Processing Steps Research Articles

Adaptive Dictionary Identification Framework and its Application to Sparsity-Optimized Harmonic Noise Separation

Coherent noise separation stands as a critical step in seismic data processing. The Morphological Component Analysis (MCA)-based separation method, treating coherent noise and signal as distinct components and representing them sparsely with dictionaries, has found widespread application in the effective suppression of coherent noise. In practice, when constructing effective fixed dictionaries for MCA-based separation methods, a common approach involves conducting numerous experiments to meticulously select appropriate transform basis functions from an extensive dictionary library and tune their parameters. To mitigate time consumption and ensure optimal dictionary construction, we introduce a framework for adaptive identification of the optimal dictionaries used in MCA-based coherent noise separation. Specifically, we initially characterize the notion of a fixed dictionary library, which encompasses dictionaries constructed using various transform basis functions and their corresponding parameters. Then, we present a Relative Sparsity Minimization Problem (RSMP), formulated to identify the optimal fixed dictionaries that lead to minimum relative sparsity within the predefined library. Finally, we design a genetic algorithm to solve RSMP. The optimal dictionaries obtained for representing signal and coherent noise, respectively, are applied to MCA-based coherent noise separation. Synthetic and field data examples demonstrate the effectiveness of the proposed method.

Deciphering hematopoietic stem cell development: key signaling pathways and mechanisms

Most blood cells derive from hematopoietic stem cells (HSCs), originating from endothelial cells. The induction of HSCs from endothelial cells occurs during mid-gestation, and research has revealed multiple steps in this induction process. Hemogenic endothelial cells emerge within the endothelium, transition to hematopoietic cells (pre-HSCs), and subsequently mature into functional HSCs. Reports indicate transcription factors and external signals are involved in these processes. In this review, we discuss the timing and role of these transcription factors and summarize the external signals that have demonstrated efficacy in an in vitro culture. A precise understanding of the signals at each step is expected to advance the development of methods for inducing HSCs from pluripotent stem cells.

Enhancing the Efficiency and Sustainability of Producing Curcumin‐Infused Plant‐Based Milk Alternatives With a Two‐in‐One Post pH‐Driven Processing Strategy

ABSTRACTIncreasing food sustainability and health benefits is essential to meet the demands of a growing global population while minimizing environmental impact. This study used a green two‐in‐one post pH‐driven processing strategy to develop a sustainable and healthy plant‐based milk alternative. It achieved both extraction and encapsulation in one step by directly incorporating the health‐promoting curcumin from turmeric into soymilk. A high processing efficiency was observed, 94.2% ± 1.6%, with a high extraction efficiency of 96.4% ± 0.5%. Using raw turmeric instead of a purified curcumin significantly enhanced the sustainability in the use of raw materials, for example, reducing the CO2‐eq emissions by 22 times and energy use by 10 times, even with a very small percentage of curcumin (∼0.03 wt%) in the formulation. This strategy underscores the importance of using raw materials and minimizing processing steps to develop more sustainable foods. Additionally, the incorporation of curcumin was found to impart a yellow color to soymilk. No significant changes were observed in other physicochemical properties like particle size, zeta potential, and melting behavior, as most curcumin molecules were encapsulated within the lipid phase of soymilk. The curcumin‐infused soymilk powders also maintained excellent storage stability for 1 month under freezing temperature.

Development of Tools for the Automatic Processing of Advanced Driver Assistance System Test Data

Advanced driver assistance system (ADAS) technologies are key to improving road safety and promoting innovation in the automotive sector. The approval and analysis of ADAS systems, especially automatic emergency braking (AEB) tests, require complex procedures and in-depth data management. This work presents innovative tools developed to facilitate the post-processing of ADAS AEB test data, created in collaboration with Nardò Technical Center. The tool, called Autonomous Code Generation Intelligence (ACGI), introduces an intuitive and intelligent user interface that helps analyze and interpret ADAS test approval regulations. ACGI automates the generation of code sections within a data analytics framework, streamlining the compliance process and significantly reducing the time and programming skills required. This tool allows engineers to focus on high-value tasks, improving overall process efficiency. To achieve this objective, the tool encodes the DAART code framework (Data Analysis and Automated Report Tool) which allows users to carry out real post-processing of the tests conducted on the track. The results demonstrate that both tools simplify and automate critical steps in the ADAS automatic emergency braking test data analysis process. In fact, the tool not only improves the accuracy and efficiency of the analyses but also offers a high degree of customization, making it a flexible and adaptable tool to meet the specific needs of users. In future developments, ACGI could be extended to cover additional ADAS tests and could be equipped with artificial intelligence to suggest configurations based on new regulations.

