Semi-Automated Extraction of Lens Fragments via a Surgical Robot Using Semantic Segmentation of OCT Images with Deep Learning - Experimental Results in ex vivo Animal Model.

The overarching goal of this work is to demonstrate the feasibility of using optical coherence tomography (OCT) to guide a robotic system to extract lens fragments from ex vivo pig eyes. A convolutional neural network (CNN) was developed to semantically segment four intraocular structures (lens material, capsule, cornea, and iris) from OCT images. The neural network was trained on images from ten pig eyes, validated on images from eight different eyes, and tested on images from another ten eyes. This segmentation algorithm was incorporated into the Intraocular Robotic Interventional Surgical System (IRISS) to realize semi-automated detection and extraction of lens material. To demonstrate the system, the semi-automated detection and extraction task was performed on seven separate ex vivo pig eyes. The developed neural network exhibited 78.20% for the validation set and 83.89% for the test set in mean intersection over union metrics. Successful implementation and efficacy of the developed method were confirmed by comparing the preoperative and postoperative OCT volume scans from the seven experiments.

Free-Breathing Volumetric Liver R2* and Proton Density Fat Fraction Quantification in Pediatric Patients Using Stack-of-Radial MRI With Self-Gating Motion Compensation.

BACKGROUND: Stack-of-radial multiecho gradient-echo MRI is promising for free-breathing liver R2* quantification and may benefit children. PURPOSE: To validate stack-of-radial MRI with self-gating motion compensation in phantoms, and to evaluate it in children. STUDY TYPE: Prospective. PHANTOMS: Four vials with different R2* driven by a motion stage. SUBJECTS: Sixteen pediatric patients with suspected nonalcoholic fatty liver disease or steatohepatitis (five females, 13 ± 4 years, body mass index 29.2 ± 8.6 kg/m2 ). FIELD STRENGTH/SEQUENCES: Stack-of-radial, and 2D and 3D Cartesian multiecho gradient-echo sequences at 3T. ASSESSMENT: Ungated and gated stack-of-radial proton density fat fraction (PDFF) and R2* maps were reconstructed without and with self-gating motion compensation. Stack-of-radial R2* measurements of phantoms without and with motion were validated against reference 2D Cartesian results of phantoms without motion. In subjects, free-breathing stack-of-radial and reference breath-hold 3D Cartesian were acquired. Subject inclusion for statistical analysis and region of interest placement were determined independently by two observers. STATISTICAL TESTS: Phantom results were fitted with a weighted linear model. Demographic differences between excluded and included subjects were tested by multivariate analysis of variance. PDFF and R2* measurements were compared using Bland-Altman analysis. Interobserver agreement was assessed by the intraclass correlation coefficient (ICC). RESULTS: Ungated stack-of-radial R2* inside moving phantom vials showed a significant positive bias of 64.3 s-1 (P < 0.00001), unlike gated results (P > 0.31). Subject inclusion decisions for statistical analysis from two observers were consistent. No significant differences were found between four excluded and 12 included subjects (P = 0.14). Compared to breath-hold Cartesian, ungated and gated free-breathing stack-of-radial exhibited mean R2* differences of 18.5 s-1 and 3.6 s-1 . Mean PDFF differences were 1.1% and 1.0% for ungated and gated measurements, respectively. Interobserver agreement was excellent (ICC for PDFF = 0.99, ICC for R2* = 0.90; P < 0.0003). DATA CONCLUSION: Stack-of-radial MRI with self-gating motion compensation seems to allow free-breathing liver R2* and PDFF quantification in children. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 2.

Automatic needle tracking using Mask R-CNN for MRI-guided percutaneous interventions.

