A machine learning study on the fatigue crack path of short crack on an α titanium alloy.

In the present study, a physics-informed neural network model based on Bayesian hyperparameter optimization is proposed for the prediction of short crack growth paths. A large number of cyclic loadings at a lower amplitude were applied to an α titanium sample by an ultrasonic fatigue machine to ensure a sufficient amount of data for machine learning. The grain size, grain orientation and grain boundary direction on the path, as well as crack growth direction, were selected as feature data for training the prediction model. The optimizations of the size ratio and the angle operation were conducted to compare different data processing methods, respectively. After evaluation, eventually, a model for predicting crack growth path is obtained with a reliable performance of 10% tolerance on the path angle at each grain boundary. And the prediction effect of the proposed model is better than that of some classic machine learning models and slip trace analysis. This article is part of the theme issue 'Physics-informed machine learning and its structural integrity applications (Part 1)'.

Toward Extending Concentric Tube Robot Kinematics for Large Clearance and Impulse Curvature.

Concentric Tube Robots (CTRs) have been proposed to operate within the unstructured environment for minimally invasive surgeries. In this letter, we consider the operation scenario where the tubes travel inside the channels with a large clearance or large curvature, such as aortas or industrial pipes. Accurate kinematic modeling of CTRs is required for the development of motion planning and control within these operation scenarios. To this end, we extended the conventional CTR kinematics model to a more general case with large tube-to-tube clearance and large centerline curvature. Numerical simulations and experimental validations are conducted to compare our model with respect to the conventional CTR kinematic model. In the physical experiments, our proposed model achieved a tip position error of 1.53 mm in the 2D planer case and 3.86 mm in 3D case, outperforming the state-of-the-art model by 71% and 61%, respectively.

Enhancing Persistent Luminescence through Synergy between Optimal Electron Traps and Dye Sensitization.

Due to their unique afterglow ability, long-wavelength-light rechargeable persistent luminescence (PersL) nanoparticles (PLNPs) have been emerging as an important category of imaging probes. Among them, ZnGa2O4:0.6% Cr3+ (ZGC) PLNPs have gained widespread recognition due to the ease of synthesis and uniform morphology. Unfortunately, the limited absorption arising from the low molar extinction coefficient of Cr3+ results in relatively low afterglow intensity and rapid decay after long-wavelength LED light irradiation. Herein, we discovered a strategy that boosting dye-sensitization performance was able to effectively amplify the PersL signal under white LED light. Specifically, Dil served as a highly efficient sensitizer for Cr3+, promoting the absorption of the excitation light. By adjusting the Pr dopant concentrations, ZGCP0.5 PLNPs with optimal trap densities were obtained, which showed the highest PersL intensity and dye-sensitized performance. Strikingly, ZGCP0.5-Dil PLNPs exhibited a 24.3-fold enhancement in intensity and a 2-fold prolongation of decay time over bare ZGC PLNPs through the synergy effect of optimal electron traps and dye sensitization. Photostable ZGCP0.5-Dil PLNPs enabled imaging of the HepG2 tumor and effectively guided tumor surgical resection verified by the H&E staining analysis. This strategy could be a significant reference in other dye-sensitization PLNPs to enhance longer-wavelength rechargeable PersL.

Persistent luminescence nanoparticles with high intensity for colorectal cancer surgery navigation and precision resection.

Surgical resection remains the main treatment for malignant tumors. Image-guided surgery aims to remove tumor tissue completely while preserving normal tissue, thereby reducing tumor recurrence rates and injury. However, challenges like tissue autofluorescence, limited probe penetration and low contrast restrict its use. Near-infrared (NIR) persistent luminescent nanoparticles (PLNPs) provide a solution by emitting persistent luminescence (PersL) even after excitation ceases, thus circumventing autofluorescence and enabling deep tumor imaging. In this study, we prepared nano-sized (140 nm hydrodynamic size) Cr3+ doped zinc gallogermanate (ZGC) using a removable template method and modified it with folate acid to obtain ZGC-FA, which exhibits NIR (695 nm) PersL with a signal-to-noise ratio of 23.9 in vivo. We utilized a colon cancer model that selectively expressed luciferase for the first time to validate the guiding efficacy of ZGC-FA in precision surgical resection. Post-intraperitoneal injection at 50 minutes, the PersL closely matched the tumor boundaries, achieving an overlap rate of approximately 98%. Complete tumor resection was achieved under PersL guidance, with only 2.3% of healthy tissue removed. This research underscores the potential of ZGC-FA in the field of surgical oncology. The precision of the ZGC-FA guided surgical approach holds promise to enhance surgical outcomes and facilitate postoperative recovery in patients.

pH-Responsive Plasmon-Enhanced Persistent Luminescent ZnGa2O4:Cr3+ Nanopomegranate for Tumor Imaging.

Noble metal compositing is a promising method to enhance radiance intensity of persistent luminescent (PersL) nanoparticles (NPs) via surface plasmon resonance (SPR) for better tumor imaging, but it rarely unites with the pH-response strategy due to the challenge of realizing rigorous pH-responsive spatial distance control as a "button switch" of SPR. Here, ZnGa2O4:Cr3+ (ZGC) NPs as "pomegranate seeds" are cladded with sodium alginate to form nanoclusters (ZGC-SA), subsequently coated with carboxyl-rich polymers to acquire "pomegranate rind" (ZSPB) and finally decorated with 10 nm gold NPs (AuNPs) on the surface to obtain nanopomegranate structure (ZSPB@AuNPs). Though without deliberate distance control, there are plenty of "seeds" inside ZSPB@AuNPs fortunately at appropriate positions, which could be plasmon-enhanced by AuNPs. Furthermore, triggered by carboxyl protonation in subacid tumor, ZSPB@AuNPs aggregate and subsequently facilitate such plasmon enhancement effect, resulting in 4.4-fold PersL promotion at pH 5.5 (tumor microenvironment, TME) over pH 7.4 and in a maximum "tumor to normal tissue ratio" of PersL imaging signals of 125.9. Under surgical navigation of ZSPB@AuNPs, intramuscular tumors of mice could be resected without residue signals left. This nanopomegranate achieves TME pH-responsive plasmon-enhanced PersL for the first time and broadens the way for designing plasmon-enhanced PersL nanosystems.

Electrospun Mussel-derived Silk Fibers.

BACKGROUND: Though there are many patents on silk, patents on sea silk are rare. Sea silk is one of the most coveted materials in the world, and the technology to make sea silk is at an extremely high risk of extinction. Unlike spider dragline silk and silkworm silk, this natural silk has been forgotten in the academic commune for millennia, though it has many fascinating properties: high strength, remarkable adhesion, extreme lightweight, and others. METHODS: Here we report that mussel-derived silk fibers can be fabricated by electrospinning. Instead of extracting proteins from byssus, we directly use the protein solution from alive blue mussels, which are intensely commercially used. The protein solution and the polyvinyl alcohol solution are mixed together to produce mussel-based silk fibers. RESULTS: The mussel-based silk fibers have many special properties like high mechanical strength, remarkable super-contraction and good wetting properties. CONCLUSION: The electrospinning mussel-based silk fibers have the potential for use as a replacement for the rarest sea silk and as a new bio-inspired material with multi-functions.