This conference presentation was prepared for the Molecular-Guided Surgery: Molecules, Devices, and Applications IX conference at SPIE BiOS, SPIE Photonics West 2023.

Organ perfusion has been of interest to surgeons unremittingly as it is generally understood that adequate tissue perfusion prevents morbidity and mortality. A promising objective imaging method that can provide objective and reproducible perfusion imaging is called laser speckle contrast imaging. For this study, a dye-free, instantaneous, continuous and real-time laparoscopic perfusion imaging device called PerfusiX-Imaging was developed. The technology was validated in multiple pre-clinical models as well as a multi-centre study in colorectal surgery. The basic and fundamental (pre-)clinical studies provide evidence that laparoscopic laser speckle contrast imaging is capable of measuring the slightest of perfusion differences in real-time.

With the rise in minimally invasive surgery and machine learning, there are emerging opportunities to improve patient outcomes with endoscopic techniques that quantify tissue shape and optical properties. We introduce a speckle-illumination stereo endoscope (SSE) that utilizes structured illumination to enhance both depth and optical property mapping. An SSE prototype was constructed and applied to fresh pig colon samples. SSE-estimated depth and optical property maps compare favorably to gold standard techniques. Requiring only minor modifications to existing commercial stereoscopes, the SSE could provide surgeons with improved visual depth perception and maps of biomarkers in vivo.

Simulations are indispensable in the field of biomedical optical imaging, particularly in functional imaging. Given the recent rise of artificial intelligence and the lack of labeled in vivo data, synthetic data is not only important for the validation of algorithms but also crucial for training machine learning methods. To support research based on synthetic data, we present a new framework for assessing the quality of synthetic spectral data. Experiments with more than 10,000 hyperspectral in vivo images obtained from multiple species and various organ classes indicate that our framework could become an important tool for researchers working with simulations.

Standard-of-care endoscopy and laparoscopy require multiple cameras to enable molecular imaging, which leads to challenges of 3D image registration and overlay. To address this, we are developing a targeted multispectral imaging (MSI) camera using custom multispectral filter arrays integrated onto image sensors. The design augments RGB-Bayer filters with sub-pixel narrowband filters, thereby maintaining white-light imaging through pixel binning while adding MSI contrast enhancement. Prototypes were tested by imaging tissue-mimicking phantoms containing blood of varied oxygenation, ICG dye and color charts. Future work will examine the translation potential for chip-on-tip endoscopy.

Negative surgical margins can be difficult to confirm intraoperatively. We propose a workflow of immunostaining and surface imaging of fresh excised tissue using highly sensitive and spectrally separable SERS nanoparticles as the targeted contrast agent. The adaptive focus capabilities of an advanced Raman instrument, combined with our rotational accessory tool for exposing each surface of the stained specimen to the objective lens, enables topographic mapping of the entire excised specimen’s surface. Detailed surface renderings color-encoded according to unmixed SERS nanoparticle abundances show a path forward for high-content, interactive surgical margin assessment.

Fluorescence-guided surgery (FGS) provides real-time visualisation of tumours with molecular specificity, but intensity-based measurement of fluorescence is prone to errors. Multispectral imaging (MSI) in the short-wave infrared (SWIR) has the potential to improve tumour delineation by enabling machine-learning-based classification of pixels based on their spectral characteristics. In this work, we demonstrate the ability of this approach to provide a robust method of visualizing tumour tissue during FGS.

Colorectal cancer (CRC) is the third most common and the second most deadly type of cancer worldwide. Currently, early-stage CRC detection is hindered by the limited information gold standard screening and diagnostic procedures (colonoscopies) provide on tissue structural changes. Accuracy of early CRC detection during colonoscopy and CRC delineation during microsurgery can potentially be increased by adding tissue molecular information in real-time. We developed a molecular-sensitive tool capable of determining depth-resolved fluorophore and chromophore concentrations, as well as scattering properties from wavelength-resolved fluorescence spectra alone. Normal mucosa and CRC exhibited differences in elastin, flavin, bile and lipid concentrations.

The current fluorescence image-guided surgery technique does not provide accurate estimates of target depth and transverse margins of fluorescing tumors. We adapted Spatial Frequency Domain Fluorescence Diffuse Optical Tomography to rapidly acquire the depth and transverse margins of fluorescent inclusions in a turbid media in two steps. First, we derive estimates of depth from normalized fluorescence responses to the spatially modulated light patterns. Second, using the estimated depth, we reconstruct the transverse margin in the target plane. We demonstrate the performance of our instrumentation and approach using a series of phantom experiments.

Fluorescence imaging during to oral cancer surgery is typically 2D, yielding limited information on tumor depth. Here, we continue the development of a spatial frequency domain imaging (SFDI) system for 3D fluorescence imaging. A deep convolutional neural network takes as inputs SFDI-computed absorption, scattering and spatial-frequency fluorescence images, and yields images of fluorescence concentration and tumour depth. The model is trained using in silico data from Monte Carlo simulations of geometric tumor shapes (e.g., cylinder, spherical harmonics). Initial results yield average depth errors of <0.1 mm. Experiments are conducted in agar phantoms based on patient imaging.

Recent advancements in micro- and nanofabrication have enabled new class of multispectral imaging devices that integrated imaging arrays with pixelated filters on the same substrate. These new spectral imaging devices can image multiple near infrared molecular probes in the near infrared spectrum. In this talk, we will discuss the details of these new imaging devices and present clinical data with folate targeted probes and proteases activated probes for imaging lung cancer.

