Automatic Detection and Classification of Modic Changes in MRI Images Using Deep Learning: Intelligent Assisted Diagnosis System.

OBJECTIVE: Modic changes (MCs) are the most prevalent classification system for describing intravertebral MRI signal intensity changes. However, interpreting these intricate MRI images is a complex and time-consuming process. This study investigates the performance of single shot multibox detector (SSD) and ResNet18 network-based automatic detection and classification of MCs. Additionally, it compares the inter-observer agreement and observer-classifier agreement in MCs diagnosis to validate the feasibility of deep learning network-assisted detection of classified MCs. METHOD: A retrospective analysis of 140 patients with MCs who underwent MRI diagnosis and met the inclusion and exclusion criteria in Tianjin Hospital from June 2020 to June 2021 was used as the internal dataset. This group consisted of 55 males and 85 females, aged 25 to 89 years, with a mean age of (59.0 ± 13.7) years. An external test dataset of 28 patients, who met the same criteria and were assessed using different MRI equipment at Tianjin Hospital, was also gathered, including 11 males and 17 females, aged 31 to 84 years, with a mean age of 62.7 ± 10.9 years. After Physician 1 (with 15 years of experience) annotated all MRI images, the internal dataset was imported into the deep learning model for training. The model comprises an SSD network for lesion localization and a ResNet18 network for lesion classification. Performance metrics, including accuracy, recall, precision, F1 score, confusion matrix, and inter-observer agreement parameter Kappa value, were used to evaluate the model's performance on the internal and external datasets. Physician 2 (with 1 year of experience) re-labeled the internal and external test datasets to compare the inter-observer agreement and observer-classifier agreement. RESULTS: In the internal dataset, when models were utilized for the detection and classification of MCs, the accuracy, recall, precision and F1 score reached 86.25%, 87.77%, 84.92% and 85.60%, respectively. The Kappa value of the inter-observer agreement was 0.768 (95% CI: 0.656, 0.847),while observer-classifier agreement was 0.717 (95% CI: 0.589, 0.809).In the external test dataset, the model's the accuracy, recall, precision and F1 scores for diagnosing MCs reached 75%, 77.08%, 77.80% and 74.97%, respectively. The inter-observer agreement was 0.681 (95% CI: 0.512, 0.677), and observer-classifier agreement was 0.519 (95% CI: 0.290, 0.690). CONCLUSION: The model demonstrated strong performance in detecting and classifying MCs, achieving high agreement with physicians in MCs diagnosis. These results suggest that deep learning models have the potential to facilitate the application of intelligent assisted diagnosis techniques in the field of spine research.

[Phase amplitude coupling analysis of local field potentials in working memory of rats affected by transcranial magneto-acoustic-electrical stimulation].

Transcranial magneto-acoustic-electrical stimulation is a new non-invasive neuromodulation technology, in which the induced electric field generated by the coupling effect of ultrasound and static magnetic field are used to regulate the neural rhythm oscillation activity in the corresponding brain region. The purpose of this paper is to investigate the effects of transcranial magneto-acoustic-electrical stimulation on the information transfer and communication in neuronal clusters during memory. In the experiment, twenty healthy adult Wistar rats were randomly divided into a control group (five rats) and stimulation groups (fifteen rats). Transcranial magneto-acoustic-electrical stimulation of 0.05~0.15 T and 2.66~13.33 W/cm 2 was applied to the rats in stimulation groups, and no stimulation was applied to the rats in the control group. The local field potentials signals in the prefrontal cortex of rats during the T-maze working memory tasks were acquired. Then the coupling differences between delta rhythm phase, theta rhythm phase and gamma rhythm amplitude of rats in different parameter stimulation groups and control group were compared. The experimental results showed that the coupling intensity of delta and gamma rhythm in stimulation groups was significantly lower than that in the control group ( P<0.05), while the coupling intensity of theta and gamma rhythm was significantly higher than that in the control group ( P<0.05). With the increase of stimulation parameters, the degree of coupling between delta and gamma rhythm showed a decreasing trend, while the degree of coupling between theta and gamma rhythm tended to increase. The preliminary results of this paper indicated that transcranial magneto-acoustic-electrical stimulation inhibited delta rhythmic neuronal activity and enhanced the oscillation of theta and gamma rhythm in the prefrontal cortex, thus promoted the exchange and transmission of information between neuronal clusters in different spatial scales. This lays the foundation for further exploring the mechanism of transcranial magneto-acoustic-electrical stimulation in regulating brain memory function.

A novel environmental fate of graphene oxide: Biodegradation by a bacterium Labrys sp. WJW to support growth.

