HSG-MGAF Net: Heterogeneous subgraph-guided multiscale graph attention fusion network for interpretable prediction of whole-slide image.

BACKGROUND AND OBJECTIVE: Pathological whole slide image (WSI) prediction and region of interest (ROI) localization are important issues in computer-aided diagnosis and postoperative analysis in clinical applications. Existing computer-aided methods for predicting WSI are mainly based on multiple instance learning (MIL) and its variants. However, most of the methods are based on instance independence and identical distribution assumption and performed at a single scale, which not fully exploit the hierarchical multiscale heterogeneous information contained in WSI. METHODS: Heterogeneous Subgraph-Guided Multiscale Graph Attention Fusion Network (HSG-MGAF Net) is proposed to build the topology of critical image patches at two scales for adaptive WSI prediction and lesion localization. The HSG-MGAF Net simulates the hierarchical heterogeneous information of WSI through graph and hypergraph at two scales, respectively. This framework not only fully exploits the low-order and potential high-order correlations of image patches at each scale, but also leverages the heterogeneous information of the two scales for adaptive WSI prediction. RESULTS: We validate the superiority of the proposed method on the CAMELYON16 and the TCGA- NSCLC, and the results show that HSG-MGAF Net outperforms the state-of-the-art method on both datasets. The average ACC, AUC and F1 score of HSG-MGAF Net can reach 92.7 %/0.951/0.892 and 92.2 %/0.957/0.919, respectively. The obtained heatmaps can also localize the positive regions more accurately, which have great consistency with the pixel-level labels. CONCLUSIONS: The results demonstrate that HSG-MGAF Net outperforms existing weakly supervised learning methods by introducing critical heterogeneous information between the two scales. This approach paves the way for further research on light weighted heterogeneous graph-based WSI prediction and ROI localization.

A novel strategy regarding geometric product for liquids discrimination based on THz reflection spectroscopy.

A novel expression of geometric product that associated with the geometric relationship from geometric algebra constructed by the vectorized refractive index and absorption coefficient in THz region is proposed, which could provide a new insight into the THz properties of materials. From the novel expression, the candidate characteristic parameters are extracted for liquids discrimination and present the abundant second order correlation information of optical parameters with the consideration of dimension rising. Three groups of liquids, containing C-reactive protein calibrators and alpha fetoprotein calibrators, were selected as examples to validate the feasibility of the proposed strategy. Comparing with the traditional THz parameters, including refractive index, absorption coefficient, and complex permittivity, the novel approach exhibits notable superiority for differentiation with the evaluation of statistical differences and effect sizes. The proposed geometric product expression could have a large potential on promoting the substance identification in some applications of THz technology.

Diagnosis of hepatocellular carcinoma based on a terahertz signal and VMD-CWSE.

A novel strategy on combining variational mode decomposition (VMD) and composite weighted-scale sample entropy (CWSE) modified from composite multiscale entropy (CMSE) is proposed to screen hepatocellular carcinoma (HCC) by measuring the terahertz (THz) pulse signals of ten normal and ten HCC serums. Eight measured HCC specimens are negative in serum biomarker alpha fetoprotein (AFP) determination. In CWSE, the time series with weighted-scales are generated from the weighted average processing in the coarse-grained time series corresponding to each scale of the CMSE algorithm. VMD served as a preprocessing method was introduced into decomposing THz signal to obtain the mode functions of specific bandwidth for identification. Final results reveal that more obtainable entropy values of CWSE for recognition in comparison to those of CMSE on the basis of the rule of statistically significant difference and effect size and also manifest the stronger discriminability than the traditional THz parameters. This study provides a new potential auxiliary tool for diagnosis HCC and develops the methodology on the discrimination for similar THz signals.

Chiral metallic glass nanolattices with combined lower density and improved auxeticity.