PURPOSE: Accurate needle tracking provides essential information for MRI-guided percutaneous interventions. Passive needle tracking using MR images is challenged by variations of the needle-induced signal void feature in different situations. This work aimed to develop an automatic needle tracking algorithm for MRI-guided interventions based on the Mask Region Proposal-Based Convolutional Neural Network (R-CNN). METHODS: Mask R-CNN was adapted and trained to segment the needle feature using 250 intra-procedural images from 85 MRI-guided prostate biopsy cases and 180 real-time images from MRI-guided needle insertion in ex vivo tissue. The segmentation masks were passed into the needle feature localization algorithm to extract the needle feature tip location and axis orientation. The proposed algorithm was tested using 208 intra-procedural images from 40 MRI-guided prostate biopsy cases, and 3 real-time MRI datasets in ex vivo tissue. The algorithm results were compared with human-annotated references. RESULTS: In prostate datasets, the proposed algorithm achieved needle feature tip localization error with median Euclidean distance (dxy) of 0.71 mm and median difference in axis orientation angle (dθ) of 1.28°, respectively. In 3 real-time MRI datasets, the proposed algorithm achieved consistent dynamic needle feature tracking performance with processing time of 75 ms/image: (a) median dxy = 0.90 mm, median dθ = 1.53°; (b) median dxy = 1.31 mm, median dθ = 1.9°; (c) median dxy = 1.09 mm, median dθ = 0.91°. CONCLUSIONS: The proposed algorithm using Mask R-CNN can accurately track the needle feature tip and axis on MR images from in vivo intra-procedural prostate biopsy cases and ex vivo real-time MRI experiments with a range of different conditions. The algorithm achieved pixel-level tracking accuracy in real time and has potential to assist MRI-guided percutaneous interventions.

MRI-guided targeted needle placement during motion using hydrostatic actuators.

BACKGROUND: Magnetic resonance imaging (MRI) has unique advantages for guiding interventions, but the narrow space is a major challenge. This study evaluates the feasibility of a remote-controlled hydrostatic actuator system for MRI-guided targeted needle placement. METHODS: The effects of the hydrostatic actuator system on MR image quality were evaluated. Using a reference step-and-shoot method (SS) and the proposed actuator-assisted method (AA), two operators performed MRI-guided needle placement in targets (n = 12) in a motion phantom. RESULTS: The hydrostatic actuator system exhibited negligible impact on MR image quality. In dynamic targets, AA was significantly more accurate and precise than SS, with mean ± SD needle-to-target error of 1.8 ± 1.0 mm (operator 1) and 1.3 ± 0.5 mm (operator 2). AA reduced the insertion time by 50% to 80% and total procedure time by 25%, compared to SS. CONCLUSIONS: The proposed hydrostatic actuator system may improve accuracy and reduce procedure time for MRI-guided targeted needle placement during motion.

Respiratory Motion Prediction Using Fusion-Based Multi-Rate Kalman Filtering and Real-Time Golden-Angle Radial MRI.

OBJECTIVE: Magnetic resonance imaging (MRI) can provide guidance for interventions in organs affected by respiration (e.g., liver). This study aims to: 1) investigate image-based and surrogate-based motion tracking methods using real-time golden-angle radial MRI; and 2) propose and evaluate a new fusion-based respiratory motion prediction framework with multi-rate Kalman filtering. METHODS: Images with different temporal footprints were reconstructed from the same golden-angle radial MRI data stream to simultaneously enable image-based and surrogate-based tracking at 10 Hz. A custom software pipeline was constructed to perform online tracking and calibrate tracking error and latency using a programmable motion phantom. A fusion-based motion prediction method was developed to combine the lower tracking error of image-based tracking with the lower latency of surrogate-based tracking. The fusion-based method was evaluated in retrospective studies using in vivo real-time free-breathing liver MRI. RESULTS: Phantom experiments confirmed that the median online tracking error of image-based tracking was lower than surrogate-based methods, however, with higher median system latency (350 ms vs. 150 ms). In retrospective in vivo studies, 75 respiratory waveforms of target features from eight subjects were analyzed. The median root-mean-squared prediction error (RMSE) for the fusion-based method (0.97 mm) was reduced (Wilcoxon signed rank test p < 0.05) compared to surrogate-based (1.18 mm) and image-based (1.3 mm) methods. CONCLUSION: The proposed fusion-based respiratory motion prediction framework using golden-angle radial MRI can achieve low-latency feedback with improved accuracy. SIGNIFICANCE: Respiratory motion prediction using the fusion-based method can overcome system latency to provide accurate feedback information for MRI-guided interventions.

