Post-processing Step Research Articles

Optical Fresnel zone plate flat lenses made entirely of colored photoresist through an i-line stepper

Light manipulation and control are essential in various contemporary technologies, and as these technologies evolve, the demand for miniaturized optical components increases. Planar-lens technologies, such as metasurfaces and diffractive optical elements, have gained attention in recent years for their potential to dramatically reduce the thickness of traditional refractive optical systems. However, their fabrication, particularly for visible wavelengths, involves complex and costly processes, such as high-resolution lithography and dry-etching, which has limited their availability. In this study, we present a simplified method for fabricating visible Fresnel zone plate (FZP) planar lenses, a type of diffractive optical element, using an i-line stepper and a special photoresist (color resist) that only necessitates coating, exposure, and development, eliminating the need for etching or other post-processing steps. We fabricated visible FZP lens patterns using conventional photolithography equipment on 8-inch silica glass wafers, and demonstrated focusing of 550 nm light to a diameter of 1.1 μm with a focusing efficiency of 7.2%. Numerical simulations showed excellent agreement with experimental results, confirming the high precision and designability of our method. Our lenses were also able to image objects with features down to 1.1 μm, showcasing their potential for practical applications in imaging. Our method is a cost-effective, simple, and scalable solution for mass production of planar lenses and other optical components operating in the visible region. It enables the development of advanced, miniaturized optical systems to meet modern technology demand, making it a valuable contribution to optical component manufacturing.

Artificial Intelligence in Radiology: A Leadership Survey.

Surveys to assess views about artificial intelligence (AI) of various diagnostic radiology constituencies have revealed interesting combinations of enthusiasm, caution, and implementation priorities. We surveyed academic radiology leaders about their views on AI and how they intend to approach AI implementation in their departments. We conducted a web survey of Society of Chairs of Academic Radiology Departments (SCARD) members between October 5 and October 31, 2023 to solicit optimism or pessimism about AI, target use cases, planned implementation, and perceptions of their workforce. P-values are provided only for descriptive purposes and have not been adjusted for multiple testing in this exploratory research. The survey was sent to the 112 SCARD members and 43 responded (38%). Chairs were optimistic, with no statistical difference between views of AI in general versus generative AI. Chairs plan to implement AI to improve quality and efficiency (43/43, 100%), burnout (41/43, 95%), healthcare costs (22/43, 51%), and equity (27/43, 63%) and most likely will target the post-processing (26/43, 60%), interpretation workflow (26/43, 60%), and image acquisition (18/43, 42%) steps in the imaging value chain. Chairs perceived that radiologists (36/43, 84%) and technologists (38/43, 88%) were not particularly worried about being displaced but saw trainees as slightly less confident (31/43, 72%). Free text responses revealed concerns about the cost of AI and emphasized trade-offs that needed to be balanced. Radiology Chairs are optimistic about AI and poised to tackle departmental challenges. Concerns about generative AI and workforce replacement are minimal.

Semblans: Automated assembly and processing of RNA-Seq data.

Recent advancements in parallel sequencing methods have precipitated a surge in publicly available short-read sequence data. This has encouraged the development of novel computational tools for the de novo assembly of transcriptomes from RNA-seq data. Despite the availability of these tools, performing an end-to-end transcriptome assembly remains a programmatically involved task necessitating familiarity with best practices. Aside from quality control steps, including error correction, adapter trimming, and chimera filtration needing to be correctly employed, moving data between programs often requires manual reformatting or restructuring, which can further impede throughput. Here, we introduce Semblans, a tool for streamlining the assembly process that efficiently and consistently produces high-quality transcriptome assemblies. Semblans abstracts the key quality control, reconstitution, and postprocessing steps of transcriptome assembly from raw short-read sequences to annotated coding sequences. Evaluating its performance against previously assembled transcriptomes on the basis of assembly quality, we find that Semblans produced higher quality assemblies for 98 of the 101 short-read runs tested. Semblans is written in C ++ and runs on Unix-compliant operating systems. Source code, documentation, and compiled binaries are hosted under the GNU General Public License at https://github.com/gladshire/Semblans. Supplementary data are available at Journal Name online.

