Search

To establish a breast lesion risk stratification system using ultrasound images to predict breast malignancy and assess Breast Imaging Reporting and Data System (BI-RADS) categories simultaneously.This multicenter study prospectively collected a dataset of ultrasound images for 5012 patients at thirty-two hospitals from December 2018 to December 2020. A deep learning (DL) model was developed to conduct binary categorization (benign and malignant) and BI-RADS categories (2, 3, 4a, 4b, 4c, and 5) simultaneously. The training set of 4212 patients and the internal test set of 416 patients were from thirty hospitals. The remaining two hospitals with 384 patients were used as an external test set. Three experienced radiologists performed a reader study on 324 patients randomly selected from the test sets. We compared the performance of the DL model with that of three radiologists and the consensus of the three radiologists.In the external test set, the DL model achieved areas under the receiver operating characteristic curve (AUCs) of 0.980 and 0.945 for the binary categorization and six-way categorizations, respectively. In the reader study set, the DL BI-RADS categories achieved a similar AUC (0.901 vs. 0.933, p = 0.0632), sensitivity (90.98% vs. 95.90%, p = 0.1094), and accuracy (83.33% vs. 79.01%, p = 0.0541), but higher specificity (78.71% vs. 68.81%, p = 0.0012) than those of the consensus of the three radiologists.The DL model performed well in distinguishing benign from malignant breast lesions and yielded outcomes similar to experienced radiologists. This indicates the potential applicability of the DL model in clinical diagnosis.• The DL model can achieve binary categorization for benign and malignant breast lesions and six-way BI-RADS categorizations for categories 2, 3, 4a, 4b, 4c, and 5, simultaneously. • The DL model showed acceptable agreement with radiologists for the classification of breast lesions. • The DL model performed well in distinguishing benign from malignant breast lesions and had promise in helping reduce unnecessary biopsies of BI-RADS 4a lesions.

A mixture of ceramide (CER) and lecithin (LEC) at specific ratios was capable of forming oleogels in sunflower oil triggered by adding a trace amount of water. It was noted that the addition of water at different temperatures (TW) resulted in different gelation behaviors and microstructures. To better illuminate the assembly mechanism at different TW, samples with water added at different TW (20 °C, 45 °C, 70 °C and 95 °C) were prepared. The viscoelastic properties, microstructures, and the crystal packing of these samples were investigated. It was observed that all samples prepared at TW of 20 °C and 95 °C formed gels, while most samples prepared at TW of 45 °C and 70 °C were too weak to form gels. Gels prepared at 95 °C were stronger but more fragile in texture compared to gels produced at 20 °C. The crystal morphology of gels drastically changed with TW. Spindle-shaped crystals were observed in gels prepared at low TW (20 °C), while gels prepared at high TW (95 °C) exhibited a network with packed oil droplets stabilized by lamellar shells together with fibrillar crystals in the bulk phase. X-ray diffractograms showed a different reflection peak (d-spacing of 14.5 Å) in gel prepared at 20 °C, compared to the d-spacing in oleogels with a single gelator (13.14 Å and 15.33 Å, respectively, for CER and LEC). Gel prepared at 95 °C showed two long-spacing characteristic peaks, which correspond to the characteristic peaks of CER gel (∼13 Å) and LEC gel (∼12 Å). Fourier transform infrared spectroscopy results indicated that the different gelation behaviors at different TW were mainly caused by vibrational changes in the amide bond of CER. Our hypothesized assembly mechanism can be concluded as: increasing TW resulted in the conversion of CER and LEC crystallization from co-assembly (TW = 20 °C) to self-sorting by individual gelators (TW = 95 °C). In this study, novel water-induced oleogels were produced by manipulating TW, and such information further assists the rational design of lipid-based healthy fat products.

Ceramide (CER) is a novel food-grade organogelator with beneficial health effects. Lecithin (LEC) is not an effective gelator; however, it may alter the crystal morphology of the host gelator in a multicomponent system. In this paper, LEC and CER were mixed at various molar ratios in sunflower oil leading to different gelation behaviors. It was interesting since in this multicomponent system, gels formed when there was more less-effective gelator (LEC), while gels hardly formed when there was more effective gelator (CER). This drew our attention since there might not be only one kind of assembly mode between the LEC and the CER. A comprehensive rheological investigation was conducted. It was found that at specific ratios (6L4C and 5L5C), strong gels (G' > 1.0 × 105 Pa) formed with superior oil binding capacity (up to 99.84%). Meanwhile, these gels exhibited higher tolerance level to permanent deformation than the monocomponent gel. However, weak gels were observed off the optimal ratio (8L2C, 7L3C, 4L6C and 3L7C). The crystal morphology of gels drastically changed with change in gelator proportion. Short needle-like crystals and small rosette crystals were observed in 6L4C and 5L5C, respectively, while other samples exhibited spherulite-shaped crystals (8L2C, 7L3C, 4L6C, and 3L7C), which differed from any of the monocomponent gel structures (10L0C and 0L10C). Results from differential scanning calorimetry and polarized light microscopy suggested that the macroscopic properties are determined by the morphology and distribution of crystals rather than the crystallinity of the matrix. Fourier transform infrared spectroscopy results indicated the presence of van der Waals forces and the formation of hydrogen bonding between the phosphate of the LEC and the amide group of the CER. The above results indicated that the LEC and CER co-assembled at approximately equal molar ratio, and the redundant LEC or CER at other ratios self-sorted to combine with the co-assembled fibers by lateral association, leading to potentially different underlying microstructures. These multicomponent supramolecular oleogels with high tunability may further broaden their applicability in various healthy lipid-based product formats.

Ceramide (CER) is a highly effective organogelator with numerous health implications. Lecithin (LEC) is often used as a crystal modifier to improve gel properties. It was found in this study that in the presence of a small amount of water, CER and LEC self-assembled at specific ratios and form oleogels in sunflower oil (SFO), and the structure of which differed from any of the mono-component gel structures. To the best of our knowledge, this is the first water-induced multi-component oleogelation reported to date. To clarify when gelation occurs, a phase diagram of the four-component system SFO/CER/LEC/water was established and characterized. In the gel region, confocal laser scanning microscopy revealed that water was evenly distributed in the self-assembled microstructure. To clarify the vital role of water in formation of stable gel, a series of samples with varying ratios in CER and LEC with and without water were explored. The thermal properties, viscoelastic properties, microstructures, as well as crystal packing were investigated. With water added, the microscopic structures of CER/LEC gels changed from three-dimensional dendritic crystals to two-dimensional rod-like or platelet-like crystals, which are more effective at structuring liquid oil. A different reflection was observed (d-spacing of 14.9 Å), suggesting the rearrangement in crystal packing of ceramide and lecithin molecules, compared to the d-spacing in oleogels with single gelator (13.1 Å and 15.3 Å, respectively for ceramide and lecithin). Fourier transform infrared spectroscopy results indicated the presence of interaction between water and gelators, which confirmed the importance of water in self-assembly of this mixed-gelator system. Water-induced conversion of oleogelation provides alternatives in producing hard-stock fat replacers as well as possibilities to incorporate water-soluble nutrients into oil-based products.