About

Utkan Demirci is a Professor of Radiology at Stanford University and is affiliated with the Canary Cancer Center. He is also an affiliated faculty member in Electrical Engineering. His work is centered on the intersection of artificial intelligence, medicine, and imaging, contributing to the development of innovative solutions in healthcare through advanced research in these fields. As a leader within the Center for Artificial Intelligence in Medicine & Imaging (AIMI), he is involved in advancing research, education, and collaboration efforts aimed at improving medical diagnostics and treatment through technological innovation.

Research topics

• Computer Science
• Materials science
• Biology
• Nanotechnology
• Bioinformatics
• Medicine
• Cell biology
• Pathology
• Chemistry
• Computational biology
• Oncology
• Optoelectronics
• Electronic engineering
• Cancer research
• Immunology
• Internal medicine

Selected publications

• Glio-SERS: Label-Free Molecular Profiling of Plasma Extracellular Vesicles in Brain Tumors Using SERS and Artificial Intelligence

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-04-23

  articleOpen accessSenior authorCorresponding

  Abstract Extracellular vesicles are increasingly recognized as important carriers of disease-associated molecular information, yet robust methods for their isolation and molecular characterization from limited clinical samples remain challenging. Here, we present an integrated approach combining standardized EV isolation, label-free Surface-Enhanced Raman Spectroscopy (SERS), and artificial intelligence (AI) for comprehensive molecular profiling of small extracellular vesicles (sEVs) from human plasma. Here, we show systematically isolated and characterized plasma sEVs using ExoTIC in accordance with MISEV2023 guidelines, with SERS analysis revealing quantifiable spectral differences across samples from patients with glioblastoma (n=20) and meningioma (n=23) compared to healthy controls (n=30). Among the evaluated AI models, the convolutional neural network most effectively captured group-level spectral differences in sEVs, achieving accuracies up to 88% in this pilot cohort. Further, an EGFR-based spectral regression model was explored to examine molecular variability across sEV samples. Parallel proteomic analysis presented statistically significant differences in several proteins elevated in glioblastoma or meningioma. This label-free, rapid approach provides a proof-of-concept framework for sEV molecular profiling establishing the basis for broad validation studies across diverse diseases.

  PublisherOA PDFDOI

• Early detection of pancreatic cancer by a high-throughput protease-activated nanosensor assay

  Science Translational Medicine · 2025-02-12 · 34 citations

  article

  Pancreatic ductal adenocarcinoma (PDAC) is among the top causes of cancer-related death. Patients are frequently diagnosed in the more advanced stages when effective treatment options are limited; however, earlier detection of PDAC by liquid biopsy may expand treatment options and improve survival outcomes. Here, we developed a noninvasive detection assay for PDAC based on serum protease activity to leverage the increase in cancer-associated protease activity in the peripheral blood of patients with PDAC. We screened a series of protease-cleavable peptide probes for the discrimination of PDAC samples versus healthy controls and noncancerous pancreatic disease. We identified a single MMP-sensitive probe, which could distinguish PDAC from controls with 79 ± 6% accuracy. We further developed this probe into a rapid magnetic nanosensor assay, termed PAC-MANN, that measures serum protease cleavage of a target-probe nanosensor with a simple fluorescent readout. In a longitudinal cohort of patients undergoing surgical removal of the primary tumor, the probe cleavage signal was reduced by 16 ± 24% after surgery. In a separate blinded retrospective study, the PAC-MANN assay identified PDAC samples with 98% specificity and 73% sensitivity across all stages and distinguished 100% of patients with noncancer pancreatic disease relative to patients with PDAC. The PAC-MANN assay combined with the clinical biomarker CA 19-9 was 85% sensitive for detection of stage I PDAC with 96% specificity. Therefore, the PAC-MANN assay is a rapid, high-throughput method that uses small blood volumes with the potential to enhance early PDAC detection, specifically among individuals at high risk of developing PDAC.

  PublisherDOI

• Circulating extracellular vesicles in serum carry Trop2 marker for prostate cancer liquid biopsy and clinical care

  medRxiv · 2025-04-04

  preprintOpen accessSenior authorCorresponding

  ABSTRACT Extracellular vesicles (EVs) are lipid nano-to-micro-sized vesicles increasingly identified as valuable liquid biopsy tools for medical applications. However, the heterogeneity of cargo and the lack of convenient quantification methods to characterize EVs pose challenges in identifying vesicles with specific markers. In this study, we show the isolation, characterization, detection, and quantification of a cancer-specific marker, Trop2, on circulating extracellular vesicles in serum (EV-Trop2). This work combines the unique advantages of our user-friendly isolation method with serum diagnostics to identify high-risk prostate cancer cases and predict recurrence after prostate surgery. To our knowledge, this is the first demonstration to isolate and quantify EV-Trop2 from prostate cancer patient serum to study its analytical validity and potential clinical utility as an EV-based liquid biopsy. Initial study with patient serum samples from three clinical groups: high- risk prostate cancer (n = 22), low-risk prostate cancer (n = 23), and cancer-free groups (n = 21), demonstrates the potential of this approach in distinguishing prostate cancer aggressiveness. We observed significantly different levels of EV-Trop2 expression between the high-risk and low-risk patient groups (p = 0.0015), and between high-risk patient and cancer-free groups (p < 0.0001). Furthermore, employing machine learning algorithms, EV-Trop2 was shown to enhance classifier metrics across the three sample groups, aiding both in risk stratification and predicting recurrence post-prostatectomy. The availability of such tool could have a broad impact across multiple cancers by enabling minimally invasive liquid biopsy sampling. Abstract Figure