AB-Type Zwitterionic Hydrogel Paint.

Zwitterionic hydrogels exhibit excellent nonfouling and hemocompatibility. However, the practical application of these materials as antifouling coatings for biomedical devices is hindered by several key challenges, including the harsh preparation conditions and the weak coating stability. Here, we present a two-component zwitterionic hydrogel paint for the in situ preparation of robust zwitterionic hydrogel coatings on various substrate surfaces without UV assistance. It is performed by the curing and adhesion of a zwitterionic hydrogel simultaneously through the ring opening reaction of epoxy and amino inspired by the successful commercial two-component epoxy structural glue. The obtained AB-type PSBMA coating can withstand water flow velocities of up to 15 m/s and still maintain its structural integrity and functional stability. It is noteworthy that the coating preparation process does not require the use of any organic solvent, which greatly simplifies the postprocessing steps for its application in medical devices. Moreover, the coating not only resists bacterial and cell adhesion but also exhibits favorable hemocompatibility. This approach offers a novel concept for the design of zwitterionic hydrogel coatings for biomedical devices.

RegBoost: Enhancing mouse brain image registration using geometric priors and Laplacian interpolation.

We show in this work that incorporating geometric features and geometry processing algorithms for mouse brain image registration broadens the applicability of registration algorithms and improves the registration accuracy of existing methods. We introduce the preprocessing and postprocessing steps in our proposed framework as RegBoost. We develop a method to align the axis of 3D image stacks by detecting the central planes that pass symmetrically through the image volumes. We then find geometric contours by defining external and internal structures to facilitate image correspondences. We establish Dirichlet boundary conditions at these correspondences and find the displacement map throughout the volume using Laplacian interpolation. We discuss the challenges in our standalone framework and demonstrate how our new approaches can improve the results of existing image registration methods. We expect our new approach and algorithms will have critical applications in brain mapping projects.

The Effect of Pre- and Post-Processing Techniques on Tree Detection in Young Forest Stands from Images of Snow Cover Using YOLO Neural Networks

A neural network model for individual tree detection was developed based on the YOLOv4 architecture, which underwent additional preprocessing and postprocessing steps. The preprocessing step involved expanding the dataset by randomly cutting fragments from images, calculating anchor box sizes using the K-means clustering algorithm, and discarding anchor boxes that were too small a priori. The existing post-processing block of the YOLO architecture was modified by giving more weight to false positives in the error function and using the non-maximum suppression algorithm. Baseline neural networks from the YOLOv4 and YOLOv5 architectures, each in two versions (pre-trained and not pre-trained on the MS COCO dataset), were used for comparison without any additional modifications. In the overgrown experimental field, multi-season aerial copter surveys and ground counts were conducted on several sample plots to gather data. Comparison of multi-season aerial photographs with ground-count data showed that the best images in terms of the percentage of visually identifiable trees were those taken during the snowy season and when there was no foliage. Using these images and some additional images, we manually created a dataset on which we trained and tested neural network models. The model we developed showed significantly better results (2 to 10 times better) on the mAP 0.5 metric compared to the alternatives we considered.

Additive manufacturing of 3D flow-focusing millifluidics for the production of curable microdroplets.

Microfluidics plays a crucial role in the generation of mono-sized microdroplet emulsions. Traditional glass microfluidic chips typically lack versatility in generating curable droplets of arbitrary liquids due to the inherent hydrophilic nature of glass and to fabrication constraints. To overcome this, we designed a microdroplet generator with 3D flow-focusing capabilities that can be 3D-printed. The chip can handle oil-in-water emulsions despite its lipophilicity. Operating in the jetting regime, the chip exploits the Rayleigh-Plateau instability to enable high throughput. With its versatile design, the chip is capable of producing both single and double emulsions within the same channel. We utilize a thermoset (epoxy-melamine) based system to test its ability to handle curable chemicals and to produce in a post-processing step both solid particles and filled capsules. With a low solvent concentration in the curable material, the present system can encapsulate water-based cores of a wide range of sizes.

