Search

Objective To develop and validate a deep learning (DL) model based on CT for differentiating bone islands and osteoblastic bone metastases. Materials and methods The patients with sclerosing bone lesions (SBLs) were retrospectively included in three hospitals. The images from site 1 were randomly assigned to the training (70%) and intrinsic verification (10%) datasets for developing the two-dimensional (2D) DL model (single-slice input) and “2.5-dimensional” (2.5D) DL model (three-slice input) and to the internal validation dataset (20%) for evaluating the performance of both models. The diagnostic performance was evaluated using the internal validation set from site 1 and additional external validation datasets from site 2 and site 3. And statistically analyze the performance of 2D and 2.5D DL models. Results In total, 1918 SBLs in 728 patients in site 1, 122 SBLs in 71 patients in site 2, and 71 SBLs in 47 patients in site 3 were used to develop and test the 2D and 2.5D DL models. The best performance was obtained using the 2.5D DL model, which achieved an AUC of 0.996 (95% confidence interval [CI], 0.995–0.996), 0.958 (95% CI, 0.958–0.960), and 0.952 (95% CI, 0.951–0.953) and accuracies of 0.950, 0.902, and 0.863 for the internal validation set, the external validation set from site 2 and site 3, respectively. Conclusion A DL model based on a three-slice CT image input (2.5D DL model) can improve the prediction of osteoblastic bone metastases, which can facilitate clinical decision-making. Key Points • This study investigated the value of deep learning models in identifying bone islands and osteoblastic bone metastases. • Three-slice CT image input (2.5D DL model) outweighed the 2D model in the classification of sclerosing bone lesions. • The 2.5D deep learning model showed excellent performance using the internal (AUC, 0.996) and two external (AUC, 0.958; AUC, 0.952) validation sets.

Sasobit® can be used not only as a warm mix additive in asphalt mixtures but also to further modify highly viscous polymer-modified asphalt (PMA) to achieve viscosity reduction, energy consumption reduction, and environmentally friendly engineering purposes. This paper demonstrates innovative research to evaluate several properties of asphalt binders modified with Sasobit® and polymers, including the exploration of the modification mechanism, fatigue damage characteristics, and healing potential at intermediate temperatures. X-ray diffraction (XRD) and Fourier transform infrared (FTIR) spectroscopy were conducted to explore the effectiveness of Sasobit® on the crystallization performance and chemical structure of styrene–butadienestyrene-modified asphalt (SBSMA), crumb rubber-modified asphalt (CRMA), and SBS/crumb rubber composite modified asphalt (SBS + CRMA). The accelerated fatigue approach, the linear amplitude sweep (LAS) test, was adopted to develop fatigue damage evolution of the Sasobit®/polymer-modified binders. A fatigue failure definition and the corresponding failure criteria based on the pseudo-strain energy (PSE) approach were adopted under the framework of the viscoelastic continuum damage (VECD) theoretical model. Subsequently, a healing process was extended. The LAS-based healing (LASH) step was introduced to assess the effectiveness of Sasobit® on the healing potential and fatigue life of the PMA. The XRD results suggested that Sasobit® weakened the crystallization effect of the PMA to some extent. Sasobit® affected the absorption of the polymers into the lighter components of the binder without creating new chemical components. The Sasobit® additives significantly raised the potential of PMA samples to resist fatigue failure. The LASH test results illustrated that Sasobit® increased the healing capacities of SBSMA and CRMA, but decreased the healing potential of SBS + CRMA. Furthermore, compared to the virgin binder, the Sasobit®/polymer-modified binder exhibited a weakened capacity to resist fatigue damage after short-term aging but a strengthened capability after appropriate long-term ultraviolet aging.

This study employs two carbon nanomaterial (i.e., graphene oxide, GO; carbon nanotubes, CNT)/polymer (i.e., styrene-butadienestyrene, SBS) composites to modify asphalt binder for the optimal design of asphalt pavement materials. The feasibility of utilizing GO + SBS or CNT + SBS composites in asphalt binders is investigated from the perspectives of the modification mechanism, aging resistance, high- and low-temperature rheological properties, and fatigue performance. To reveal the modification mechanism, X-ray diffraction (XRD) and Raman spectroscopy were performed to detect the composition, crystal structure, and defects of the GO/CNT/SBS modifiers and the modified asphalt binders. Fourier transform infrared (FTIR) spectroscopy characterization was performed to reveal the interaction mechanism between the GO/CNT/SBS modifiers and the base binder. The variations in oxygen-containing groups were detected by FTIR to assess the aging resistance of all the modified binders. The high- and low- temperature properties were determined by multiple stress creep recovery (MSCR) tests and bending beam rheometer (BBR) tests, respectively. Moreover, the fatigue performance of the GO + SBS- and CNT + SBS-modified binder was examined by linear amplitude sweep (LAS) tests. The XRD and Raman spectroscopy results suggested that the GO/CNT/SBS modifiers were of high quality, and showed that GO was beneficial to the crystallization of SBS-modified binders, while CNT had the opposite effect. The FTIR characterization results showed that there was no chemical reaction between GO (or CNT) and SBS or the base binder. The promotion of the anti-aging resistance of the asphalt binders by GO and CNT was proven by FTIR analysis. A favorable high-temperature rheological performance was obtained for the binders with GO or CNT additives, and the low-temperature performance of the SBS-modified binders was also promoted by GO and CNT. Moreover, the nano-reinforcement effect of carbon nanomaterials can extend the asphalt binder’s fatigue life predicted under the framework of the linear viscoelastic continuum damage theory.

The utilization of carbon nanomaterials such as graphene oxide (GO) and carbon nanotube (CNT) to modify asphalt binders is an innovative process. The feasibility of modifying asphalt binder with carbon nanomaterials was evaluated from the perspectives of storage stability, thermodynamic property, thermal stability and chemical structure. Accordingly, a combination of the segregation tests, dynamic mechanical rheological analyses and micro-morphology observation methods were adopted to verify whether the carbon-based nanoparticles could possibly segregate in the asphalt binders. The sessile drop method based on the surface free energy (SFE) theory was utilized to calculate the thermodynamic parameters of carbon nanomaterials-modified binders and the aggregates, and the effect of aging on the thermodynamic characteristics of the modified binders was evaluated. Moreover, the thermal gravimetric analysis (TGA) was performed to assess the effect of GO and CNT on the thermal characteristics of the binders. Finally, changes in the chemical structure were detected via Fourier transform infrared (FTIR) spectroscopy. The physical tests of the softening point differences, the rheological responses and the microscopic morphologies showed that the carbon nanomaterial-modified binders could achieve storage stability characteristics. The results of the SFE and related parameters suggested that the carbon-based nanoparticles can promote the thermodynamic characteristics of the asphalt binder. The rolling thin film oven (RTFO) process promoted the thermodynamic properties of the modified binders while the 12-day ultraviolet aging had an adverse effectiveness on the thermodynamic properties of the modified binders. For TGA tests, three indicators such as the residues at 600 °C revealed that CNT enhanced the thermal stability of the binders while GO weakened the thermal characteristics of the binders. The FTIR data demonstrated that no chemical reactions occurred in the process of preparing carbon nanomaterial-modified binders.