Flow Alignment Research Articles

Evaluation of ploidy and the DNA index by flow cytometry in central nervous system tumors: a review.

Research on central nervous system tumors (CNSTs) has a significant impact on the diagnosis and prognosis of patients. Currently, CNSTs are classified according to the schema proposed by the World Health Organization (WHO), which considers clinical, histopathological, and molecular characteristics, highlighting the importance of tumor biology for accurate diagnosis and optimal treatment approaches. Despite these advances, assessing DNA ploidy-a marker of tumor aggressiveness-remains complex in CNSTs. This review investigates the utility of DNA index (DNAi) and DNA ploidy analysis by flow cytometry in diagnosing CNSTs and prognosing their outcomes. We systematically reviewed studies in the PubMed database from 1990 to the present using the keywords "DNA Index", "Brain", "Flow cytometry", and "Ploidy". We identified 151 studies, 36 of which met our inclusion criteria. We found considerable variation in sample sizes and methodological variation across the studies. Discrepancies between the reported DNAi and ploidy values were observed. Aneuploidy is generally associated with more aggressive tumors, although exceptions exist. Higher DNAi levels correlate with increased malignancy, notably in glioblastomas, astrocytomas, and meningiomas, whereas diploid astrocytomas and oligodendrogliomas are associated with shorter survival rates. DNA ploidy assessment via flow cytometry could predict CNST behavior, yet methodological issues with tissue selection, adequate control samples, and technique variability remain. DNAi and ploidy assessments show promise as prognostic markers in CNSTs. However, the standardization of flow cytometry protocols and alignment with the current WHO classification schema are essential steps to integrate ploidy analysis in routine CNST assessment.

Graphene Oxide and its Composites: Advanced Membranes for Selective Water Permeation.

Graphene oxide (GO) membranes have gained significant attention as a promising material for separation by selective permeation processes due to their advantageous structural and chemical properties, including high water permeability, chemical resistance, and mechanical strength. In this study, we explore the potential applications of GO membranes in pervaporation to separate liquid mixtures. The layered structure and hydrophilic nature of GO membrane facilitate rapid and selective water transport through angstrom-scale interlayer spacings, resulting in superior performance over conventional polymeric and inorganic membranes. The unique mass transport mechanisms - slip flow and molecular alignment - enable GO membranes to selectively permeate water over organic solvents. For chemical dehydration, GO membranes are the most potential candidates. Furthermore, advancements in composite GO membranes and cross-linking techniques that improve their stability and separation performance are discussed. This study highlights the advantages of GO membranes and their potential to replace or complement existing technologies, by emphasizing their role in advancing membrane-based separation and promoting environmental sustainability. Future research is expected to optimize the fabrication techniques for GO membranes and expand their application scope.

A Geometric Significance-Aware Deep Mutual Learning Network for Building Extraction from Aerial Images

Knowledge-driven building extraction method exhibits a restricted adaptability scope and is vulnerable to external factors that affect its extraction accuracy. On the other hand, data-driven building extraction method lacks interpretability, heavily relies on extensive training data, and may result in extraction outcomes with building boundary blur issues. The integration of pre-existing knowledge with data-driven learning is essential for the intelligent identification and extraction of buildings from high-resolution aerial images. To overcome the limitations of current deep learning building extraction networks in effectively leveraging prior knowledge of aerial images, a geometric significance-aware deep mutual learning network (GSDMLNet) is proposed. Firstly, the GeoSay algorithm is utilized to derive building geometric significance feature maps as prior knowledge and integrate them into the deep learning network to enhance the targeted extraction of building features. Secondly, a bi-directional guidance attention module (BGAM) is developed to facilitate deep mutual learning between the building feature map and the building geometric significance feature map within the dual-branch network. Furthermore, the deployment of an enhanced flow alignment module (FAM++) is utilized to produce high-resolution, robust semantic feature maps with strong interpretability. Ultimately, a multi-objective loss function is crafted to refine the network’s performance. Experimental results demonstrate that the GSDMLNet excels in building extraction tasks within densely populated and diverse urban areas, reducing misidentification of shadow-obscured regions and color-similar terrains lacking building structural features. This approach effectively ensures the precise acquisition of urban building information in aerial images.