Graphene oxide (GO) is a new type of nanomaterial with unique physicochemical properties and diverse applications, whereas it poses potential risk to human and environment. By screening from natural soil exposed to GO in the laboratory, we successfully obtained a novel bacterium, Labrys sp. WJW, which was able to use GO as the sole carbon source for growth. Within 8 days, cell numbers increased 16.76 ±â€¯3.21 folds using 100 mg/L GO as the carbon source by qPCR analysis. The bacterial biodegradation which resulted in formation of holes and functional group changes of GO was proved by Raman spectroscopy, atomic force microscopy, transmission electron microscopy and X-ray photoelectron spectroscopy analyses. Aromatic intermediates with structures of benzoic acid and phenol were identified using gas chromatograph-mass spectrometry and liquid chromatography/time-of-flight/mass spectrometry. Combination of genomic and proteomic analyses were performed to explore the proteins associated with GO degradation. A total of 644 proteins were significantly shifted. Bioinformatics analysis indicated that part of the up-regulated proteins were related to oxidation, ring cleavage and intermediates transmembrane processes, and GO was supposed to be degraded to benzoate and further degraded for downstream processes. This study enriches our understanding and provides new insights into the environmental fate of GO.

Comparison of gold nanoparticles biosynthesized by cell-free extracts of Labrys, Trichosporon montevideense, and Aspergillus.

Biosynthesis of gold nanoparticles (AuNPs) by microbes has received much attention as an efficient and eco-friendly process. However, the characteristics of AuNPs biosynthesized by different microbial cell-free extracts are rarely comparatively studied. In this study, three locally isolated strains, i.e., bacteria Labrys sp. WJW, yeast Trichosporon montevideense WIN, and filamentous fungus Aspergillus sp. WL-Au, were selected for AuNPs biosynthesis. UV-Vis absorption bands at 538, 539, and 543 nm confirmed the formation of AuNPs by these strains. Transmission electron microscopy and selected area electron diffraction analyses revealed that the as-synthesized AuNPs were crystalline with spherical or pseudo-spherical shapes. However, the average sizes of these AuNPs were diverse, which were 18.8, 22.2 and 9.5 nm, respectively. The biomolecules involved in nanoparticles stabilization were demonstrated by Fourier transform infrared spectroscopy analysis. Four common functional groups such as -N-H, -C=C, -N=O, and -S=O groups were detected in these AuNPs, while a distinct -C=O group was involved in WL-Au-AuNPs. The catalytic rate of WL-Au-AuNPs toward 4-nitrophenol reduction (0.37 min-1) was much higher than those of others (WJW-AuNPs 0.27 min-1 and WIN-AuNPs 0.23 min-1). This research would provide useful information for exploring efficient microbial candidates to synthesize AuNPs with excellent performances.

Biosynthesis of gold nanoparticles using cell-free extracts of Magnusiomyces ingens LH-F1 for nitrophenols reduction.

A green and eco-friendly method for the synthesis of gold nanoparticles (AuNPs) was developed using the cell-free extracts of a yeast strain Magnusiomyces ingens LH-F1. UV-vis spectra showed a distinct absorption band at ~ 540 nm, corresponding to the surface plasmon resonance of AuNPs. Transmission electron microscopy images revealed that the shapes of AuNPs were almost spherical and pseudo-spherical. Fourier transform infrared spectroscopy and sodium dodecyl sulfate-polyacrylamide gel electrophoresis analyses suggested that some proteins containing amino- and carboxyl-groups in the cell-free extracts were absorbed on the surface of nanoparticles, which could act as reducing and capping agents for AuNPs synthesis. Furthermore, with the concentration of cell-free extracts increasing from 25 to 200 mg L-1, the average size of AuNPs decreased from 28.3 to 20.3 nm. Meanwhile, the morphology became more uniform with less irregular shapes. In addition, the as-synthesized AuNPs showed an excellent catalytic activity for nitrophenols reduction (i.e., 4-nitrophenol, 3-nitrophenol and 2-nitrophenol) in the presence of excess NaBH4. The catalytic rate constant of nitrophenols reduction was also dependent on cell-free extract concentration. The larger AuNPs synthesized by less cell-free extracts were covered with a thinner corona and showed better capacity for reducing nitrophenols. This study suggested that the as-synthesized AuNPs could be employed as efficient catalysts in reduction of organic contaminants.

Biosynthesis of gold nanoparticles by Trichoderma sp. WL-Go for azo dyes decolorization.