  PublisherOA PDFDOI

• Engineering and biofabrication of early cancer models

  Nature Reviews Bioengineering · 2025-11-03 · 2 citations

  article
  PublisherDOI

• Plasma Preparation Strategies for Extracellular Vesicle‐Based Biomarkers in Metastatic Castration‐Resistant Prostate Cancer

  Journal of Extracellular Biology · 2025-07-31 · 5 citations

  articleOpen accessSenior authorCorresponding

  ABSTRACT Extracellular vesicles (EVs) offer a minimally invasive approach for cancer detection and monitoring. However, the lack of standardized methods for clinical biospecimen preparation and EV isolation limits the clinical utility of EV‐based biomarker assessments. A targeted need exists for detailed analysis of plasma EV content. Our study investigates the impact of clinical sample preparation and our ExoTIC device on the quality of plasma‐derived EVs and their RNA/protein cargo in metastatic castration‐resistant prostate cancer (mCRPC) patients. We assessed sample preparation variables: blood anti‐coagulant choice (EDTA or sodium citrate), type of plasma platelet fraction (platelet‐rich or platelet‐poor), and use of protease inhibitors. EVs were isolated via ExoTIC device, followed by EV characterization and biomarker analysis using nanoparticle tracking analysis (NTA), cryogenic electron microscopy, Western blot, and digital PCR (dPCR). We detected mCRPC‐relevant proteins (ARv7 and PSMA) in EVs from all plasma sample types with different sample preparation variables. Additionally, our findings indicate that platelet‐poor plasma (PPP) is optimal for detecting EV‐ and biologically associated mCRPC biomarker miR‐375. In this pilot study ( n = 3), elevated EV miR‐375 levels in PPP samples from mCRPC patients experiencing disease progression during docetaxel treatment were associated with poor therapeutic response to docetaxel chemotherapy, which aligns with our preceding in vitro and in vivo study. Optimal biospecimen preparation for EV analysis could enhance detection accuracy and patient management, highlighting detection of plasma EV‐associated mCRPC‐specific marker proteins (ARv7 and PSMA) and microRNA miR‐375.

  PublisherOA PDFDOI

• Screw‐Based Pill for Intelligent Robotic Extraction of Viscous Fluids in Medical Applications

  Advanced Intelligent Systems · 2025-07-13

  articleOpen accessSenior authorCorresponding

  Smart capsules are promising tools for minimally invasive sampling. However, existing designs rely on passive diffusion, which is ineffective for viscous samples such as mucus. To address this limitation, we present screw‐based pill for intelligent robotic extraction (S‐PIRE), designed for active sampling of viscous materials. S‐PIRE features a motorized 3D‐printed hydrodynamic screw, remotely activated via a Bluetooth microcontroller. Its magnetic actuation system ensures controlled positioning, while its integrated 3‐axis Hall Effect magnetic sensor provides spatial localization. S‐PIRE is magnetically guided for collecting viscous samples in vitro to demonstrate its sampling mechanism. Samples are securely stored in a detachable, disposable chamber for analysis. This proof of concept demonstrates S‐PIRE potential for in vivo applications, offering a precise, minimally invasive approach for future targeted mucus sampling from difficult‐to‐reach regions of the body like the gastrointestinal tract, which could aid in early disease detection and biomarker discovery.

  PublisherOA PDFDOI

• A biosensor-integrated filtration device for nanoparticle isolation and label-free imaging

  Lab on a Chip · 2025-01-01 · 7 citations

  articleOpen accessCorresponding

  Rapid, efficient, simple approaches for biological nanoparticle recovery from bodily fluids are required for translating detection strategies from lab diagnostics to low-resource settings, where expensive sample processing instruments such as an ultracentrifuge are not accessible. In this work, we characterize an alternative approach in which intact nanoparticles are filtered from plasma with a nanoporous filtration device that separates particulates within a 100-200 nm diameter range followed by detection on a photonic crystal (PC) biosensor with a portable photonic resonator interferometric scattering microscopy (PRISM) instrument. The biosensor-integrated recovery device's (BIRD) collection efficiency is initially characterized using gold nanoparticles and fluorescent nanobeads suspended in buffer solution and plasma, followed by spiking intact HIV pseudovirus into the same media. We demonstrate a recovery rate of 55.0% for 100 nm diameter AuNP and HIV spiked into the buffer and 11.9% for 100 nm diameter FluoSpheres spiked in human plasma. Using PRISM, we observed the Brownian motion of filtered nanoparticles and virions eluted into the detection compartment, with concentration-dependent counting of transient contact events between the nanoparticles and the PC surface.