Dual Input Stream Transformer for Vertical Drift Correction in Eye-Tracking Reading Data.

We introduce a novel Dual Input Stream Transformer (DIST) for the challenging problem of assigning fixation points from eye-tracking data collected during passage reading to the line of text that the reader was actually focused on. This post-processing step is crucial for analysis of the reading data due to the presence of noise in the form of vertical drift. We evaluate DIST against eleven classical approaches on a comprehensive suite of nine diverse datasets. We demonstrate that combining multiple instances of the DIST model in an ensemble achieves high accuracy across all datasets. Further combining the DIST ensemble with the best classical approach yields an average accuracy of 98.17%. Our approach presents a significant step towards addressing the bottleneck of manual line assignment in reading research. Through extensive analysis and ablation studies, we identify key factors that contribute to DIST's success, including the incorporation of line overlap features and the use of a second input stream. Via rigorous evaluation, we demonstrate that DIST is robust to various experimental setups, making it a safe first choice for practitioners in the field.

A 3D printed microfluidic device for centrifugal droplet generation

PurposeThis study aims to use an additive process for the first time to develop a microfluidic device that uses centrifugal technique for precise and repeatable generation of microdroplets. Droplets have versatile applications in life sciences, but so far centrifugal devices for their production have been made mainly using standard subtractive techniques. This study focused on evaluating the applicability of 3D printing technology in the development of centrifugal microfluidic devices and investigating their properties and future applications.Design/methodology/approachFirst, the background of this interdisciplinary research, including the principle of droplet microfluidics and the centrifugal technique, is explained. The developed device has the form of a disc (similar to an audio CD), containing an integrated microfluidic system for droplet generation. The disc is rotated at a specific spin profile to induce controlled liquid flow and accurate production of oil-in-water microdroplets. The device was fabricated using material jetting technology. The design, operation principles, printing process parameters and post-processing steps are explained in detail.FindingsThe device was thoroughly characterised, including its mechanical properties, the impact of chemical treatment and the flow measurement of the liquids. The study confirms that the disc can be applied to produce various emulsions using centrifugal force alone. 3D printing technology enables potential mass production and other applications of the device.Originality/valueThe 3D printing process allowed for easy design, fabrication and duplication of the device. Compared to standard PMMA discs, a simpler fabrication protocol and a more flexible and monolithic structure were obtained. The device can be adapted to other microfluidic processes in a lab with high potential for point-of-care applications.

Microfluidic wet spinning of soft polydimethylsiloxane polymer optical fibers

Polymer optical fibers (POFs) are an essential component of photonic textile sensors for the development of healthcare@home technologies. However, fabricating tailored POFs exhibiting the required properties for such applications remains challenging. Here, an innovative method to fabricate soft POFs based on polydimethylsiloxane is introduced: the hydrogel-assisted microfluidic wet spinning (HA-MWS) technology. Combined with a straightforward post-processing step, the HA-MWS enabled the production of soft POFs with tailored moduli (0.35 MPa to 5.0 MPa), low surface roughness (Rq< 6 nm), and, consequently, low attenuation (<0.25 dB/cm). The quasi-static and dynamic mechanical, chemical, optical, thermal, and surface properties of the soft POFs produced by HA-MWS were investigated in detail. A photonic textile sensor demonstrator was built with these soft POFs with tailored pressure sensitivity in the range of 2–300 kPa, modulus close to that of skin, and high-temperature stability between -30C∘ and 90C∘. This methodology can potentially become a standard tool for designing POFs with tunable properties for healthcare monitoring, soft robotics, and biomedicine applications.