Combining optical flow and Swin Transformer for Space-Time video super-resolution

Space–time video super-resolution is a task that aims to interpolate low frame rate, low resolution videos to high frame rate, high resolution ones. While existing Transformer-based methods have achieved results comparable to convolutional neural networks-based methods, the computational cost of Transformer limits its performance with constrained computational resources. Moreover, Swin Transformer may fail to fully exploit the spatio-temporal information of video frames due to the limitation of window size, impeding its effectiveness in handling large motions. To address these limitations, we propose an end-to-end space–time video super-resolution architecture based on optical flow alignment and Swin Transformer. The alignment module is introduced to extract spatio-temporal information from adjacent frames without significantly increasing the computational burden. Additionally, we design a residual convolution layer to enhance the translational invariance of the features extracted by Swin Transformer and introduces additional nonlinear transformations. Experimental results demonstrate that our proposed method achieves superior performance on various benchmark datasets compared to state-of-the-art methods. In terms of Peak Signal-to-Noise Ratio, our method outperforms the state-of-the-art methods by at least 0.15 dB on Vimeo-Medium dataset.

Exploring the potential of graphene oxide membranes for vapor permeation

This review explores the applicability of graphene oxide (GO) membranes in vapor permeation, emphasizing their unique properties and promising capabilities. In the context of eco-friendly separation processes, such as pervaporation (PV) and vapor permeation (VP), GO membranes can selectively permeate water vapor to separate vapor mixtures. Their distinct structure, featuring slip flow and precise water molecule alignment in angstrom-scale interlayer spaces, facilitates fast water transport. This review proposes the selective adsorption–pore flow–evaporation model as a plausible mechanism for vapor permeation through GO membranes. Despite extensive research on GO membranes for PV in separating mixtures, their application in VP remains underexplored. Nevertheless, their selectivity for water vapor and effectiveness in gas dehumidification processes emphasize their applicability for chemical dehydration (purification). Future research in this field can advance membrane technologies, offering more sustainable and efficient separation processes across various industries for purifying and recycling chemicals and separating azeotropic mixtures.

Under pressure: integrated endothelial cell response to hydrostatic and shear stresses.

Blood flow within the vasculature is a critical determinant of endothelial cell (EC) identity and functionality, yet the intricate interplay of various hemodynamic forces and their collective impact on endothelial and vascular responses are not fully understood. Specifically, the role of hydrostatic pressure in the EC flow response is understudied, despite its known significance in vascular development and disease. To address this gap, we developed in vitro models to investigate how pressure influences EC responses to flow. Our study demonstrates that elevated pressure conditions significantly modify shear-induced flow alignment and increase endothelial cell density. Bulk and single-cell RNA sequencing analyses revealed that, while shear stress remains the primary driver of flow-induced transcriptional changes, pressure modulates shear-induced signaling in a dose-dependent manner. These pressure-responsive transcriptional signatures identified in human ECs were conserved during the onset of circulation in early mouse embryonic vascular development, where pressure was notably associated with transcriptional programs essential to arterial and hemogenic EC fates. Our findings suggest that pressure plays a synergistic role with shear stress on ECs and emphasizes the need for an integrative approach to endothelial cell mechanotransduction, one that encompasses the effects induced by pressure alongside other hemodynamic forces.

Comparison of the self-assembly and cytocompatibility of conjugates of Fmoc (9-fluorenylmethoxycarbonyl) with hydrophobic, aromatic, or charged amino acids.

The self-assembly in aqueous solution of three Fmoc-amino acids with hydrophobic (aliphatic or aromatic, alanine or phenylalanine) or hydrophilic cationic residues (arginine) is compared. The critical aggregation concentrations were obtained using intrinsic fluorescence or fluorescence probe measurements, and conformation was probed using circular dichroism spectroscopy. Self-assembled nanostructures were imaged using cryo-transmission electron microscopy and small-angle X-ray scattering (SAXS). Fmoc-Ala is found to form remarkable structures comprising extended fibril-like objects nucleating from spherical cores. In contrast, Fmoc-Arg self-assembles into plate-like crystals. Fmoc-Phe forms extended structures, in a mixture of straight and twisted fibrils coexisting with nanotapes. Spontaneous flow alignment of solutions of Fmoc-Phe assemblies is observed by SAXS. The cytocompatibility of the three Fmoc-amino acids was also compared via MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] mitochondrial activity assays. All three Fmoc-amino acids are cytocompatible with L929 fibroblasts at low concentration, and Fmoc-Arg shows cell viability up to comparatively high concentration (0.63 mM).