Deep Learning-Based Cloud Detection for Optical Remote Sensing Images: A Survey

In optical remote sensing images, the presence of clouds affects the completeness of the ground observation and further affects the accuracy and efficiency of remote sensing applications. Especially in quantitative analysis, the impact of cloud cover on the reliability of analysis results cannot be ignored. Therefore, high-precision cloud detection is an important step in the preprocessing of optical remote sensing images. In the past decade, with the continuous progress of artificial intelligence, algorithms based on deep learning have become one of the main methods for cloud detection. The rapid development of deep learning technology, especially the introduction of self-attention Transformer models, has greatly improved the accuracy of cloud detection tasks while achieving efficient processing of large-scale remote sensing images. This review provides a comprehensive overview of cloud detection algorithms based on deep learning from the perspective of semantic segmentation, and elaborates on the research progress, advantages, and limitations of different categories in this field. In addition, this paper introduces the publicly available datasets and accuracy evaluation indicators for cloud detection, compares the accuracy of mainstream deep learning models in cloud detection, and briefly summarizes the subsequent processing steps of cloud shadow detection and removal. Finally, this paper analyzes the current challenges faced by existing deep learning-based cloud detection algorithms and the future development direction of the field.

PyPCG: a Python toolbox specialized for phonocardiography analysis.

Phonocardiography has recently gained popularity in low-cost and remote monitoring, including passive fetal heart monitoring. The development of methods which analyse phonocardiographic data tries to capitalize on this opportunity, and in recent years a multitude of such algorithms and models have been published. In these approaches there is little to no standardization and multiple parts of these models have to be reimplemented on a case-by-case basis. Datasets containing heart sound recordings also lack standardization in both data storage and labeling, especially in fetal phonocardiography. We are presenting a toolbox that can serve as a basis for a future standard framework for heart sound analysis. This toolbox contains some of the most widely used processing steps and with these, complex analysis pipelines can be created. These functions can be tested individually. Due to the interdependence of the steps, we validated the current segmentation stage using two phonocardiogram datasets, a fetal dataset comprising 50 one-minute abdominal PCG recordings, which include 6758 S1 and 6729 S2 labels and a filtered version of the dataset used in the 2022 PhysioNet Challenge, containing 413 records with 9795 S1 and 9761 S2 labels. Our results were compared to other common and publicly available segmentation methods, such as peak detection with the Neurokit2 library, and the Hidden Semi-Markov Model by Springer et al. Our best model achieved a 96.1% F1 score and 11.7 ms mean absolute error for fetal S1 detection, and 81.3% F1 score and 50.5 ms mean absolute error for PhysioNet S1 detection. Our detection method outperformed all other tested methods on the fetal dataset and
achieved results comparable to the state of the art on the PhysioNet dataset. Accurate segmentation of signals is critical for the calculation of accurate statistical measures and the creation of classification models. Our toolbox contains functions for both feature extraction and calculation of statistics which are compatible with the previous steps. All of our methods can be fine tuned for specific datasets. pyPCG is available on https://pypcg-toolbox.readthedocs.io/en/latest/.

Supporting Patients' Use of Digital Services in Primary Health Care in England: Synthesis of Evidence From a Mixed Methods Study of "Digital Facilitation".

General medical practitioners and other staff at primary care medical practices have an important role in facilitating patient access to online services in the National Health Service in England. These services range from online ordering of repeat prescriptions to conducting online consultations with health care professionals. We have defined "digital facilitation" as that range of processes, procedures, and personnel that seeks to support patients in their uptake and use of online services. We report how we have synthesized the evidence from a mixed methods study of digital facilitation in primary care in England. The study's objectives were to identify, characterize, and explore the benefits and challenges of different models of digital facilitation in general medical practices in England and to design a framework for evaluation of the effectiveness and costs of digital facilitation interventions. Our study comprised scoping review of literature, survey of staff in general practices, survey of patients, and ethnography at case study practices plus stakeholder interviews. We compiled a triangulation matrix of the findings from individual work packages through an iterative process whereby each work package's results were first analyzed separately and were then cumulatively combined across work packages in 3 successive workshops. From the resulting matrix, we developed a program theory and an implementation theory and constructed a framework for evaluations of digital facilitation in primary care. The final step of the synthesis process was to discuss the results with national and regional National Health Service stakeholders. Triangulation yielded a combined set of findings summarized within 11 thematic groupings: 3 setting the scene within which digital facilitation takes place, and 8 related to different types of digital facilitation, their implementation, and effectiveness. Some thematic groupings were evident in the findings of all 4 of the research work packages; others were not addressed in all the work packages but were evident from those where they were addressed. Throughout the synthesis, there were no instances where findings from one work package contradicted the findings of another. Findings either reinforced each other or offered complementary or additional insights. The discussion at the stakeholder meeting held at the end of the study resulted in the research team clarifying some findings but not changing any of them. Digital facilitation can take many forms, though much of what is currently done in primary care practices in England is reactive and passive. Clear lines of responsibility, digital tools and platforms that work well for patients and practice staff, and investment in staff time and training are all needed if digital facilitation is to deliver on its promise. We propose a framework for future evaluations of the effectiveness and costs of digital facilitation interventions.