Differentiable design of compact imaging systems with curved Fresnel optics

Phase elements can improve the performance and reduce the size of imaging systems, thanks to the additional degrees of freedom that are offered by the independent phase gradient on top of a refractive/reflective surface. Possible implementations include diffractive elements or metasurfaces, but these suffer from diffractive dispersion. Similar optical functionality however can be provided by thin, curved Fresnel optics, which solely rely on refraction. In this study, a differentiable raytracing framework is presented that offers precise and rapid optimization of curved Fresnel surfaces, by modeling them as a combination of a distinct geometrical and refractive surface, both differentiable with respect to the imaging merit function. The method is demonstrated by designing a compact imaging and projection lens, both with high numerical aperture. The paper analyzes the impact of Fresnelizing the optimized "theoretical" surfaces on both the imaging performance and transmission efficiency. It furthermore shows how the system performance can be enhanced through dedicated post-processing steps, emphasizing the practical relevance of compact Fresnel optics.

Upsampling Algorithm for V-PCC-Coded 3D Point Clouds

Point cloud (PC) compression is crucial to immersive visual applications such as autonomous vehicles to classify objects on the roads. The Motion Picture Experts Group (MPEG) standardization group has achieved a notable compression efficiency, called video-based PC compression (V-PCC), which consists of an encoder-decoder. The V-PCC encoder takes original 3D PC data and projects them onto multiple 2D planes to generate several 2D feature images. These images are then compressed using the well-established High-Efficiency Video Coding (HEVC) method. The V-PCC decoder uses compressed information and decoding techniques to reconstruct the 3D PC. However, the PCs produced by V-PCC are often sparse, non-uniform, and contain artifacts. In many practical applications, it is necessary to recover complete PCs from partial ones in real time. This article presents a method for enhancing decoded PCs as a post-processing step in the V-PCC with reduced computational time. Our approach involves a 2D upsampling for the V-PCC occupancy image, which increases the density of the PC, and a 2D high-resolution auxiliary information modification algorithm for the 2D-3D conversion of high-resolution 3D PCs, which improves the uniformity and reduces the noise in the PC. The 3D high-resolution PC has been further enhanced using the developed 3D outlier removal and point regeneration algorithm. Our proposed work can significantly simplify the state-of-the-art super resolution methods for PCs and reduce the time complexity of 61–75% while maintaining a high level of quality in PCs.

A direct quantification of numerical dissipation towards improved large eddy simulations

In implicit large eddy simulations (ILES), it becomes increasingly clear that numerical errors are essential to simulation accuracy. Nevertheless, whether the numerical dissipation in a CFD solver can be regarded as a means of turbulence modeling cannot be known apriori. In the present work, we propose a general method to quantify the numerical dissipation rate for arbitrary flow solvers. Unlike previous approaches in which the numerical dissipation is estimated from the perspective of kinetic energy transfer, our method focuses on direct comparisons with the SGS dissipation from explicit models. The new method is both self-contained and self-consistent, which can be applied to any numerical solver through a simple post-processing step in the physical space. We show that for two common techniques to introduce numerical dissipation (through numerical schemes and solution filtering), the quantification results help to determine if a simulation can be considered as a legitimate ILES run and provide direct guidance for designing better models. When the numerical dissipation is already significant, an improved ILES filtering approach is proposed, which reduces the native numerical dissipation and works better for low order codes. The methods are general and work well for different Reynolds numbers, grid resolutions, and numerical schemes.

Guided Game Level Repair via Explainable AI

Procedurally generated levels created by machine learning models can be unsolvable without further editing. Various methods have been developed to automatically repair these levels by enforcing hard constraints during the post-processing step. However, as levels increase in size, these constraint-based repairs become increasingly slow. This paper proposes using explainability methods to identify specific regions of a level that contribute to its unsolvability. By assigning higher weights to these regions, constraint-based solvers can prioritize these problematic areas, enabling more efficient repairs. Our results, tested across three games, demonstrate that this approach can help to repair procedurally generated levels faster.