Hybrid control of aerofoil self-noise by coupling air blowing and trailing-edge serration

Active and passive control methods have been extensively studied and employed for reducing turbulent-boundary-layer (TBL) trailing-edge (TE) noise of aerofoils, for example air blowing and TE serration. To further explore the potential of noise reduction, this paper proposes a hybrid method coupling blowing and serration at the trailing edge, and investigates the effects of air blowing, TE serration, and the hybrid blowing-serration (HBS) on the TBL TE noise of a NACA0012 aerofoil at a Reynolds number of Rec=4×105. Large eddy simulation and Ffowcs Williams–Hawkings acoustic analogy were used to analyse these effects and the respective noise reduction mechanisms. The numerical results show that the HBS method exhibits superior acoustic performance to the blowing and serration methods, with a maximum reduction of 20.5 dB in broadband noise and 5.7 dB in overall sound pressure level. It is revealed that the HBS method functions primarily through three noise reduction mechanisms. Firstly, blowing eliminates the outward flow motion at the serration root and tip, resulting in a forced flow alignment. Secondly, it weakens the turbulence energy in the near wake as for air blowing. Finally, this method mitigates spanwise coherence at the trailing edge, thereby causing destructive interference of noise sources, similar to the effect of TE serration.

Morphology of poly-3-hexyl-thiophene blends with styrene-isoprene-styrene block-copolymer elastomers from X-ray and neutron scattering.

The nano- and micron scale morphology of poly(3-hexylthiophene) (P3HT) and polystyrene-block-polyisoprene-block-polystyrene (PS-PI-PS) elastomeric blends is investigated through the use of ultra-small and small angle X-ray and neutron scattering (USAXS, SAXS, SANS). It is demonstrated that loading P3HT into elastomer matrices is possible with little distortion of the elastomeric structure up to a loading of ∼5 wt%. Increased loadings of conjugated polymer is found to significantly distort the matrix structure. Changes in processing conditions are also found to affect the blend morphology with especially strong dependence on processing temperature. Processing temperatures above the glass transition temperature (Tg) of polystyrene and the melting temperature (Tm) of the conjugated polymer additive (P3HT) creates significantly more organized mesophase domains. P3HT blends with PS-PI-PS can also be flow-aligned through processing, which results in an anisotropic structure that could be useful for the generation of anisotropic properties (e.g. conductivity). Moreover, the extent of flow alignment is significantly affected by the P3HT loading in the PS-PI-PS matrix. The work adds insight to the morphological understanding of a complex P3HT and PS-PI-PS polymer blend as conjugated polymer is added to the system. We also provide studies isolating the effect of processing changes aiding in the understanding of the structural changes in this elastomeric conjugated polymer blend.

Self-Supervised Learning for Real-World Super-Resolution from Dual and Multiple Zoomed Observations.

In this paper, we consider two challenging issues in reference-based super-resolution (RefSR) for smartphone, (i) how to choose a proper reference image, and (ii) how to learn RefSR in a self-supervised manner. Particularly, we propose a novel self-supervised learning approach for real-world RefSR from observations at dual and multiple camera zooms. Firstly, considering the popularity of multiple cameras in modern smartphones, the more zoomed (telephoto) image can be naturally leveraged as the reference to guide the super-resolution (SR) of the lesser zoomed (ultra-wide) image, which gives us a chance to learn a deep network that performs SR from the dual zoomed observations (DZSR). Secondly, for self-supervised learning of DZSR, we take the telephoto image instead of an additional high-resolution image as the supervision information, and select a center patch from it as the reference to super-resolve the corresponding ultra-wide image patch. To mitigate the effect of the misalignment between ultra-wide low-resolution (LR) patch and telephoto ground-truth (GT) image during training, we first adopt patch-based optical flow alignment to obtain the warped LR, then further design an auxiliary-LR to guide the deforming of the warped LR features. To generate visually pleasing results, we present local overlapped sliced Wasserstein loss to better represent the perceptual difference between GT and output in the feature space. During testing, DZSR can be directly deployed to super-solve the whole ultra-wide image with the reference of the telephoto image. In addition, we further take multiple zoomed observations to explore self-supervised RefSR, and present a progressive fusion scheme for the effective utilization of reference images. Experiments show that our methods achieve better quantitative and qualitative performance against state-of-the-arts. The code and pre-trained models will be publicly available.

