Aydogan Ozcan’s research while affiliated with University of California System and other places

Imaging mass spectrometry (IMS) is a powerful tool for untargeted, highly multiplexed molecular mapping of tissue in biomedical research. IMS offers a means of mapping the spatial distributions of molecular species in biological tissue with unparalleled chemical specificity and sensitivity. However, most IMS platforms are not able to achieve microscopy-level spatial resolution and lack cellular morphological contrast, necessitating subsequent histochemical staining, microscopic imaging and advanced image registration steps to enable molecular distributions to be linked to specific tissue features and cell types. Here, we present a virtual histological staining approach that enhances spatial resolution and digitally introduces cellular morphological contrast into mass spectrometry images of label-free human tissue using a diffusion model. Blind testing on human kidney tissue demonstrated that the virtually stained images of label-free samples closely match their histochemically stained counterparts (with Periodic Acid-Schiff staining), showing high concordance in identifying key renal pathology structures despite utilizing IMS data with 10-fold larger pixel size. Additionally, our approach employs an optimized noise sampling technique during the diffusion model's inference process to reduce variance in the generated images, yielding reliable and repeatable virtual staining. We believe this virtual staining method will significantly expand the applicability of IMS in life sciences and open new avenues for mass spectrometry-based biomedical research.

Noninvasive optical imaging modalities can probe patient's tissue in 3D and over time generate gigabytes of clinically relevant data per sample. There is a need for AI models to analyze this data and assist clinical workflow. The lack of expert labelers and the large dataset required (>100,000 images) for model training and tuning are the main hurdles in creating foundation models. In this paper we introduce FoundationShift, a method to apply any AI model from computational pathology without retraining. We show our method is more accurate than state of the art models (SAM, MedSAM, SAM-Med2D, CellProfiler, Hover-Net, PLIP, UNI and ChatGPT), with multiple imaging modalities (OCT and RCM). This is achieved without the need for model retraining or fine-tuning. Applying our method to noninvasive in vivo images could enable physicians to readily incorporate optical imaging modalities into their clinical practice, providing real time tissue analysis and improving patient care.

Chronic diseases, including diabetes, cardiovascular diseases, and microvascular complications, contribute significantly to global morbidity and mortality. Current monitoring tools such as glucometers and continuous glucose monitors only measure one analyte; multiplexing technologies offer a promising approach for monitoring multiple biomarkers, enabling the management of comorbidities and providing more comprehensive disease insights. In this work, we describe a miniaturized optical “barcode” sensor with high biocompatibility for the continuous monitoring of glucose and oxygen. This enzymatic sensor relies on oxygen consumption in proportion to local glucose levels and the phosphorescence reporting of tissue oxygen with a lifetime-based probe. The sensor was specifically designed to operate in a tissue environment with low levels of dissolved oxygen. The barcode sensor consists of a poly(ethylene glycol) diacrylate (PEGDA) hydrogel with four discrete compartments separately filled with glucose- or oxygen-sensing phosphorescent microparticles. We evaluated the response of the barcode hydrogels to fluctuating glucose levels over the physiological range under low oxygen conditions, demonstrating the controlled tuning of dynamic range and sensitivity. Moreover, the barcode sensor exhibited remarkable storage stability over 12 weeks, along with full reversibility and excellent reproducibility (∼6% variability in the phosphorescence lifetime) over nearly 50 devices. Electron beam sterilization had a negligible effect on the glucose response of the barcode sensors. Furthermore, our investigation revealed minimal phosphorescence lifetime changes in oxygen compartments while exhibiting increased lifetime in glucose-responsive compartments when subjected to alternating glucose concentrations (0 and 200 mg/dL), showcasing the sensor’s multianalyte sensing capabilities without crosstalk between compartments. Additionally, the evaluation of chronic tissue response to sensors inserted in pigs revealed the appropriate biocompatibility of the barcodes as well as excellent material stability over many months. These findings support further development of similar technologies for introducing optical assays for multiple biomarkers that can provide continuous or on-demand feedback to individuals to manage chronic conditions.

Virtual staining of tissue offers a powerful tool for transforming label-free microscopy images of unstained tissue into equivalents of histochemically stained samples. This study presents a diffusion model-based super-resolution virtual staining approach utilizing a Brownian bridge process to enhance both the spatial resolution and fidelity of label-free virtual tissue staining, addressing the limitations of traditional deep learning-based methods. Our approach integrates novel sampling techniques into a diffusion model-based image inference process to significantly reduce the variance in the generated virtually stained images, resulting in more stable and accurate outputs. Blindly applied to lower-resolution auto-fluorescence images of label-free human lung tissue samples, the diffusion-based super-resolution virtual staining model consistently outperformed conventional approaches in resolution, structural similarity and perceptual accuracy, successfully achieving a super-resolution factor of 4-5x, increasing the output space-bandwidth product by 16-25-fold compared to the input label-free microscopy images. Diffusion-based super-resolved virtual tissue staining not only improves resolution and image quality but also enhances the reliability of virtual staining without traditional chemical staining, offering significant potential for clinical diagnostics.