Auxetic materials are promising structural and functional candidates due to their unique lateral expansion when stretched, however, bulk metallic glasses (MGs) could not show any auxeticity because of their intrinsic isotropic nature. Here we construct chiral Cu50Zr50 metallic glass nanolattices with cavities, and investigate their auxeticity and underlying mechanism with molecular dynamics simulations. It is found that, compared to monolithic MGs, all the chiral metallic glass nanolattices (CMGNs) exhibit improved auxeticity and lower density. For CMGNs with cavities, the negative Poisson's ratio and ultimate tensile strength (UTS) increase first and then decrease with increasing cavity radius, with the cavity radius of 2.5 nm being the most favorable for auxeticity and enhanced UTS. The auxetic mechanism is attributed to the competition between rotation behavior and non-affine deformation under tension. Our study not only reveals the mechanism of auxeticity in CMGNs having cavities but also provides a feasible method to optimize their auxetic performance and density by structure designing of MGs.

Atomic-level crystallization in selective laser melting fabricated Zr-based metallic glasses.

As a promising additive manufacturing technique, selective laser melting (SLM) provides the possibility of fabricating metallic glassy components free of the constraints of geometrical complexity and dimensions. However, unexpected crystallization greatly affects the microstructure and degrades the mechanical performance of SLM-fabricated metallic glasses (MGs). To clarify the crystallization mechanism and the effect of laser processing on the crystallization, we investigate the atomic-level crystallization in the SLM Zr90Cu10 MG by using molecular dynamics simulations. The results show that crystallization highly related to scan speed lies in the atomic-level cluster changes. Lower scan speed leads to a dramatically increased fraction of the BCC crystal phase, accompanied by the nucleation of a few HCP and FCC crystal phases. As scan speed increases, more icosahedron-like clusters are formed, leading to the formation of the MG, while the nucleation of the crystal phase is suppressed. The suppression of crystallization is further attributed to a higher average temperature variation rate induced by higher scan speed, which reduces the relaxation time, preventing the nucleation and growth of crystal phases. This work contributes to the understanding of the crystallization in MGs during the SLM process at the atomic level, providing guidance to suppress the crystallization in the SLM process of desired metallic glassy components.

Electronic specific heats for amorphous and crystallized alloys.

The low temperature specific heats of (Fe0.5Co0.5)72B20Si4Nb4 amorphous and crystallized alloys are measured and analyzed from 1.4 to 110 K. Specific heats can be well fitted by electronic and phonon contribution terms. It is found that the electronic contribution term in specific heat for amorphous alloy is larger than that for crystallized one, and this phenomenon has been interpreted in detail. The research shows that the electronic density of states at the Fermi level and the localized loose "rattler" atoms in oversized cage structure may make contributions to the enhancement of electronic specific heat coefficient γ, and result in a larger electronic contribution term. This study is significant for further understanding the structure-property relationship for amorphous alloys at low temperature.

Mechanical properties and structural features of novel Fe-based bulk metallic glasses with unprecedented plasticity.

Fe-based bulk metallic glasses (BMGs) have attracted great attention due to their unique magnetic and mechanical properties, but few applications have been materialized because of their brittleness at room temperature. Here we report a new Fe(50)Ni(30)P(13)C(7) BMG which exhibits unprecedented compressive plasticity (>20%) at room temperature without final fracture. The mechanism of unprecedented plasticity for this new Fe-based BMG was also investigated. It was discovered that the ductile Fe(50)Ni(30)P(13)C(7) BMG is composed of unique clusters mainly linked by less directional metal-metal bonds which are inclined to accommodate shear strain and absorbed energy in the front of crack tip. This conclusion was further verified by the X-ray photoelectron spectroscopy and ultraviolet photoelectron spectroscopy experiments of Fe(80-x)Ni(x)P(13)C(7) (x = 0, 10, 20, 30) and Fe(72-x)Ni(x)B(20)Si(4)Nb(4) (x = 0, 7.2, 14.4, 21.6, 28.8) glassy systems. The results also indicate a strong correlation between the p-d hybridization and plasticity, verifying that the transition from brittle to ductile induced by Ni addition is due to the change of bonding characteristics in atomic configurations. Thus, we can design the plasticity of Fe-based BMGs and open up a new possible pathway for manufacturing BMGs with high strength and plasticity.