About

Marina Marjanovic is an Associate Professor Emerita of Bioengineering at the University of Illinois Urbana-Champaign. She has held multiple positions within the university, including Teaching Associate Professor and Research Professor in the Department of Bioengineering since 2014 and 2021 respectively. She is also an Affiliate Faculty member at the Carle-Illinois College of Medicine and serves as the Associate Director of the Center for Optical Molecular Imaging at the Beckman Institute for Advanced Science and Technology. Her primary research area is biomedical imaging, with specific focus on biomedical imaging, live-cell imaging, optical diagnostics for cancer, and primary care imaging. She has contributed to the field through her work in biomedical imaging research, emphasizing optical techniques and live-cell imaging to advance diagnostics and understanding of health issues.

Research topics

• Computer science
• Medicine
• Chemistry
• Pathology
• Biomedical engineering

Selected publications

• Coherent Raman at GSK

  2026-03-05

  article
  PublisherDOI

• Spatial, metabolic, and chemical imaging for enhanced characterization and optimal selection of CHO cell lines via multimodal label-free microscopy

  2025-03-19

  article

  Developing highly productive Chinese Hamster Ovary (CHO) cell lines is critical yet resource-intensive in the biopharmaceutical industry. Multimodal label-free microscopy has potential to enhance cell line selection through integration of system that combines Simultaneous Label-free Autofluorescence Multi-harmonic (SLAM) microscopy, Fluorescence Lifetime Imaging Microscopy (FLIM), and Coherent Anti-Stokes Raman Scattering (CARS) microscopy. Each modality offers unique insights into cellular structure, metabolism, and biochemistry. SLAM provides structural and metabolic analysis, FLIM maps metabolic cofactors like NAD(P)H and FAD, and CARS reveals chemical information. Various CHO cell lines were imaged using this system to generate multichannel, co-registered images. Deep learning models were developed to interpret this high content information, classifying CHO cell lines that were imaged over two passages, improving characterization and selection for biopharmaceutical production.

  PublisherDOI

• Cold microplasma exposure as a novel therapeutic treatment: future implications for the management of bacterial otitis media

  2025-03-19

  article
  PublisherDOI

• Defining Biological Variability, Analytical Precision and Quantitative Biophysiochemical Characterization of Human Urinary Extracellular Vesicles

  Journal of Extracellular Vesicles · 2025-05-01 · 6 citations

  articleOpen access

  ABSTRACT The magnitude of combined analytical errors of urinary extracellular vesicle (uEV) preparation and measurement techniques (CV A ) has not been thoroughly investigated to determine whether it exceeds biological variations. We utilized technical replicates of human urine to assess the repeatability of uEV concentration and size measurements by nanoparticle tracking analysis (NTA) following differential velocity centrifugation (DC), silicon carbide, or polyethylene glycol uEV isolation methods. The DC method attained the highest precision. Consequently, DC‐derived uEV size, most abundant protein levels, and optical redox ratio (ORR) were further assessed by dynamic light scattering (DLS), immunoblotting or multi‐photon (SLAM) microscopy. Procedural errors primarily affected uEV counting and uEV‐associated protein quantification, while instrumental errors contributed most to the total variability of NTA‐ and DLS‐mediated uEV sizing processes. The intra‐individual variability (CV I ) of uEV counts assessed by NTA was smaller than inter‐individual variability (CV G ), resulting in an estimated index of individuality IOI < 0.6, suggesting that personalized reference interval (RI) is more suitable for interpretation of changes in subject's test results. Population‐based RI was more appropriate for ORR (IOI > 1.4). The analytical performance of DC‐NTA and DC‐SLAM techniques met optimal CV A < 0.5 × CV I criteria, indicating their suitability for further testing in clinical laboratory settings.

  PublisherDOI

• Cold microplasma exposure as a novel therapeutic treatment for bacterial otitis media: study results and conclusions in a small animal model

  2025-03-19

  article
  PublisherDOI

• Design and implementation of an advanced simultaneous label-free autofluorescence multiharmonic (SLAM) microscopy system for multimodal imaging (Conference Presentation)

  2025-03-20

  article

  Label-free nonlinear optical microscopy can be leveraged to extract structural, metabolic, and biochemical contrast from biological samples. Recently, simultaneous label-free autofluorescence multiharmonic (SLAM) microscopy has been demonstrated by using a shaped supercontinuum source to simultaneously excite second harmonic generation, third harmonic generation, two-photon fluorescence of FAD, and three-photon fluorescence of NAD(P)H. While SLAM microscopy provides four complementary channels simultaneously, analysis capabilities can be further improved by acquiring additional modalities. This presentation will provide details on the incorporation of additional modalities of fluorescence lifetime imaging microscopy (FLIM) and coherent anti-Stokes Raman scattering (CARS) into SLAM microscopy, describe the optimization procedure for excitation pulse shaping, and quantify the benefits provided by additional modalities and optimal excitation for image analysis and classification.

  PublisherDOI

• Hand-held probe integrating Raman spectroscopy and optical coherence tomography for in vivo characterization of otitis media

  2025-03-19

  article

  A handheld probe and imaging system integrating Raman spectroscopy with Optical Coherence Tomography (RS-OCT) has been developed for enhanced diagnosis of otitis media. This multimodal system combines the high-resolution structural imaging capabilities of OCT with the molecular specificity of RS, allowing for advanced characterization of bacterial species within middle ear effusions. The RS-OCT probe provides comprehensive diagnostic information, facilitating accurate and targeted treatment decisions. Designed for clinical use, the probe fits a standard otoscope head and speculum. This probe was integrated into a portable cart-based system and used for imaging participants in clinic prior to surgery. Initial studies highlight the potential of RS-OCT in improving otitis media diagnostics.

