An interpretable machine learning model for predicting 5‐year survival in breast cancer based on integration of proteomics and clinical data
Peer-Reviewed Publication
Updates every hour. Last Updated: 23-Nov-2025 10:11 ET (23-Nov-2025 15:11 GMT/UTC)
Breast cancer is a highly heterogeneous malignancy among women worldwide. Traditional prognostic models relying solely on clinicopathological features offer limited predictive accuracy and lack molecular-level insights. Unlike such conventional approaches, this study integrates proteomic and clinical data within an interpretable deep learning framework to improve prognostic precision and biological interpretability. We aimed to develop a more reliable model to accurately predict the 5-year survival status of patients with breast cancer using multi-omics data. The model integrating proteomics and clinical features demonstrated superior performance (AUC = 0.8136) compared to other feature combination models. The optimized model with 13 key features (4 clinical features and 9 proteins) achieved an AUC of 0.864 with the precision of 0.970, the recall of 0.810, and F1-score of 0.883. SHapley Additive exPlanations analysis identified MPHOSPH10, EGFR, ARL3, KRT18, lymph node status, and HER2 status as the most influential features, while Kolmogorov–Arnold Network analysis provided explicit mathematical relationships between key contributors and prediction outcomes. Collectively, our interpretable multi-modal model demonstrates robust performance in predicting 5-year survival in breast cancer patients and offers mechanistic insights, thereby enhancing its potential for clinical translation through the development of an accessible prediction tool.
Skyrmions are three-dimensional, nontrivial topological textures that have attracted extensive attention in magnetism, condensed matter physics, and beyond due to their unique stability and rich physical implications. In recent years, researchers have succeeded in creating optical skyrmions, using light’s polarization and orbital angular momentum (OAM) to generate complex polarization topologies in the spatial domain. However, so far all optical skyrmions have been confined to the spatial domain, driven by longitudinal OAM to form twisted “helical tubes” along the propagation direction. This raises a natural question: can we break this limitation and bring skyrmions into spatiotemporal domain?
Recently, the team led by Prof. Qiwen Zhan at the University of Shanghai for Science and Technology has, for the first time, proposed and experimentally demonstrated a novel optical spatiotemporal skyrmion. By employing vectorial sculpturing of picosecond pulse wave packets, the researchers combined two orthogonally polarized beams—spatiotemporal Gaussian pulses and spatiotemporal vortex pulses—to generate a spatiotemporal topological structure covering the full set of polarization states: the spatiotemporal skyrmion.
The work, entitled “Construction of Optical Spatiotemporal Skyrmions”, is published in Light: Science & Applications. This breakthrough extends the topological concept of optical skyrmions from the spatial domain to the spatiotemporal domain, opening new directions for structured light and topological optics research.
A research paper by scientists from Liaoning Cancer Hospital & Institute, The First Affiliated Hospital of Ningbo University, and other institutions proposed a hollow mesoporous carbon (HMC) nanoparticle prepared via the sacrificial template method, featuring a porphyrin-like structure that enables efficient singlet oxygen generation and synergistic sono-immunotherapy for pancreatic cancer.
The new research paper, published on May 9 in the journal Cyborg and Bionic Systems, presented the preparation, characterization, and therapeutic application of MOF-derived HMC nanoparticles, and demonstrated their potential to enhance sono-immunotherapy efficacy by inducing tumor cell apoptosis and activating the immune
system.Recently, Professor Lu Zhengang's team at Harbin Institute of Technology proposed a non-microscope objective lithography method that utilizes the spherical convex lens aberration, enabling laser beams converged layer-by-layer axially. This technique can modulate collimated hollow beams into finer annular spots, directly generating annular patterns on curved substrates through a single laser pulse. The lithography device based on this method demonstrates superior performance. It achieves significantly finer line width and broader diameter adjustment ranges comparing to conventional annular lithography techniques. Moreover, compared to traditional laser direct-writing methods, it offers an extended depth of field and working distance and reduces hardware requirements while providing greater spatial redundancy for substrate positioning. This approach combines cost-effectiveness, high efficiency, and high performance. It is not only applicable to manufacturing ring-shaped metal mesh gratings and metasurface unit cells on curved substrates but also holds promise for providing viable solutions in various laser processing applications.