Generative models cover various application areas, including image, video and music synthesis, natural language processing, and molecular design, among many others. As digital generative models become larger, scalable inference in a fast and energy-efficient manner becomes a challenge. Here, we present optical generative models inspired by diffusion models, where a shallow and fast digital encoder first maps random noise into phase patterns that serve as optical generative seeds for a desired data distribution; a jointly-trained free-space-based reconfigurable decoder all-optically processes these generative seeds to create novel images (never seen before) following the target data distribution. Except for the illumination power and the random seed generation through a shallow encoder, these optical generative models do not consume computing power during the synthesis of novel images. We report the optical generation of monochrome and multi-color novel images of handwritten digits, fashion products, butterflies, and human faces, following the data distributions of MNIST, Fashion MNIST, Butterflies-100, and Celeb-A datasets, respectively, achieving an overall performance comparable to digital neural network-based generative models. To experimentally demonstrate optical generative models, we used visible light to generate, in a snapshot, novel images of handwritten digits and fashion products. These optical generative models might pave the way for energy-efficient, scalable and rapid inference tasks, further exploiting the potentials of optics and photonics for artificial intelligence-generated content.

We developed a rapid scanning optical microscope, termed "BlurryScope", that leverages continuous image acquisition and deep learning to provide a cost-effective and compact solution for automated inspection and analysis of tissue sections. BlurryScope integrates specialized hardware with a neural network-based model to quickly process motion-blurred histological images and perform automated pathology classification. This device offers comparable speed to commercial digital pathology scanners, but at a significantly lower price point and smaller size/weight, making it ideal for fast triaging in small clinics, as well as for resource-limited settings. To demonstrate the proof-of-concept of BlurryScope, we implemented automated classification of human epidermal growth factor receptor 2 (HER2) scores on immunohistochemically (IHC) stained breast tissue sections, achieving concordant results with those obtained from a high-end digital scanning microscope. We evaluated this approach by scanning HER2-stained tissue microarrays (TMAs) at a continuous speed of 5 mm/s, which introduces bidirectional motion blur artifacts. These compromised images were then used to train our network models. Using a test set of 284 unique patient cores, we achieved blind testing accuracies of 79.3% and 89.7% for 4-class (0, 1+, 2+, 3+) and 2-class (0/1+ , 2+/3+) HER2 score classification, respectively. BlurryScope automates the entire workflow, from image scanning to stitching and cropping of regions of interest, as well as HER2 score classification. We believe BlurryScope has the potential to enhance the current pathology infrastructure in resource-scarce environments, save diagnostician time and bolster cancer identification and classification across various clinical environments.

Background: The gold standard for crystal arthritis diagnosis relies on the identification of either monosodium urate (MSU) or calcium pyrophosphate (CPP) crystals in synovial fluid. With the goal of enhanced crystal detection, we adapted a standard compensated polarized light microscope (CPLM) with a polarized digital camera and multi-focal depth imaging capabilities to create digital images from synovial fluid mounted on microscope slides. Using this single-shot computational polarized light microscopy (SCPLM) method, we compared rates of crystal detection and raters’ preference for image. Methods: Microscope slides from patients with either CPP, MSU, or no crystals in synovial fluid were acquired using CPLM and SCPLM methodologies. Detection rate, sensitivity, and specificity were evaluated by presenting expert crystal raters with (randomly sorted) CPLM and SCPLM digital images, from FOV above clinical samples. For each FOV and each method, each rater was asked to identify crystal suspects and their level of certainty for each crystal suspect and crystal type (MSU vs. CPP). Results: For the 283 crystal suspects evaluated, SCPLM resulted in higher crystal detection rates than did CPLM, for both CPP (51%. vs. 28%) and MSU (78% vs. 46%) crystals. Similarly, sensitivity was greater for SCPLM for CPP (0.63 vs. 0.35) and MSU (0.88 vs. 0.52) without giving up much specificity resulting in higher AUC. Conclusions: Subjective and objective measures of greater detection and higher certainty were observed for SCPLM over CPLM, particularly for CPP crystals. The digital data associated with these images can ultimately be incorporated into an automated crystal detection system that provides a quantitative report on crystal count, size, and morphology.