HAIPO: Hybrid AI Algorithm-Based Post-Fabrication Optimization for Modern 3D NAND Flash Memory

To successfully meet the various requirements of modern storage systems, NAND flash memory should be highly optimized by precisely tuning a huge number of internal operating parameters. Although 3D NAND flash memory succeeds in increasing the capacity of storage systems, its complex architecture and unique error behavior make such optimization a more difficult and time-consuming process during NAND manufacturing. In this paper, we introduce HAIPO, a novel methodology for post-fabrication optimization of NAND flash memory, which is an essential step in the manufacturing process of modern 3D NAND flash memory to simultaneously meet various requirements on reliability, performance, yield, etc. HAIPO is based on simple machine-learning approaches that consist of (i) a lightweight deep-learning (DL) model to generate initial device parameters and (ii) an evolutionary algorithm (EA) to explore device parameters automatically. To more effectively explore device parameters, we introduce three key guidelines for each generation in the EA: (1) domain-specific rules, (2) recent optimization results, and (3) online Bayesian simulation, respectively, to enable quick optimization for a huge number of device parameters within the limited product turnaround time (TAT). In addition, we integrate two optimization modules with HAIPO to improve optimization efficiency even in environments with severe process variation. We demonstrate the feasibility and effectiveness of HAIPO using real 320 3D TLC/QLC NAND flash chips, showing significant performance and reliability improvements by up to 8.8% and 12% on average, respectively, within a quite limited optimization TAT.

Radiation protection and personal dosimetry in a core facility for multimodal small animal imaging.

Clinical imaging techniques such as positron emission tomography (PET) in combination with computed tomography (CT) are increasingly being used in biomedical research involving small animal models. The handling of open radioactive substances (radiopharmaceuticals) necessary for PET imaging requires prior official authorization for handling, the application of radiation protection principles, and regular training. The overriding aim of radiation protection is to protect the personnel directly involved, other persons, and the environment from the harmful effects of ionizing radiation.This paper aims to provide an overview of the regulatory requirements of the Radiation Protection Act (StrlSchG), the Radiation Protection Ordinance (StrlSchV), and the associated standards and guidelines. Furthermore, their implementation in practical work in small animal imaging using PET/CT is shown. We will focus on the individual steps of the imaging process, from delivery of the radiopharmaceuticals to waste disposal. This should provide interested researchers with an initial overview of the safe and successful use of the method. In addition, exposure values from the last six years in the literature were analyzed. While personal dosimetric monitoring in clinical PET/CT imaging has been extensively published, there is no published data known to us for personnel for PET/CT research with small animals. The evaluation of the personal dosimetric monitoring of our small animal imaging facility with 7 employees over 4 years revealed an increased personal and finger dose normalized to the injected activity and compared to human PET/CT imaging. Nevertheless, the annual personal dose or annual finger dose in small animal imaging (Hp(10): 1.7 mSv, Hp(0.07): 64 mSv) is lower than for personnel performing human PET/CT imaging at the local University Department of Nuclear Medicine (Hp(10): 3.8 mSv, Hp(0.07): 156 mSv) or published values, and is well below the legally permissible maximum dose of 20 or 500 mSv per year.The increasing use of PET/CT in small animal research can be safely utilized if the radiation protection principles are implemented and continuously trained. · PET/CT imaging in small animals is increasingly used in biomedical research.. · Radiation protection laws and guidelines have to be known and are relevant in animal experiments.. · Compared to published values from human medicine, activity-specific employee doses are increased in the presented imaging facility.. · The legal personal dose in the studied imaging facility is below legal limits.. · Schildt A, Sänger P, Lütgens M et al. Radiation protection and personal dosimetry in a core facility for multimodal small animal imaging. Fortschr Röntgenstr 2024; DOI 10.1055/a-2462-2419.