The research, titled “Ultra-Long Focal Depth Annular Lithography for Fabricating Micro Ring-Shaped Metasurface Unit Cells on Highly Curved Substrates”, was published in the top-tier optical journal Light: Advanced Manufacturing.
This review introduces an innovative and integrative perspective on nanoemulsion technology by linking formulation parameters directly to cosmetic performance outcomes—a relationship that has been largely overlooked in previous studies. Unlike earlier reviews that provided only general overviews, this paper critically synthesizes experimental findings to establish how specific formulation choices, such as surfactant selection, oil phase composition, and preparation method, affect stability, penetration depth, and active ingredient release in cosmetic products.
The review also highlights emerging strategies for achieving safer, more sustainable nanoemulsions through the use of natural surfactants, biodegradable oils, and environmentally responsible production methods. By combining insights from material science, dermatology, and cosmetic engineering, it provides a scientific framework for optimizing nanoemulsion design and guiding future product development.
Ultimately, this work not only consolidates the current understanding of nanoemulsions in cosmetics but also sets the foundation for the next generation of high-performance, biocompatible, and eco-friendly skincare formulations—bridging the gap between scientific innovation and consumer demand.
This review presents an innovative perspective on the replacement of conventional petrochemical-based surfactants with microbial biosurfactants such as rhamnolipids and sophorolipids in shampoo formulations. It emphasizes the integration of green chemistry principles and circular bioeconomy concepts into personal care product development. Unlike synthetic surfactants like SLS, biosurfactants are biodegradable, skin-compatible, and eco-friendly, reducing both dermatological irritation and environmental pollution. The article innovatively links biotechnological fermentation processes with cosmetic formulation science, providing comparative analyses of cleansing efficacy, foam stability, biodegradability, and protein protection in skin models. Moreover, it explores formulation optimization strategies, including hybrid surfactant systems and pH buffering to overcome stability and cost barriers. By combining environmental metrics with dermatological performance, this review establishes a holistic framework for sustainable shampoo design. The innovative aspect lies in positioning biosurfactants not merely as replacements for synthetics, but as multifunctional active ingredients that enhance scalp microbiome health while advancing global sustainability goals in the cosmetics industry.
Zn-I2 batteries have emerged as promising next-generation energy storage systems owing to their inherent safety, environmental compatibility, rapid reaction kinetics, and small voltage hysteresis. Nevertheless, two critical challenges, i.e., zinc dendrite growth and polyiodide shuttle effect, severely impede their commercial viability. To conquer these limitations, this study develops a multifunctional separator fabricated from straw-derived carboxylated nanocellulose, with its negative charge density further reinforced by anionic polyacrylamide incorporation. This modification simultaneously improves the separator’s mechanical properties, ionic conductivity, and Zn2+ ion transfer number. Remarkably, despite its ultrathin 20 μm profile, the engineered separator demonstrates exceptional dendrite suppression and parasitic reaction inhibition, enabling Zn//Zn symmetric cells to achieve impressive cycle life (> 1800 h at 2 mA cm−2/2 mAh cm−2) while maintaining robust performance even at ultrahigh areal capacities (25 mAh cm−2). Additionally, the separator’s anionic characteristic effectively blocks polyiodide migration through electrostatic repulsion, yielding Zn-I2 batteries with outstanding rate capability (120.7 mAh g−1 at 5 A g−1) and excellent cyclability (94.2% capacity retention after 10,000 cycles). And superior cycling stability can still be achieved under zinc-deficient condition and pouch cell configuration. This work establishes a new paradigm for designing high-performance zinc-based energy storage systems through rational separator engineering.