Fluorescence surgical navigation helps surgeons to see molecular information about tissue near the surface and distinguish malignancies from healthy tissue. It is possible to label tumors with specific fluorescent molecules, but not to easily see into tissue beyond microscopy limits of near a millimeter. Furthermore, the molecular targeting generally suffers from poor tumor-to-background contrast due to non-specific signal from healthy tissue. Both of these challenges can be addressed by time-resolved, single-photon sensitive imaging utilizing emerging SPAD sensors. Here we review current achievements in 3D depth sensing and functional imaging in vivo, and provide guidance for further design of surgical SPAD cameras. We hypothesize that the utility of surgical guidance will grow with the combination of new fast sensors and functional fluorescence imaging methods that utilize existing FDA-approved labels.

This conference presentation was prepared for the Molecular-Guided Surgery: Molecules, Devices, and Applications IX conference at SPIE BiOS, SPIE Photonics West 2023.

While ICG-based NIR imaging has shown great potential in intraoperative surgery, there are two fundamental and unsolved problems facing medical imaging: 1) nonspecific uptake of intravenously administered diagnostic and/or therapeutic agents by normal tissues and organs and 2) incomplete elimination of unbound targeted agents from the body. These problems make image-guided cancer surgery extremely difficult because the background signal is high, and therefore the TBR is low. Designing a targeted contrast agent that shows fast clearance from the background tissues and eventually from the body after complete targeting is the key to the success of image-guided interventions. “Structure-Inherent Targeting” is a strategy that combines tissue-specific targeting components and imaging domain into a single molecule for targeting and imaging specific tissues in real-time, where the compact structural design enables the unbound contrast agent to be easily cleared from the body after targeting.

Fluorescence paired-agent imaging (PAI) is presented as a method to rapidly screen en face margins during Mohs surgery to reduce the time required for pathological assessment of intraoperative frozen tissue sections. PAI was applied to mouse models of squamous cell carcinoma, and positive tumor burden was detected using both the mean and maximum signal intensity. It was determined that PAI BP had higher tumor detection accuracy as compared to single-agent fluorescence, while the en face margins provided detailed visualization of the tumor burden. PAI is a promising methodology for rapidly screening positive margins during Mohs surgery.

Publisher's Note: This conference presentation, originally published on 17 March 2023, was withdrawn on 21 March 2023 per author request.

Pancreatic cancer and colon cancer metastasis are recalcitrant cancers that are often difficult to detect. Mucin 5AC (MUC5A) and Mucin 4 (MUC4) have been found to be overexpressed in pancreatic and colon cancers, respectively, while having minimal expression in normal tissue. Using Mucin antibodies conjugated to a fluorescent dye, we demonstrate their specific labeling of human derived pancreatic and colon cancers in both subcutaneous and orthotopic mouse models. Tumor-specific fluorescent antibodies are clinically promising tools for improving both oncologic resection and patient survival.

Nerve damage ruins the lives of many patients post surgery, significantly affecting post-surgical quality of life. Intraoperative nerve detection is completed using anatomical knowledge and conventional white light visualization when possible. However, nerves can be difficult or impossible to identify by white light visualization and neuroanatomy is often varied between patients. We have developed nerve specific fluorescence guided surgery (FGS) contrast agents that provide real time direct visualization of nerves intraoperatively. These nerve-specific fluorophores represent the first of their kind and are capable of translation to clinical studies using existing clinical infrastructure of FGS systems. Work is underway to complete the preclinical pharmacology and toxicology testing required for a successful investigational new drug application to the FDA for first-in-human clinical trials and translation to surgical use should be feasible within the next five years.

Iatrogenic nerve injury is a major source of morbidity common to all surgical specialties. Prostate cancer, the second leading cause of cancer-related death among men in the U.S, is often treated surgically via prostatectomy. But visibility of the nerve plexus is extremely limited and nerve damage affects 60% of patients leading to post-surgical comorbidities. We’ve developed a synthetic strategy to improve key properties of fluorophores with potential clinical translatability to generate an optimal 700 nm fluorophore to pair with a fluorescently labeled probe optimized for the 800 nm channel in FGS systems targeting PSMA via the EUK targeting sequence for use in two-color prostatectomy. These new water-soluble, NIR, nerve-specific fluorophores show improved nerve specificity and in vivo brightness, require a lower dose to achieve contrast of superficial and buried nerve tissue and negate formulation development, improving safety profiles and lowering the cost of clinical translation.

Cranial and spinal nerve repair occurs at a very slow rate, and in most cases the iatrogenic injury can’t be fully repaired, leading to permanent motor or sensory disabilities as well as incurable neuropathies. The visualization and evaluation of tumor-involved nerves is extremely difficult during minimally invasive surgical procedures such as through at the skull base. Recently, our group developed a library of nerve-specific near infrared (NIR) oxazine scaffold dyes that have high specificity for cranial nerves, and the ability to permeate the Blood-Brain Barrier (BBB), which resulted in different degrees of the obtained cranial nerves SBR. These cranial nerve-specific fluorophores will significantly improve nerve visualization at depth, enhancing the ability to visualize and evaluate buried and tumor-involved cranial nerves. This could significantly decrease post-surgical morbidity rates and could solve the unmet clinical need for an intraoperative tool that enhances visualization.

This conference presentation was prepared for the Molecular-Guided Surgery: Molecules, Devices, and Applications IX conference at SPIE BiOS, SPIE Photonics West 2023.

This conference presentation was prepared for the Molecular-Guided Surgery: Molecules, Devices, and Applications IX conference at SPIE BiOS, SPIE Photonics West 2023.

This conference presentation was prepared for the Molecular-Guided Surgery: Molecules, Devices, and Applications IX conference at SPIE BiOS, SPIE Photonics West 2023.