Multi-scale feature flow alignment fusion with Transformer for the microscopic images segmentation of activated sludge

Accurate microscopic images segmentation of activated sludge is essential for monitoring wastewater treatment processes. However, it is a challenging task due to poor contrast, artifacts, morphological similarities, and distribution imbalance. A novel image segmentation model (FafFormer) was developed in the work based on Transformer that incorporated pyramid pooling and flow alignment fusion. Pyramid Pooling Module was used to extract multi-scale features of flocs and filamentous bacteria with different morphology in the encoder. Multi-scale features were fused by flow alignment fusion module in the decoder. The module used generated semantic flow as auxiliary information to restore boundary details and facilitate fine-grained upsampling. The Focal–Lovász Loss was designed to handle class imbalance for filamentous bacteria and flocs. Image-segmentation experiments were conducted on an activated sludge dataset from a municipal wastewater treatment plant. FafFormer showed relative superiority in accuracy and reliability, especially for filamentous bacteria compared to existing models.

SGNet: a fast and accurate semantic segmentation network based on semantic guidance

ABSTRACT In this paper, we design a fast and accurate lightweight semantic segmentation method for mobile robots. Accurate semantic segmentation usually requires obtaining high-resolution feature maps with strong semantic representation. Among the widely utilized methods, dilated convolution is computationally intensive and the feature pyramid accuracy is not as accurate. Inspired by the Flow Alignment Module (FAM) proposed by SFNet, we propose a Semantic Guided Upsampling Module (SGUM) for learning semantic offsets between neighboring level feature maps to solve the semantic misalignment problem and effectively fuse high level semantic features into low-level high-resolution feature maps. The integration of the SGUM into the semantic segmentation network with different backbones shows better performance than other methods on various datasets, showing the effectiveness of the SGUM on semantic feature fusion as well as its generalizability.

Sfnet: Faster and Accurate Semantic Segmentation Via Semantic Flow

In this paper, we focus on exploring effective methods for faster and accurate semantic segmentation. A common practice to improve the performance is to attain high-resolution feature maps with strong semantic representation. Two strategies are widely used: atrous convolutions and feature pyramid fusion, while both are either computationally intensive or ineffective. Inspired by the Optical Flow for motion alignment between adjacent video frames, we propose a Flow Alignment Module (FAM) to learn Semantic Flow between feature maps of adjacent levels and broadcast high-level features to high-resolution features effectively and efficiently. Furthermore, integrating our FAM to a standard feature pyramid structure exhibits superior performance over other real-time methods, even on lightweight backbone networks, such as ResNet-18 and DFNet. Then to further speed up the inference procedure, we also present a novel Gated Dual Flow Alignment Module to directly align high-resolution feature maps and low-resolution feature maps where we term the improved version network as SFNet-Lite. Extensive experiments are conducted on several challenging datasets, where results show the effectiveness of both SFNet and SFNet-Lite. In particular, when using Cityscapes test set, the SFNet-Lite series achieve 80.1 mIoU while running at 60 FPS using ResNet-18 backbone and 78.8 mIoU while running at 120 FPS using STDC backbone on RTX-3090. Moreover, we unify four challenging driving datasets (i.e., Cityscapes, Mapillary, IDD, and BDD) into one large dataset, which we named Unified Driving Segmentation (UDS) dataset. It contains diverse domain and style information. We benchmark several representative works on UDS. Both SFNet and SFNet-Lite still achieve the best speed and accuracy trade-off on UDS, which serves as a strong baseline in such a challenging setting. The code and models are publicly available at https://github.com/lxtGH/SFSegNets.

Multi-focus image fusion using structure-guided flow

Existing deep learning based methods have shown their advantages in multi-focus image fusion task. However, most methods still suffer from inaccurate focus region detection. In this paper, we employ the property of part-whole relationships embedded by the Capsule Network (CapsNet) to solve the problem. Specifically, we introduce CapsNet in multi-focus image fusion task, and design a structure-guided flow module, which fully utilizes structure information to help locate focus regions. CapsNet is introduced to extract structure features by supervising gradient information of the image. Compared with traditional convolutional neural networks (CNNs), CapsNet takes into account the correlation of features from different positions, such that it encodes more compact features. Once structure features are obtained, a flow alignment module is introduced to learn flow field between structure and image features, and propagate effectively structure features to image features to make confident focus region detection. Experimental results show the proposed method achieves robust fusion performance on three publicly available multi-focus datasets, and outperforms or is comparable to the state-of-the-art methods.

