FLGS-04. Fluorescence lifetime imaging (FLIm) is a dye-free, high sensitivity approach for fluorescence guided surgery in high-grade and low-grade gliomas

INTRODUCTION Fluorescence-guided surgery can improve tumor identification and extent of surgical resection. 5-ALA is the standard for GBM, but is limited by lack of quantitative fluorescence, a need to work in the dark, and a lack of sensitivity for low grade tumors. We have developed a novel instrument for dye-free tissue autofluorescence lifetime imaging (FLIm) to identify glioma tissue during resection. This approach utilizes time-resolved autofluorescence measurements in narrow-band channels to assess markers of tissue metabolism. Compared to intensity-based imaging of exogenous fluorophores, FLIm has greater sensitivity without dependence on background lighting. The advantages of FLIm include quantitative tissue analysis, the ability to work under full light conditions, sensitivity for high and low grade gliomas, and the potential ability to identify IDH mutational status. In this study, we validated the use of FLIm for identification of glioma tissue at tumor resection margins. METHODS FLIm was used to image tissue margins during glioma resections and compared to microbiopsies from imaged regions to correlate fluorescence with histopathology. RESULTS FLIm was applied intraoperatively to 11 GBM and 5 LGG patients (7 imaged biopsies per patient). In GBM, tumor infiltration of cortex was associated with significantly decreased fluorescence lifetime (FL) in channels 2 (470/28nm;p<0.05) and 3 (542/50nm;p<0.002). In subcortical margins, FL was inversely proportional to the density of tumor in channels 2,3 (p<0.05). When IDH wild-type GBMs were compared to IDH1-mutant tumors, FL was noted to be significantly longer in channel 1 (390/40nm;p<0.05), and trended towards longer FL in channel 2, shorter FL in channel 3. In LGG, FL was significantly correlated with tumor density in channel 2 (p<0.01). CONCLUSIONS FLIm is a dye-free, quantitative alternative to 5-ALA for fluorescence guided glioma resections with sensitivity to high and low-grade tumors, and the ability to predict IDH mutations in GBM. Further validation studies are on-going.

Automated operative workflow analysis of endoscopic pituitary surgery using machine learning: development and preclinical evaluation (IDEAL stage 0)

OBJECTIVE Surgical workflow analysis involves systematically breaking down operations into key phases and steps. Automatic analysis of this workflow has potential uses for surgical training, preoperative planning, and outcome prediction. Recent advances in machine learning (ML) and computer vision have allowed accurate automated workflow analysis of operative videos. In this Idea, Development, Exploration, Assessment, Long-term study (IDEAL) stage 0 study, the authors sought to use Touch Surgery for the development and validation of an ML-powered analysis of phases and steps in the endoscopic transsphenoidal approach (eTSA) for pituitary adenoma resection, a first for neurosurgery. METHODS The surgical phases and steps of 50 anonymized eTSA operative videos were labeled by expert surgeons. Forty videos were used to train a combined convolutional and recurrent neural network model by Touch Surgery. Ten videos were used for model evaluation (accuracy, F1 score), comparing the phase and step recognition of surgeons to the automatic detection of the ML model. RESULTS The longest phase was the sellar phase (median 28 minutes), followed by the nasal phase (median 22 minutes) and the closure phase (median 14 minutes). The longest steps were step 5 (tumor identification and excision, median 17 minutes); step 3 (posterior septectomy and removal of sphenoid septations, median 14 minutes); and step 4 (anterior sellar wall removal, median 10 minutes). There were substantial variations within the recorded procedures in terms of video appearances, step duration, and step order, with only 50% of videos containing all 7 steps performed sequentially in numerical order. Despite this, the model was able to output accurate recognition of surgical phases (91% accuracy, 90% F1 score) and steps (76% accuracy, 75% F1 score). CONCLUSIONS In this IDEAL stage 0 study, ML techniques have been developed to automatically analyze operative videos of eTSA pituitary surgery. This technology has previously been shown to be acceptable to neurosurgical teams and patients. ML-based surgical workflow analysis has numerous potential uses—such as education (e.g., automatic indexing of contemporary operative videos for teaching), improved operative efficiency (e.g., orchestrating the entire surgical team to a common workflow), and improved patient outcomes (e.g., comparison of surgical techniques or early detection of adverse events). Future directions include the real-time integration of Touch Surgery into the live operative environment as an IDEAL stage 1 (first-in-human) study, and further development of underpinning ML models using larger data sets.