An alternating minimization algorithm with trajectory for direct exoplanet detection. The AMAT algorithm

Effective image post-processing algorithms are vital for the successful direct imaging of exoplanets. Standard point spread function (PSF) subtraction methods use techniques based on a low-rank approximation to separate the rotating planet signal from the quasi-static speckles and rely on signal-to-noise ratio maps to detect the planet. These steps do not interact or feed each other, leading to potential limitations in the accuracy and efficiency of exoplanet detection. We aim to develop a novel approach that iteratively finds the flux of the planet and the low-rank approximation of quasi-static signals in an attempt to improve upon current PSF subtraction techniques. In this study, we extend the standard L2 norm minimization paradigm to an L1 norm minimization framework in order to better account for noise statistics in the high contrast images. Then, we propose a new method, referred to as the alternating minimization algorithm with trajectory (AMAT), that makes more advanced use of estimating the low-rank approximation of the speckle field and the planet flux by alternating between them and utilizing both L1 and L2 norms. For the L1 norm minimization, we propose using L1 norm low-rank approximation (L1-LRA), a low-rank approximation computed using an exact block-cyclic coordinate descent method, while we use randomized singular value decomposition for the L2 norm minimization. Additionally, we enhance the visibility of the planet signal using a likelihood ratio as a post-processing step. Numerical experiments performed on a VLT/SPHERE-IRDIS dataset show the potential of AMAT to improve upon the existing approaches in terms of higher S/N, sensitivity limits (contrast curves), and receiver operating characteristic curves. Moreover, for a systematic comparison, we used datasets from the exoplanet data challenge to compare our algorithm with other algorithms in the challenge, and we find AMAT with a likelihood ratio map performs better than most algorithms tested on the exoplanet data challenge.

Automated lung segmentation on chest MRI in children with cystic fibrosis.

Segmentation of lung structures in medical imaging is crucial for the application of automated post-processing steps on lung diseases like cystic fibrosis (CF). Recently, machine learning methods, particularly neural networks, have demonstrated remarkable improvements, often outperforming conventional segmentation methods. Nonetheless, challenges still remain when attempting to segment various imaging modalities and diseases, especially when the visual characteristics of pathologic findings significantly deviate from healthy tissue. Our study focuses on imaging of pediatric CF patients [mean age, standard deviation (7.50 ± 4.6)], utilizing deep learning-based methods for automated lung segmentation from chest magnetic resonance imaging (MRI). A total of 165 standardized annual surveillance MRI scans from 84 patients with CF were segmented using the nnU-Net framework. Patient cases represented a range of disease severities and ages. The nnU-Net was trained and evaluated on three MRI sequences (BLADE, VIBE, and HASTE), which are highly relevant for the evaluation of CF induced lung changes. We utilized 40 cases for training per sequence, and tested with 15 cases per sequence, using the Sørensen-Dice-Score, Pearson's correlation coefficient (r), a segmentation questionnaire, and slice-based analysis. The results demonstrated a high level of segmentation performance across all sequences, with only minor differences observed in the mean Dice coefficient: BLADE (0.96 ± 0.05), VIBE (0.96 ± 0.04), and HASTE (0.95 ± 0.05). Additionally, the segmentation quality was consistent across different disease severities, patient ages, and sizes. Manual evaluation identified specific challenges, such as incomplete segmentations near the diaphragm and dorsal regions. Validation on a separate, external dataset of nine toddlers (2-24 months) demonstrated generalizability of the trained model achieving a Dice coefficient of 0.85 ± 0.03. Overall, our study demonstrates the feasibility and effectiveness of using nnU-Net for automated segmentation of lung halves in pediatric CF patients, showing promising directions for advanced image analysis techniques to assist in clinical decision-making and monitoring of CF lung disease progression. Despite these achievements, further improvements are needed to address specific segmentation challenges and enhance generalizability.

Capturing cell heterogeneity in representations of cell populations for image-based profiling using contrastive learning.

Image-based cell profiling is a powerful tool that compares perturbed cell populations by measuring thousands of single-cell features and summarizing them into profiles. Typically a sample is represented by averaging across cells, but this fails to capture the heterogeneity within cell populations. We introduce CytoSummaryNet: a Deep Sets-based approach that improves mechanism of action prediction by 30-68% in mean average precision compared to average profiling on a public dataset. CytoSummaryNet uses self-supervised contrastive learning in a multiple-instance learning framework, providing an easier-to-apply method for aggregating single-cell feature data than previously published strategies. Interpretability analysis suggests that the model achieves this improvement by downweighting small mitotic cells or those with debris and prioritizing large uncrowded cells. The approach requires only perturbation labels for training, which are readily available in all cell profiling datasets. CytoSummaryNet offers a straightforward post-processing step for single-cell profiles that can significantly boost retrieval performance on image-based profiling datasets.