Influence of the spatial variation of upstream velocity on a vertical-axis tidal turbine performance

Assessment of the upstream flow available to the tidal energy converters (TEC) is key to evaluate its performance. Simultaneously, TEC technology has been innovating on its concepts and designs to expand the potential sites to harvest energy generated by tidal currents and rivers. The Gkinetic CEFA 12 is an easy-to-deploy device suitable to operate on estuary environments. The design consists of two 1.2 m vertical axis turbines attached to the sides of a buoyant platform, which uses a bluff body to accelerate the incoming flow to the rotors. The device was deployed on a single point mooring enabling passive flow alignment. As part of the Vertical axis tidal turbines in Strangford lough project (VATTS), the upstream flow has been measured using acoustic doppler profilers (ADP) mounted on the TEC during its operation in Strangford lough. Recommendations of IEC-200 were followed when mounting the ADPs relative to Gkinetic. However, the continuous repositioning of the TEC according to the prevailing tidal regime affects the ADPs heading. The implications of the spatial variation of the rotor’s upstream velocity on the resource assessment and device performance are not clear since this situation is not typical.To investigate the spatial variation of the upstream velocity two ADCP deployment locations were made. A seabed (upward facing) ADCP as per the current standards and a device mounted (downward facing) ADCP upstream of the rotor plane. Evaluation of the influence of the following three factors were made: i) the ADPs direction repositioning according to prevailing tidal regime, ii) the proximity of the sensors to the sea-surface, and iii) the proximity of the sensors to the TEC.These three factors will be evaluated using ADP datasets collected at Strangford narrows during the VATTS project. The datasets were obtained at approximately the same location of the TEC operation. They enable the study of the following scenarios 1) incoming flow to the rotors during operation, 2) incoming flow on undisturbed conditions (no turbine operation), and 3) a harmonic analysis prediction of the undisturbed incoming flow, which solely captures the tidal-driven flow.The investigation of these three scenarios will provide a better understanding on the rotor’s upstream flow spatial variation, and the influence of the device’s proximity and near sea surface conditions on the mean flow. These findings would benefit developers of alternative TEC designs that operate near the sea-surface.

High-Resolution Remote Sensing Image Change Detection Method Based on Improved Siamese U-Net

Focusing on problems of blurred detection boundary, small target miss detection, and more pseudo changes in high-resolution remote sensing image change detection, a change detection algorithm based on Siamese neural networks is proposed. Siam-FAUnet can implement end-to-end change detection tasks. Firstly, the improved VGG16 is utilized as an encoder to extract the image features. Secondly, the atrous spatial pyramid pooling module is used to increase the receptive field of the model to make full use of the global information of the image and obtain the multi-scale contextual information of the image. The flow alignment module is used to fuse the low-level features in the encoder to the decoder and solve the problem of semantic misalignment caused by the direct concatenation of features when the features are fused, so as to obtain the change region of the image. The experiments are trained and tested using publicly available CDD and SZTAKI datasets. The results show that the evaluation metrics of the Siam-FAUnet model are improved compared to the baseline model, in which the F1-score is improved by 4.00% on the CDD and by 7.32% and 2.62% on the sub-datasets of SZTAKI (SZADA and TISZADOB), respectively; compared to other state-of-the-art methods, the Siam-FAUnet model has improved in both evaluation metrics, indicating that the model has a good detection performance.

Flow-Induced Protein Chain Deformation, Segmental Orientation, and Phase Separation in Native Silk Feedstock.

The ability of many arthropods to spin silk and its many uses bear testament to its importance in Nature. Despite over a century of research, however, the spinning process is still not fully understood. While it is widely accepted that flow and chain alignment may be involved, the link to protein gelation remains obscure. Using combinations of rheology, polarized light imaging, and infrared spectroscopy to probe different length scales, this work explored flow-induced gelation of native silk feedstock from Bombyx mori larvae. Protein chain deformation, orientation, and microphase separation were observed, culminating in the formation of antiparallel β-sheet structures while the work rate during flow appeared as an important criterion. Moreover, infrared spectroscopy provided direct observations suggesting a loss of protein hydration during flow-induced gelation of fibroin in native silk feedstock, which is consistent with recently reported hypotheses.

