Marine microalgae play an increasingly significant role in addressing the issues of environmental monitoring and disease treatment, making the analysis of marine microalgae at the single-cell level an essential technique. For this, we put forward accurate and fast microfluidic impedance cytometry to analyze microalgal cells by assembling two cylindrical electrodes and microchannels to form a three-dimensional detection zone. Firstly, we established a mathematical model of microalgal cell detection based on Maxwell's mixture theory and numerically investigated the effects of the electrode gap, microalgal positions, and ion concentrations of the solution on detection to optimize detection conditions. Secondly, 80 μm stainless steel wires were used to construct flat-ended cylindrical electrodes and were then inserted into two collinear channels fabricated using standard photolithography techniques to form a spatially uniform electric field to promote the detection throughput and sensitivity. Thirdly, based on the validation of this method, we measured the impedance of living Euglena and Haematococcus pluvialis to study parametric influences, including ion concentration, cell density and electrode gap. The throughput of this method was also investigated, which reached 1800 cells per s in the detection of Haematococcus pluvialis. Fourthly, we analyzed live and dead Euglena to prove the ability of this method to detect the physiological status of cells and obtained impedances of 124.3 Ω and 31.0 Ω with proportions of 15.9% and 84.1%, respectively. Finally, this method was engineered for the analysis of marine microalgae, measuring living Euglena with an impedance of 159.61 Ω accounting for 3.9%, dead Euglena with an impedance of 36.43 Ω accounting for 10.1% and Oocystis sp. with an impedance of 55.00 Ω accounting for about 81.0%. This method could provide a reliable tool to analyze marine microalgae for monitoring the marine environment and treatment of diseases owing to its outstanding advantages of low cost, high throughput and high corrosion resistance.

Multivariate time series forecasting estimates future development by capturing variable relationships and constructing temporal regular, which is widely used in many scenarios, including industrial production, economic development, and disease prediction. Although the existing deep learning methods have achieved impressive results in multivariate time series forecasting tasks, the existing methods only emphasize the prediction performance and ignore the widespread issue of missing data in the real world. This article proposes a robust multivariate correlation self-distillation Transformer framework for incomplete time series data forecasting. The proposed method first decouples the interinference of historical series and the exter-inference of future series into two stages. The first stage focuses on the reconstruction of historical series, while the second stage focuses on the prediction of future series. Then, a novel multivariate correlation Transformer is designed as the basic component of the network, which can perform feature inference from both multivariate relationships and single-variate temporal regular. Finally, a variable correlation self-distillation method is proposed to self-distill the more complete variable relationship from the exter-inference stage to the interinference stage. The proposed method is verified on eight real-world datasets, and both qualitative and quantitative results show that the proposed method has good performance.

Non-spherical flagellate algae play an increasingly significant role in handling problematic issues as versatile biological micro/nanorobots and resources of valuable bioproducts. However, the commensalism of flagellate algae with distinct structures and constituents causes considerable difficulties in their further biological utilization. Therefore, it is imperative to develop a novel method to realize high-efficiency selection of non-spherical flagellate algae in a non-invasive manner. Enthused by these, we proposed a novel method to accomplish the selection of flagellate algae based on the numerical and experimental investigation of dielectrophoretic characterizations of flagellate algae. Firstly, an arbitrary Lagrangian–Eulerian method was utilized to study the electro-orientation and dielectrophoretic assembly process of spindle-shaped and ellipsoid-shaped cells in a uniform electric field. Secondly, we studied the equilibrium state of spherical, ellipsoid-shaped, and spindle-shaped cells under positive DEP forces actuated by right-angle bipolar electrodes. Thirdly, we investigated the dielectrophoretic assembly and escape processes of the non-spherical flagellate algae in continuous flow to explore their influences on the selection. Fourthly, freshwater flagellate algae (Euglena, H. pluvialis, and C. reinhardtii) and marine ones (Euglena, Dunaliella salina, and Platymonas) were separated to validate the feasibility and adaptability of this method. Finally, this approach was engineered in the selection of Euglena cells with high viability and motility. This method presents immense prospects in the selection of pure non-spherical flagellate algae with high motility for chronic wound healing, bio-micromotor construction, and decontamination with advantages of no sheath, strong reliability, and shape-insensitivity.

Adsorption of urea from dialysate is essential for wearable artificial kidneys (WRK). Molecularly imprinted microspheres with nanoporous and multilayered structures are prepared based on liquid–liquid phase separation (LLPS), which can selectively adsorb urea. In addition, we combine the microspheres with a designed polydimethylsiloxane (PDMS) chip to propose an efficient urea adsorption platform. In this work, we propose a formulation of LLPS including Tripropylene glycol diacrylate (TPGDA), ethanol, and acrylic acid (30% v/v), to prepare urea molecularly imprinted microspheres in a simple and highly controllable method. These microspheres have urea molecular imprinting sites on the surface and inside, allowing selective adsorption of urea and preservation of other essential constituents. Previous static studies on urea adsorption have not considered the combination between urea adsorbent and WRK. Therefore, we design the platform embedded with urea molecular imprinted microspheres, which can disturb the fluid motion and improve the efficiency of urea adsorption. These advantages enable the urea absorption platform to be highly promising for dialysate regeneration in WRK.