A robotic system for telementoring and training in laparoscopic surgery.

A laparoscopic surgical training system, the LapaRobot, is introduced. The system is composed of an expert station and a trainee station connected through the Internet. Embedded actuators allow the trainee station to be driven by an expert surgeon so that a trainee learns proper technique through physical feedback. The surgical-tool trajectory and video feed can be recorded and later "played back" to a trainee to hone operative skills through guided repetition without the need for expert supervision. The system is designed to create a high-fidelity approximation of the intracorporeal workspace, incorporate commercially available surgical instruments, and provide a wealth of high-resolution data for quantitative analysis and feedback. Experimental evaluation demonstrated a 55% improvement in surgical performance with use of our system. In this paper, we introduce the details of the design and fabrication of the LapaRobot, illustrate the mechatronics and software-control schemes, and evaluate the system in a study.

Semiautomated optical coherence tomography-guided robotic surgery for porcine lens removal.

PURPOSE: To evaluate semiautomated surgical lens extraction procedures using the optical coherence tomography (OCT)-integrated Intraocular Robotic Interventional Surgical System. SETTING: Stein Eye Institute and Department of Mechanical and Aerospace Engineering, University of California, Los Angeles, USA. DESIGN: Experimental study. METHODS: Semiautomated lens extraction was performed on postmortem pig eyes using a robotic platform integrated with an OCT imaging system. Lens extraction was performed using a series of automated steps including robot-to-eye alignment, irrigation/aspiration (I/A) handpiece insertion, anatomic modeling, surgical path planning, and I/A handpiece navigation. Intraoperative surgical supervision and human intervention were enabled by real-time OCT image feedback to the surgeon via a graphical user interface. Manual preparation of the pig-eye models, including the corneal incision and capsulorhexis, was performed by a trained cataract surgeon before the semiautomated lens extraction procedures. A scoring system was used to assess surgical complications in a postoperative evaluation. RESULTS: Complete lens extraction was achieved in 25 of 30 eyes. In the remaining 5 eyes, small lens pieces (≤1.0 mm3) were detected near the lens equator, where transpupillary OCT could not image. No posterior capsule rupture or corneal leakage occurred. The mean surgical duration was 277 seconds ± 42 (SD). Based on a 3-point scale (0 = no damage), damage to the iris was 0.33 ± 0.20, damage to the cornea was 1.47 ± 0.20 (due to tissue dehydration), and stress at the incision was 0.97 ± 0.11. CONCLUSIONS: No posterior capsule rupture was reported. Complete lens removal was achieved in 25 trials without significant surgical complications. Refinements to the procedures are required before fully automated lens extraction can be realized.

Intraocular robotic interventional surgical system (IRISS): Semi-automated OCT-guided cataract removal.

BACKGROUND: With the development of laser-assisted platforms, the outcomes of cataract surgery have been improved by automating several procedures. The cataract-extraction step continues to be manually performed, but due to deficiencies in sensing capabilities, surgical complications such as posterior capsule rupture and incomplete cataract removal remain. METHODS: An optical coherence tomography (OCT) system is integrated into our intraocular robotic interventional surgical system (IRISS) robot. The OCT images are used for preoperative planning and intraoperative intervention in a series of automated procedures. Real-time intervention allows surgeons to evaluate the progress and override the operation. RESULTS: The developed system was validated by performing lens extraction on 30 postmortem pig eyes. Complete lens extraction was achieved on 25 eyes, and "almost complete" extraction was achieved on the remainder due to an inability to image small lens particles behind the iris. No capsule rupture was found. CONCLUSION: The IRISS successfully demonstrated semiautomated OCT-guided lens removal with real-time supervision and intervention.