  PublisherOA PDFDOI

• Dynamically reconfigurable acoustofluidic metasurface for subwavelength particle manipulation and assembly

  Nature Communications · 2025-01-15 · 26 citations

  articleOpen accessSenior author

  Particle manipulation plays a pivotal role in scientific and technological domains such as materials science, physics, and the life sciences. Here, we present a dynamically reconfigurable acoustofluidic metasurface that enables precise trapping and positioning of microscale particles in fluidic environments. By harnessing acoustic-structure interaction in a passive membrane resonator array, we generate localized standing acoustic waves that can be reconfigured in real-time. The resulting radiation force allows for subwavelength manipulation and patterning of particles on the metasurface at individual and collective scales, with actuation frequencies below 2 MHz. We further demonstrate the capabilities of the reconfigurable metasurface in trapping and enriching beads and biological cells flowing in microfluidic channels, showcasing its potential in high-throughput bioanalytical applications. Our versatile and biocompatible particle manipulation platform is suitable for applications ranging from the assembly of colloidal particles to enrichment of rare cells.

  PublisherOA PDFDOI

• A microneedle device for rapid dermal interstitial fluid sampling

  Science Advances · 2025-09-24 · 4 citations

  articleOpen access

  Dermal interstitial fluid (ISF) offers a promising alternative to invasive blood tests and opportunities for skin diagnostics. Progress in both the understanding and adoption of ISF tests is hindered by sampling challenges, including lengthy collection times, non-negligible failure rates, variable collection volumes, and inconsistent bioanalyte levels. The causes of many of these issues are not well understood. We demonstrate a microneedle device that is several times faster than state of the art, collecting an average of 15.5 mg of ISF in 5 minutes in humans with near-zero failure rate. This improvement was achieved by designing the spatial pressure gradient driving ISF flow. The influence of penetration depth, collection time, pressure, and age on ISF collection was elucidated, with Darcy's law explaining multiple observations. A data-driven acceptance criterion of <1% blood contamination for ISF is proposed. The device and findings presented will empower researchers to better conduct robust studies in the development of ISF diagnostics.

  PublisherDOI

• Extracellular Vesicles in Serum Carry Trop2 Protein as a Potential Molecular Indicator in Prostate Cancer

  Journal of Extracellular Biology · 2025-09-01 · 1 citations

  articleOpen accessSenior authorCorresponding

  ABSTRACT Extracellular vesicles (EVs) are lipid nano‐to‐micro‐sized vesicles increasingly studied for their role in intercellular communication and their potential as minimally invasive molecular indicators in various diseases. However, challenges remain in characterizing specific surface molecules on EVs due to cargo heterogeneity and the lack of convenient quantification methods. In this study, we show the isolation, characterization, detection, and quantification of Trop2‐carrying EVs (EV‐Trop2) in serum of prostate cancer patients. This work combines the unique advantages of our EV isolation method with ELISA to enable surface‐protein‐specific EV analysis directly from serum. This is, to our knowledge, the first demonstration to isolate and quantify EV‐Trop2 from prostate cancer patient serum to study its expression patterns in relation to prostate cancer status. Analysis of serum samples from three patient groups: high‐risk prostate cancer ( n = 22), low‐risk prostate cancer ( n = 23), and cancer‐free groups ( n = 21), revealed significantly different levels of EV‐Trop2 expression between the high‐risk and low‐risk patient groups ( p = 0.0015) and between high‐risk patient and cancer‐free groups ( p &lt; 0.0001). Multivariate modeling further showed that EV‐Trop2 contributed to improved classifier metrics across the three sample groups. These findings highlight a strategy for probing EV‐associated surface targets and suggest broader applicability of this approach across multiple cancers.

  PublisherOA PDFDOI

Recent grants

• NIH Grant R21HL112114

  NIH · $452k · 2014

• CAREER: Non-invasive fields for directed 3D microgel assembly for tissue engineering

  NSF · $64k · 2014–2017

• NIH Grant R21AI110277

  NIH · $474k · 2017

• NIH Grant R21HL095960

  NIH · $491k · 2014

• Exosome separation and digital resolution detection of blood-based nucleic acid biomarkers for noninvasive therapeutic diagnostics in cancer

  NIH · $556k · 2020–2024

Frequent coauthors

• Fatih İnci

  Bilkent University

  190 shared

• Umut A. Gürkan

  Shaker Heights Public Library

  181 shared

• Ali Khademhosseini

  Terasaki Foundation

  146 shared

• Feng Xu

  126 shared

• Sangjun Moon

  115 shared

• Savaş Taşoğlu

  Boğaziçi University

  96 shared

• Shuqi Wang

  Henan Agricultural University

  94 shared

• Hadi Shafiee

  90 shared