Enhancing the Interlaminar Shear Strength and Void Control of 3D-Printed Continuous Carbon-Fiber-Reinforced Polymer Composites Using a Robotic Magnetic Compaction Force-Assisted Additive Manufacturing (MCFA-AM) Process and Carbon-Nanofiber Z-Threads

Three-dimensional (3D) printing with continuous carbon-fiber-reinforced polymer (C-CFRP) composites is under increasing development, as it offers more versatility than traditional molding processes, such as the out-of-autoclave-vacuum bag only (OOA-VBO) process. However, due to the layer-by-layer deposition of materials, voids can form between the layers and weaken some of the parts’ properties, such as the interlaminar shear strength (ILSS). In this paper, a novel mold-less magnetic compaction force-assisted additive manufacturing (MCFA-AM) method was used to print carbon nanofiber (CNF) z-threaded CFRP (ZT-CFRP) laminates with significantly improved ILSS and reduced void content compared to traditional C-CFRP laminates, which are printed using a no-pressure 3D-printing process (similar to the fused-deposition-modeling process). The radial flow alignment (RFA) and resin-blending techniques were utilized to manufacture a printing-compatible fast-curing ZT-CFRP prepreg tape to act as the feedstock for a MCFA-AM printhead, which was mounted on a robotic arm. In terms of the ILSS, the MCFA-AM method coupled with ZT-CFRP nanomaterial technology significantly outperformed the C-CFRP made with both the traditional no-pressure 3D-printing process and the OOA-VBO molding process. Furthermore, the mold-less MCFA-AM process more than doubled the production speed of the OOA-VBO molding process. This demonstrates that through the integration of new nanomaterials and 3D-printing techniques, a paradigm shift in C-CFRP manufacturing with significantly better performance, versatility, agility, efficiency, and lower cost is achievable.

Rethinking 1D convolution for lightweight semantic segmentation.

Lightweight semantic segmentation promotes the application of semantic segmentation in tiny devices. The existing lightweight semantic segmentation network (LSNet) has the problems of low precision and a large number of parameters. In response to the above problems, we designed a full 1D convolutional LSNet. The tremendous success of this network is attributed to the following three modules: 1D multi-layer space module (1D-MS), 1D multi-layer channel module (1D-MC), and flow alignment module (FA). The 1D-MS and the 1D-MC add global feature extraction operations based on the multi-layer perceptron (MLP) idea. This module uses 1D convolutional coding, which is more flexible than MLP. It increases the global information operation, improving features' coding ability. The FA module fuses high-level and low-level semantic information, which solves the problem of precision loss caused by the misalignment of features. We designed a 1D-mixer encoder based on the transformer structure. It performed fusion encoding of the feature space information extracted by the 1D-MS module and the channel information extracted by the 1D-MC module. 1D-mixer obtains high-quality encoded features with very few parameters, which is the key to the network's success. The attention pyramid with FA (AP-FA) uses an AP to decode features and adds a FA module to solve the problem of feature misalignment. Our network requires no pre-training and only needs a 1080Ti GPU for training. It achieved 72.6 mIoU and 95.6 FPS on the Cityscapes dataset and 70.5 mIoU and 122 FPS on the CamVid dataset. We ported the network trained on the ADE2K dataset to mobile devices, and the latency of 224 ms proves the application value of the network on mobile devices. The results on the three datasets prove that the network generalization ability we designed is powerful. Compared to state-of-the-art lightweight semantic segmentation algorithms, our designed network achieves the best balance between segmentation accuracy and parameters. The parameters of LSNet are only 0.62 M, which is currently the network with the highest segmentation accuracy within 1 M parameters.

Dendritic patterns from shear-enhanced anisotropy in nematic liquid crystals.

Controlling the growth morphology of fluid instabilities is challenging because of their self-amplified and nonlinear growth. The viscous fingering instability, which arises when a less viscous fluid displaces a more viscous one, transitions from exhibiting dense-branching growth characterized by repeated tip splitting of the growing fingers to dendritic growth characterized by stable tips in the presence of anisotropy. We controllably induce such a morphology transition by shear-enhancing the anisotropy of nematic liquid crystal solutions. For fast enough flow induced by the finger growth, the intrinsic tumbling behavior of lyotropic chromonic liquid crystals can be suppressed, which results in a flow alignment of the material. This microscopic change in the director field occurs as the viscous torque from the shear flow becomes dominant over the elastic torque from the nematic potential and macroscopically enhances the liquid crystal anisotropy to induce the transition to dendritic growth.