Inhibition of PKC-Induced COX-2 and IL-8 Expression in Human Breast Cancer Cells by Glucosamine.

Breast cancer is a common cancer leading to many deaths among females. Cyclooxygenase-2 (COX-2) and interleukin-8 (IL-8) are two highly expressed inflammatory mediators to be induced by the protein kinase C (PKC) signaling via various inflammatory stimuli and both contribute significantly to cancer metastasis/progression. Glucosamine has been shown to act as an anti-inflammation molecule. The aim of this study was to clarify the role and acting mechanism of glucosamine during the PKC-regulation of COX-2/IL-8 expression and the associated impact on breast cancer. In MCF-7 breast cancer cells, glucosamine effectively suppresses the PKC induction of COX-2 and IL-8 promoter activity, mRNA and protein levels, as well as the production of prostaglandin E(2) (PGE(2)) and IL-8. Glucosamine is able to promote COX-2 protein degradation in a calpain-dependent manner and IL-8 protein degradation in calpain-dependent and proteasome-dependent manners. The MAPK and NF-κB pathways are involved in PKC-induced COX-2 expression, but only the NF-κB pathway is involved in PKC-induced IL-8 expression. Glucosamine attenuates PKC-mediated IκBα phosphorylation, nuclear NF-κB translocation, and NF-κB reporter activation. Both PGE(2) and IL-8 promote cell proliferation and IL-8 induces cell migration; thus, glucosamine appears to suppress PKC-induced cell proliferation and migration. Furthermore, glucosamine significantly inhibits the growth of breast cancer xenografts and this is accompanied by a reduction in COX-2 and IL-8 expression. In conclusion, glucosamine seems to attenuate the inflammatory response in vitro and in vivo and this occurs, at least in part by targeting to the NF-κB signaling pathway, resulting in an inhibition of breast cancer cell growth.

Downregulation of SUN2, a novel tumor suppressor, mediates miR-221/222-induced malignancy in central nervous system embryonal tumors.

Embryonal tumors of the central nervous system represent a highly malignant tumor group of medulloblastoma (MB), atypical teratoid/rhabdoid tumor (AT/RT) and primitive neuroectodermal tumor that frequently afflict children. AT/RT is often misdiagnosed as MB/primitive neuroectodermal tumor but with higher recurrence and lower survival rates. Pathogenesis of AT/RT is largely unknown. In this study, we report both the miRNome and transcriptome traits in AT/RT and MB by using small RNA sequencing and gene expression microarray analyses. Our findings demonstrate that the miR-221/222-encoded micro RNAs are abundantly expressed in AT/RT but not in MB, which contribute substantially to the malignancy of embryonal tumors. miR-221/222 targeted SUN2, a newly discovered tumor suppressor, directly to increase cell proliferation and tumor malignancy in vitro and in vivo. Immunohistochemistry against SUN2 in a tissue microarray of 33 AT/RT and 154 MB tumor specimens also detected less SUN2 protein in AT/RT. Collectively, this study uncovers a novel tumor suppressor, SUN2, plays a critical role in miR-221/222-mediated AT/RT malignancy as well as supports miR-221/222 and SUN2 represent new promising targets for more active therapies in AT/RT. In addition, our miRNome and transcriptome data also provide a roadmap for further embryonal tumor research.

Physicochemical and biological properties of biomimetic mineralo-protein nanoparticles formed spontaneously in biological fluids.