  PublisherDOI

• Accelerating biopharmaceutical cell line selection with label-free multimodal nonlinear optical microscopy and machine learning

  Communications Biology · 2025-02-03 · 4 citations

  articleOpen access

  The selection of high-performing cell lines is crucial for biopharmaceutical production but is often time-consuming and labor-intensive. We investigated label-free multimodal nonlinear optical microscopy for non-perturbative profiling of biopharmaceutical cell lines based on their intrinsic molecular contrast. Employing simultaneous label-free autofluorescence multiharmonic (SLAM) microscopy with fluorescence lifetime imaging microscopy (FLIM), we characterized Chinese hamster ovary (CHO) cell lines at early passages (0–2). A machine learning (ML)-assisted analysis pipeline leveraged high-dimensional information to classify single cells into their respective lines. Remarkably, the monoclonal cell line classifiers achieved balanced accuracies exceeding 96.8% as early as passage 2. Correlation features and FLIM modality played pivotal roles in early classification. This integrated optical bioimaging and machine learning approach presents a promising solution to expedite cell line selection process while ensuring identification of high-performing biopharmaceutical cell lines. The techniques have potential for broader single-cell characterization applications in stem cell research, immunology, cancer biology and beyond. Label-free multimodal nonlinear optical microscopy and machine learning enable early, non-perturbative classification of biopharmaceutical cell lines, accelerating cell line selection processes and opening avenues for broader single-cell studies.

  PublisherOA PDFDOI

• Data Science Meets Fragrance: Analyzing User Reviews to Decode Emotional Connections to Perfume Notes

  2025-01-01

  articleOpen accessSenior author

  This paper explores the emotional connections associated with perfumes by analyzing user reviews and fragrance notes for each product.Using a public dataset sourced from the Fragnatica platform, the study applies sentiment analysis techniques to categorize perfumes into six essential emotional groups: Romantic, Energizing, Melancholic, Aggressive, Relaxing, and Neutral.Sentiment analysis models, like VADER, are employed for basic sentiment scoring, while more advanced models including fine-tuned DistilBERT are incorporated to detect nuanced emotions.The emotional tones extracted from user-generated text correlate with consumer ratings and perfume characteristics.The study also investigates the relationship between fragrance notes and user emotions, identifying specific scent profiles that strongly relate to each group.Methodologies applied include sentiment analysis, clustering, and statistical visualization, utilizing a substantial dataset of perfume reviews.These strategies uncover patterns in emotional responses to scent, providing insights into how fragrance compositions influence emotional perceptions.The results bridge the gap between subjective fragrance experiences and objective data analytics, enabling more refined product categorization.Ultimately, this study offers valuable implications for the fragrance industry, helping brands improve product development and marketing strategies by better understanding the emotional resonance, leading to enhanced customer satisfaction and targeted product offerings.

  PublisherOA PDFDOI

• The Power of Words: Leveraging Deep Learning Techniques to Predict Hotel Ratings from User Reviews

  2025-03-19

  articleOpen accessSenior author

  Online reviews represent a major source of information for evaluating customer experience and supporting decision making in the hospitality industry, yet rating prediction from review content remains challenging because review text is often short, noisy, and internally inconsistent. This study presents a deep learning framework for predicting hotel ratings from guest reviews while explicitly addressing data quality before model training. Data reliability is treated as a central modeling concern. The proposed methodology combines review titles, review texts, and associated tags with a structured preprocessing pipeline that incorporates sentiment inconsistency detection, textual similarity analysis, deviation analysis based on correlation, and reviewer behavior profiling to identify unreliable observations. On the filtered corpus, we evaluate multiple predictive architectures, including LSTM, Bidirectional LSTM variants, and DistilBERT, for review-level rating prediction, and we further examine hotel-level temporal forecasting through aggregated historical review signals over a 30-day horizon. The results indicate that model performance depends strongly on both data reliability and architectural choice. Among recurrent models, BiLSTM with self-attention achieves the best performance, while DistilBERT yields the strongest overall results. Ablation analysis confirms that the full preprocessing pipeline consistently improves prediction quality, and the forecasting experiments indicate that aggregated review features contain useful information for short-term hotel rating dynamics. The study contributes a systematic and practically relevant framework for rating prediction and hospitality analytics in support of reputation management.

  PublisherOA PDFDOI

Frequent coauthors

• Stephen A. Boppart

  University of Illinois System

  222 shared

• Eric J. Chaney

  University of Illinois Urbana-Champaign

  116 shared

• Aneesh Alex

  GlaxoSmithKline (United States)

  67 shared

• Darold R. Spillman

  University of Illinois Urbana-Champaign

  65 shared

• Steve R. Hood

  University of Illinois Urbana-Champaign

  55 shared

• Steven G. Adie

  44 shared

• Haohua Tu

  41 shared

• Janet E. Sorrells

  University of Illinois Urbana-Champaign

  35 shared

Labs

• Everitt LabPI

Awards & honors

• JenLab Young Investigator Award
• 2022 McGinnis Medical Innovation Graduate Student Fellowship