. Using unsupervised trajectory-based clustering exploratory analyses, we determined standardized patterns across groups. ML models were then developed using clinical EHR data as predictor variables and prevalent ILD and all-cause mortality as outcome variables. Model performance was assessed using area under the receiver operating characteristic curve (AUC). Results: Seventy-four clinical features with low missingness, including demographic, vital sign, laboratory, and pulmonary function test data, were utilized for analyses. Four robust PFT trajectory clusters were identified that were associated with ILD prevalence and mortality in exploratory analyses. A ML model for ILD detection achieved an AUC of 0.832 and retained performance in the Yale cohort (AUC 0.754). In addition to established predictors such as autoantibodies and pulmonary function, the model identified routine laboratory measurements, including red cell distribution width (RDW), white blood cell count, and serum chloride, as important contributors. One-year mortality prediction achieved AUCs of 0.904 in the North-western cohort and 0.910 in the Yale cohort. Among patients with SSc-ILD, one-year mortality was predicted with AUCs of 0.744 and 0.902 in the Northwestern and Yale cohorts, respectively. Unexpectedly, we found that subtle laboratory abnormalities (such as change in RDW) contributed to predicting mortality. Conclusions: Our prediction models comprised of widely available EHR data are useful tools to identify SSc patients at high risk for prevalent ILD and all-cause mortality. Integration of these models into clinical practice could enable scalable risk stratification and inform individualized ILD screening and monitoring strategies for SSc patients.

  • Open Access

    Developing and validating a machine learning model to predict successful next-day extubation in the ICU

    by Samuel W Fenske, Alec Peltekian, Mengjia Kang, Nikolay S Markov, Mengou Zhu, Kevin Grudzinski, Melissa J Bak, Anna Pawlowski, Vishu Gupta, Yuwei Mao, et al.

    Background Criteria to identify patients who are ready to be liberated from mechanical ventilation are imprecise, often resulting in prolonged mechanical ventilation or reintubation, both of which are associated with adverse outcomes. Daily protocol-driven assessment of the need for mechanical ventilation leads to earlier extubation but requires dedicated personnel. We sought to determine whether machine learning applied to the electronic health record could predict successful extubation. Methods We examined 37 clinical features from patients from a single-center prospective cohort study of patients in our quaternary care medical ICU who required mechanical ventilation and underwent a bronchoalveolar lavage for known or suspected pneumonia. We also tested our models on an external test set from a community hospital ICU in our health care system. We curated electronic health record data aggregated from midnight to 8AM and labeled extubation status. We deployed three data encoding/imputation strategies and built XGBoost, LightGBM, logistic regression, LSTM, and RNN models to predict successful next-day extubation. We evaluated each model’s performance using Area Under the Receiver Operating Characteristic (AUROC), Area Under the Precision Recall Curve (AUPRC), Sensitivity (Recall), Specificity, PPV (Precision), Accuracy, and F1-Score. Results Our internal cohort included 696 patients and 9,828 ICU days, and our external cohort had 333 patients and 2,835 ICU days. The best model (LSTM) predicted successful extubation on a given ICU day with an AUROC 0.87 (95% CI 0.834-0.902) and the internal test set and 0.87 (95% CI 0.848-0.885) on the external test set. A Logistic Regression model performed similarly (AUROC 0.86 internal test, 0.83 external test). Across multiple model types, measures previously demonstrated to be important in determining readiness for extubation were found to be most informative, including plateau pressure and Richmond Agitation Sedation Scale (RASS) score. Our model often predicted patients to be stable for extubation in the days preceding their actual extubation, with 63.8% of predicted extubations occurring within three days of true extubation. We also tested the best model on cases of failed extubations (requiring reintubation within two days) not seen by the model during training. Our best model would have identified 35.4% (17/48) of these cases in the internal test set and 48.1% (13/27) cases in the external test set as unlikely to be successfully extubated. Conclusions Machine learning models can accurately predict the likelihood of extubation on a given ICU day from data available in the electronic health record. Predictions from these models are driven by clinical features that have been associated with successful extubation in clinical trials.


  • Open Access

    Neutrophil percentages in bronchoalveolar lavage fluid: Implications for diagnosing bacterial pneumonia in patients with immunocompromise and neutropenia

    by Kevin M. Grudzinski, Samuel Fenske, Alec Peltekian, Nikolay S. Markov, Anna Pawlowski, Mengjia Kang, James M. Walter, Chiagozie I. Pickens, Nandita R. Nadig, Ankit Agrawal, et al.

    Background Pneumonia is the leading cause of infectious deaths and the most common infection identified in ICU patients. Assessment of bronchoalveolar lavage fluid (BALF) cellularity can aid in pneumonia diagnosis. Low percentages (Methods We analyzed a large cohort of BALF specimens obtained for routine care for suspected pneumonia in mechanically ventilated patients and enrolled in the single-center Successful Clinical Response In Pneumonia Therapy (SCRIPT) study. BALF neutrophils were reported as a percentage of leukocytes by the clinical laboratory. The etiology of each episode of suspected pneumonia was adjudicated by a committee of critical care physicians using a predefined protocol. Immunocompromise was defined using predetermined criteria by the study research team. Neutropenia was defined here as a peripheral ANC Results 688 mechanically ventilated patients with suspected pneumonia were included. 409 (59.4%) were male; median age was 62 [51,71]. 461 patients (67.0%) were immunocompetent, 149 (21.7%) were immunocompromised without neutropenia and 78 (11.3%) were neutropenic at some point during their admission. A total of 1746 BALs were performed. Fifty-seven BALs were obtained on a day where the patient’s ANCConclusion Among patients with bacterial pneumonia, BALF neutrophil percentage was not significantly decreased by a spectrum of immunocompromise. However, the subset of patients who were acutely neutropenic at the time of BAL sampling had significantly lower BALF % neutrophils. A traditional approach using BALF<50% to suggest against bacterial pneumonia may be inaccurate in this particular population.