Recent studies indicate that mineral nanoparticles (NPs) form spontaneously in human body fluids. These biological NPs represent mineral precursors that are associated with ectopic calcifications seen in various human diseases. However, the parameters that control the formation of mineral NPs and their possible effects on human cells remain poorly understood. Here a nanomaterial approach to study the formation of biomimetic calcium phosphate NPs comparable to their physiological counterparts is described. Particle sizing using dynamic light scattering reveals that serum and ion concentrations within the physiological range yield NPs below 100 nm in diameter. While the particles are phagocytosed by macrophages in a size-independent manner, only large particles or NP aggregates in the micrometer range induce cellular responses that include production of mitochondrial reactive oxygen species, caspase-1 activation, and secretion of interleukin-1ß (IL-1ß). A comprehensive proteomic analysis reveals that the particle-bound proteins are similar in terms of their identity and number, regardless of particle size, suggesting that protein adsorption is independent of particle size and curvature. In conclusion, the conditions underlying the formation of mineralo-protein particles are similar to the ones that form in vivo. While mineral NPs do not activate immune cells, they may become pro-inflammatory and contribute to pathological processes once they aggregate and form larger mineral particles.

Intact INI1 gene region with paradoxical loss of protein expression in AT/RT: implications for a possible novel mechanism associated with absence of INI1 protein immunoreactivity.

Atypical teratoid/rhabdoid tumor (AT/RT) is a highly malignant central nervous system tumor often misdiagnosed as some other type of pediatric embryonal tumor, such as medulloblastoma (MB). Distinguishing AT/RT from primitive neuroectodermal tumor/MB is of clinical significance, as the reported survival rate of patients with AT/RT is much lower than that of patients with average-risk primitive neuroectodermal tumor/MB. The diagnosis of AT/RT currently relies primarily on the morphologic assessment and immunostaining of a few known markers, such as the lack of INI1 protein expression. Immunohistochemical staining of INI1 is considered very sensitive and is highly specific for the detection of INI1 genetic defects. Genetic studies have shown that deletion or mutation of the INI1 gene, which is located on 22q11.2, occurs in AT/RT lesions. During our gene expression microarray analysis, we unexpectedly found a subgroup of AT/RT patients still expressing INI1 mRNA, even though INI1 proteins were negative by immunohistochemistry in those cases. Direct DNA sequencing showed no INI1 sequence alternation in 3 of 4 AT/RTs. Point mutation was found in only 1 allele of the fourth case, which would result in a frameshift mutation and generate a new INI1 protein with an extra 100-aa tail. Global array comparative genomic hybridization analysis confirmed no aberration around the INI1 gene at 22q11.2. It also extended our knowledge on the chromosomal aberration situations in our series. This study reveals that a novel yet unidentified posttranscriptional regulatory mechanism(s) for INI1 protein synthesis exists in AT/RT tumor cells.

Serum-derived nanoparticles: de novo generation and growth in vitro, and internalization by mammalian cells in culture.

AIM: While nanoparticles (NPs) have been shown to form spontaneously in body fluids such as serum, the possible implications of these NPs for cell cultures that use supporting media containing serum remain unclear. To understand the de novo formation of NPs, we delineated their growth characteristics, chemical composition and interaction with cells in culture. MATERIALS & METHODS: Serum-derived particles were analyzed using a combination of dynamic light scattering, turbidity measurements, spectroscopic techniques and optical/electron microscopies. RESULTS: NPs were found in serum and in serum-containing medium and they increased in size and number during incubation. The mineral particles, consisting mainly of calcium carbonate phosphate bound to organics such as proteins, underwent an amorphous-to-crystalline transformation with time. Serum-derived particles were internalized by the cells tested, eventually reaching lysosomal compartments. CONCLUSION: The spontaneous formation of serum-derived NPs and their internalization by cells may have overlooked effects on cultured cells in vitro as well as potential pathophysiological consequences in vivo.

Rana catesbeiana ribonuclease inhibits Japanese encephalitis virus (JEV) replication and enhances apoptosis of JEV-infected BHK-21 cells.