Energy-aware scheduling in flow shops: a novel artificial neural network-driven multi-objective optimization

Group technology is a managerial strategy used to optimize production by reducing setup times, lead times, and work-in-process inventories. Research on flow-shop sequence-dependent group scheduling problems (FSDGSPs) has primarily focused on minimizing makespan and total flow time to improve efficiency. However, the need for energy-efficient scheduling in FSDGSPs remains underexplored despite increasing sustainability concerns. To address this, the energy-efficient flow-shop sequence-dependent group scheduling problem (EEFSDGSP) is introduced. A novel multi-objective optimization (MOO) technique, the artificial neural network-based multi-objective genetic algorithm (ANN-MOGA), is proposed to minimize makespan and energy consumption in EEFSDGSP. ANN-MOGA advances MOO by using a neural network to evaluate fitness and guide selection, reducing computational complexity versus traditional methods like NSGA-II and SPEA2. A post-processing step (PPANNS) further enhances solution diversity and distribution. Results show ANN-MOGA, especially with PPANNS, outperforms NSGA-II and competes effectively with SPEA2 in larger problem instances.

Research on lane detection algorithms based on GTNs

Abstract. Lane detection technology plays an important role in the lane departure warning and adaptive cruise control functions of autonomous vehicle systems. Traditional lane line detection includes experimental steps such as preyreatment, color space conversion, feature extraction, and lane line tracking, but there are still some problems such as recognition accuracy in complex environments and parameter adjustment dependency. To overcome these limitations, this study uses a novel method that can directly predict the parameters of the lane shape model, thus avoiding complex post-processing steps. This research method uses the network structure based on Graph Transformer Networks (GTNS) to capture richer structural information and contextual relationships, thereby improving the accuracy of detection. In terms of feature extraction, Graph Transformer Networks is used to accurately capture the structural information of the lane through the attention mechanism. At the same time, the results in the later stage show that our method has better advantages in accuracy and speed. In short, this optimization scheme not only improves the detection speed but also ensures a certain degree of accuracy. The method will perform better in the future after further optimization such as multi-sensor fusion and real-time optimization.

An automated AI-powered IoT algorithm with data processing and noise elimination for plant monitoring and actuating.

This article aims to develop a novel Artificial Intelligence-powered Internet of Things (AI-powered IoT) system that can automatically monitor the conditions of the plant (crop) and apply the necessary action without human interaction. The system can remotely send a report on the plant conditions to the farmers through IoT, enabling them for tracking the healthiness of plants. Chili plant has been selected to test the proposed AI-powered IoT monitoring and actuating system as it is so sensitive to the soil moisture, weather changes and can be attacked by several types of diseases. The structure of the proposed system is passed through five main stages, namely, AI-powered IoT system design, prototype fabrication, signal and image processing, noise elimination and proposed system testing. The prototype for monitoring is equipped with multiple sensors, namely, soil moisture, carbon dioxide (CO2) detector, temperature, and camera sensors, which are utilized to continuously monitor the conditions of the plant. Several signal and image processing operations have been applied on the acquired sensors data to prepare them for further post-processing stage. In the post processing step, a new AI based noise elimination algorithm has been introduced to eliminate the noise in the images and take the right actions which are performed using actuators such as pumps, fans to make the necessary actions. The experimental results show that the prototype is functioning well with the proposed AI-powered IoT algorithm, where the water pump, exhausted fan and pesticide pump are actuated when the sensors detect a low moisture level, high CO2 concentration level, and video processing-based pests' detection, respectively. The results also show that the algorithm is capable to detect the pests on the leaves with 75% successful rate.