Brain Tumor Identification and Locating

Various poorer-grade glioblastoma subtypes related with outline attribute have been identified in a late research. poorer-grade glioblastoma can have symptoms ranging from slight seizures to extensive seizures, affecting the capability to talk or even lift your arms and legs. This poorer-grade glioblastoma is handled with a combination of surgery and examination by monitoring the tumor with brain MRI scans. The study of this project enabled us to build a completely self-working system for segmentation of tumor utilizing computer vision techniques, and deploying models that would allow high- quality LGG detection in the brain MRI would potentially be self-working to identify the genomic subtype of the tumor by quick and low-cost imaging. The methodology, procedures, pros and their boundaries and their future ultimatums are discussed in this project.

Molecular persistent luminescence imaging with porphyrin derivatives for ultrasensitive image-guided cancer surgery and drug screening

Persistent luminescence without excitation light and tissue autofluorescence interference holds great promise for in vivo imaging and sensing. However, the availability of persistence luminescence materials is largely limited by potential toxicity, instability, short-wavelength emissions, and poor clinical potential for currently available ones. Here we report a series of porphyrin derivatives with near-infrared (NIR) persistence luminescence for image-guided cancer surgery and drug screening. These porphyrin derivatives showed NIR persistence luminescence over 760 nm after cessation of excitation light or upon interaction with peroxynitrite (ONOO-), and a plausible mechanism of ordered oxidation of vinylene bond is proposed. Through molecular engineering with adaptive peptides bearing the functions of β-sheet-formatting and cancer cell targeting, the resultant Ppa-FFGYSA supermolecular probe showed enhanced photoacoustic and persistence luminescence signals, facilitating preoperative photoacoustic tumor identification and intraoperative persistence luminescence image-guided tumor resection with outperformed signal-to-background ratio. In addition, the activated persistence luminescence in recognition of ONOO- also permits the specific monitoring of neutrophil infiltration and screening of immunogenic cell death (ICD) drugs with high sensitivity and specificity.

ANALYSIS OF TUMOR DETECTION USING UNETS AND VGG16 WEIGHTS

Tumor detection from Brain MRI images Abstract: Detecting tumors in the human brain has become the most challenging medical science issue. Recognition of tumors in MRIs is vital as it offers the aberrant relevant data for therapeutic interventions. MRI includes details on malignant tissue. An abnormal tissue growing and multiplying in the brain is a brain tumor. Physical examination is the standard approach for brain tumor identification, which takes much time and is not accurate every time. So, automated brain tumor identification methods are establishing to save time. Image segmentation utilizes to detect the brain's abnormal portion, which gives the tumor's location. This work uses the UNETS with VGG16 weights model to see and segment tumors from the rest of the brain tissue. The accurate detection of the tumors helps reduce the delay between diagnostic testing and therapy. Therefore, there is a significant demand for computer algorithms to be precise, speedy, time-efficient, and dependable. The technology described relates to detecting and analyzing brain cancers automatically via U-Net and the VGG16 CNN.

New Perspectives in Imaging of Uveal Melanoma

Uveal melanoma is the most common primary intraocular malignancy in adults characterized by insidious onset and poor prognosis strongly associated with tumor size and the presence of distant metastases, most commonly in the liver. Contrary to most tumor identification, biopsy followed by pathological exam is not recommended in ophthalmic oncology. Therefore, early and non-invasive diagnosis is essential to enhance patients’ chances for early treatment possibilities. We reviewed imaging modalities currently used in the diagnosis of uveal melanoma, i.e., fundus imaging, ultrasonography (US), optical coherence tomography (OCT), single-photon emission computed tomography (SPECT), positron emission tomography/computed tomography (PET/CT), magnetic resonance imaging (MRI), fundus fluorescein angiography (FFA), indocyanine green angiography (ICGA), fundus autofluorescence (FAF). The principle of imaging techniques was briefly explained, along with their role in the diagnostic process and a summary of their advantages and limitations. Further, the experimental data and the advancements in imaging modalities were searched. We described their innovations, showed current usage and research, and explained the possibilities of utilizing them to diagnose uveal melanoma and their potential application in personalized medicine such as theranostics.