Developing an eco-friendly approach for metallic nanoparticles synthesis is important in current nanotechnology research. In this study, green synthesis of gold nanoparticles (AuNPs) was carried out by a newly isolated strain Trichoderma sp. WL-Go. UV-vis spectra of AuNPs showed a surface plasmon resonance peak at 550nm, and transmission electron microscopy images revealed that the AuNPs were of varied shape with well dispersibility. The optimal conditions for AuNPs synthesis were HAuCl4 1.0mmol/L, biomass 0.5g and pH7-11. Moreover, the bio-AuNPs could efficiently catalyze the decolorization of various azo dyes. This research provided a new microbial resource candidate for green synthesis of AuNPs and demonstrated the potential application of bio-AuNPs for azo dye decolorization.

Control of supramolecular nanoassemblies by tuning the interactions of bent-shaped rod-coil molecules.

Rod-coil molecules 1a, 1b and 2a, 2b, consisting of biphenyl and phenyl units connected by an acetylene bond as the rod segment and oligo(ethylene glycol) (OEG) as the coil segment, were synthesized and characterized. Molecules 1a and 1b incorporate a butoxy group at the apex of their bent-shaped rigid building blocks, while both 1b and 2b contain a lateral methyl group between the rod and coil segments. The self-assembling behavior of these molecules was investigated using DSC, SAXS, CD, AFM, and TEM in bulk and aqueous solutions. In the bulk state, 1a self-assembles into oblique columnar structures, whereas 1b, incorporating butoxy and lateral methyl groups, self-assembles into three-dimensional body-centered tetragonal structures. Molecules 2a and 2b with no butoxy groups, and 2b incorporating a lateral methyl group, self-assemble into hexagonal perforated lamellar and oblique columnar structures, respectively. In dilute aqueous solutions, 1a assembles into tubular nanoassemblies, while 1b self-organizes into micelles and nanoparticles. On the other hand, 2a and 2b spontaneously aggregate into nanoribbons and nanofibers. Furthermore, CD experiments together with AFM investigations of 2b indicate the creation of self-organized helical fibers, implying that the lateral methyl group induces the helical stacking of the rod building block. These results reveal that the butoxy and lateral methyl groups between the rod and coil segments dramatically influence the creation of supramolecular nanostructures and morphologies.

Catalytic reduction of 4-nitrophenol using gold nanoparticles biosynthesized by cell-free extracts of Aspergillus sp. WL-Au.

A facile one-pot eco-friendly process for synthesis of gold nanoparticles (AuNPs) with high catalytic activity was achieved using cell-free extracts of Aspergillus sp. WL-Au as reducing, capping and stabilizing agents. The surface plasmon resonance band of UV-vis spectrum at 532nm confirmed the presence of AuNPs. Transmission electron microscopy images showed that quite uniform spherical AuNPs were synthesized and the average size of nanoparticles increased from 4nm to 29nm with reaction time. X-ray diffraction analysis verified the formation of nano-crystalline gold particles. Fourier transform infrared spectra showed the presence of functional groups on the surface of biosynthesized AuNPs, such as OH, NH, CO, CH, COH and COC groups, which increased the stability of AuNPs. The biogenic AuNPs could serve as a highly efficient catalyst for 4-nitrophenol reduction. The reaction rate constant was linearly correlated with the concentration of AuNPs, which increased from 0.59min-1 to 1.51min-1 with the amount of AuNPs increasing form 1.46×10-6 to 17.47×10-6mmol. Moreover, the as-synthesized AuNPs exhibited a remarkable normalized catalytic activity (4.04×105min-1mol-1), which was much higher than that observed for AuNPs synthesized by other biological and conventional chemical methods.

Illumina MiSeq sequencing reveals long-term impacts of single-walled carbon nanotubes on microbial communities of wastewater treatment systems.

In this study, phenol wastewater treatment systems treated with different concentrations of single-walled carbon nanotubes (SWCNTs) (0-3.5g/L) were exposed to phenol and carbon nanotubes (CNTs) shock loadings to investigate the long-term impacts of SWCNTs on microbial communities. Phenol removal remained high efficiency (>98%) in SWCNTs-treated groups but decreased in non-treated group (85.1±1.9%) when exposed to high concentration of phenol (500mg/L). However, secondary dosing of SWCNTs in SWCNTs-treated groups would decrease the phenol removal efficiency. Illumina MiSeq sequencing revealed that the diversity, richness and structure of microbial communities were shifted under phenol shock loading, especially under high phenol concentration, but not under CNTs shock loading. In response to phenol and CNTs shock loadings, Rudaea, Burkholderia, Sphingomonas, Acinetobacter, Methylocystis and Thauera became dominant genera, which should be involved in phenol removal. These results suggested that a proper amount of SWCNTs might have positive effects on phenol wastewater treatment systems.