Rana catesbeiana ribonuclease (RC-RNase) is a cytotoxic and antitumor RNase isolated from the oocyte yolk granules of the bullfrog R. catesbeiana. Our previous studies have shown that RC-RNase possesses antitumor activity by activating proapoptotic caspases. Here, we demonstrate that RC-RNase also possesses antiviral activity. By using cell viability and caspase activation assays, we show that RC-RNase largely enhances apoptosis of Japanese encephalitis virus (JEV)-infected BHK-21 cells by activating caspase-3, caspase-8, and caspase-9. In addition, immunoblotting experiments revealed that JEV infection enhances the internalization of RC-RNase by cells. In sum, these results indicate that RC-RNase provides a beneficial effect on JEV-infected cells by enhancing apoptosis.

Activation of the ERK signal transduction pathway by Epstein-Barr virus immediate-early protein Rta.

BRCA1-associated protein 2 (BRAP2) is known to interact with the kinase suppressor of Ras 1 (KSR1), inhibiting the ERK signal transduction cascade. This study found that an Epstein-Barr virus (EBV) immediate-early protein, Rta, is a binding partner of BRAP2 in yeast and confirmed the binding in vitro by a glutathione S-transferase pull-down assay and in vivo by co-immunoprecipitation in 293(maxi-EBV) cells. Binding studies also showed that Rta and KSR1 interacted with the C-terminal 202 aa region in BRAP2. Additionally, Rta appeared to prevent the binding of KSR1 to BRAP2, activating the ERK signal transduction pathway and the transcription of an EBV immediate-early gene, BZLF1. Activation of the ERK signal transduction pathway by Rta may be critical for the maintenance of the lytic state of EBV.

Nonribosomal synthesis of fengycin on an enzyme complex formed by fengycin synthetases.

Fengycin, a lipopeptidic antibiotic, is synthesized nonribosomally by five fengycin synthetases (FenC, FenD, FenE, FenA, and FenB) in Bacillus subtilis F29-3. This work demonstrates that these fengycin synthetases interlock to form a chain, which coils into a 14.5-nm structure. In this chain, fengycin synthetases are linked in the order FenC-FenD-FenE-FenA-FenB by interactions between the C-terminal region of an upstream enzyme and the N-terminal region of its downstream partner enzyme, with their amino acid activation modules arranged colinearly with the amino acids in fengycin. This work also reveals that fengycin is synthesized on this fengycin synthetase chain, explaining how fengycin is synthesized efficiently and accurately. The results from this investigation demonstrate that forming a peptide synthetase complex is crucial to nonribosomal peptide synthesis.

Post-translational modification of Rta of Epstein-Barr virus by SUMO-1.

Epstein-Barr virus (EBV) expresses an immediate-early protein, Rta, to activate the transcription of EBV lytic genes and the lytic cycle. This work identifies Ubc9 and PIAS1 as binding partners of Rta in a yeast two-hybrid screen. These bindings are verified by glutathione S-transferase pull-down assay, coimmunoprecipitation, and confocal microscopy. The interactions appear to cause Rta sumoylation, because not only can Rta be sumoylated in vitro but also sumoylated Rta can be detected in P3HR1 cells following lytic induction and in 293T cells after transfecting plasmids that express Rta and SUMO-1. Moreover, PIAS1 stimulates conjugation of SUMO-1 to Rta, thus acting as an E3 ligase. Furthermore, transfecting plasmids that express Ubc9, PIAS1, and SUMO-1 increases the capacity of Rta to transactivate the promoter that includes an Rta response element, indicating that the modification by SUMO-1 increases the transactivation activity of Rta. This study reveals that Rta is sumoylated at the Lys-19, Lys-213, and Lys-517 residues and that SUMO-1 conjugation at the Lys-19 residue is crucial for enhancing the transactivation activity of Rta. These results indicate that sumoylation of Rta may be important in EBV lytic activation.