Monitoring Experiment of Melon Greenhouse’s Environment in Tropical Climate

Greenhouses in tropical climates are designed to control passively the environment, protecting plants from pest and extreme climate condition, which is increasingly important due to climate change. This research aims to monitor a melon greenhouse's environment in a tropical climate to understand light intensity, pollutants, and climate conditions. Indoor and outdoor conditions of melon greenhouse were real-time monitored by Vantage VUE model, DAVIS weather station, PM2.5 meter and noise meter. The findings examined that peak light intensities were recorded at 135,600 lux outdoors and 32,050 lux indoors at noon, with an average light transmittance of 38%. Additionally, PM2.5 levels remained stable around 26-30 µg/m³, and sound levels decreased from 60 dB in the morning to 45 dB. These pollution levels did not disturb farmer and indoor melon in winter season. However, other seasoning period needs to be monitored for long term adaptation of application and climate change mitigation. These research findings will support greenhouse design for human comfort and plant growth, considering and optimizing temperature and humidity conditions. IoTs mechanisms and devices were proposed high costly potential for monitoring sensor, networking process, comparative and reliable data collection for further next step of greenhouse integration. Lastly, upcycled transparent roof from LDPE were suggested to be continually used with minor development or plug-in devices for increasing light shade during the mid-daytime.

Ultra-small Metallic Nickel Nanoparticles on Dealuminated Zeolite for Active and Durable Catalytic Dehydrogenation.

Each step in the catalyst synthesis process plays an important role in tuning the catalyst structures. For zeolite-supported nickel catalysts, we found the conventional calcination-reduction method typically leads to the formation of large nickel particles, but a pre-aging in hydrogen or nitrogen at a low temperature prior to final reduction can result in ultra-small nickel nanoparticles in a metallic state. This pre-aging treatment facilitates the interaction between Ni2+ cations and silanol nests on zeolite before the decomposition of the metal salt, leading to the formation of nanoparticles with an average diameter of ~1.2 nm. In contrast, the pre-calcination in oxygen caused the Ni2+ aggregation before the decomposition of the metal salt precursor, yielding nickel nanoparticles larger than 5 nm. Given the structure sensitivity of nickel in cyclohexane dehydrogenation for hydrogen production, the ultra-small nickel nanoparticles exhibited significantly enhanced activity and durability compared to previous nickel catalysts.

Ultra‐small Metallic Nickel Nanoparticles on Dealuminated Zeolite for Active and Durable Catalytic Dehydrogenation

Each step in the catalyst synthesis process plays an important role in tuning the catalyst structures. For zeolite‐supported nickel catalysts, we found the conventional calcination‐reduction method typically leads to the formation of large nickel particles, but a pre‐aging in hydrogen or nitrogen at a low temperature prior to final reduction can result in ultra‐small nickel nanoparticles in a metallic state. This pre‐aging treatment facilitates the interaction between Ni2+ cations and silanol nests on zeolite before the decomposition of the metal salt, leading to the formation of nanoparticles with an average diameter of ~1.2 nm. In contrast, the pre‐calcination in oxygen caused the Ni2+ aggregation before the decomposition of the metal salt precursor, yielding nickel nanoparticles larger than 5 nm. Given the structure sensitivity of nickel in cyclohexane dehydrogenation for hydrogen production, the ultra‐small nickel nanoparticles exhibited significantly enhanced activity and durability compared to previous nickel catalysts.

3D Nanofabrication via Directed Material Assembly: Mechanism, Method, and Future.

Fabrication of complex three-dimensional (3D) structures at nanoscale is the core of nanotechnology, as it enables the creation of various micro-/nano-devices such as micro-robots, metamaterials, sensors, photonic devices, etc. Among most 3D nanofabrication strategies, the guided material assembly, an efficient bottom-up approach capable of directly constructing designed structures from precise integration of material building blocks, possesses compelling advantages in diverse material compatibility, sufficient driving forces, facile processing steps, and nanoscale resolution. In this review, we focus on assembly-based fabrication methods capable of creating complex 3D nanostructures (instead of periodic or 2.5D-only structures). Recent advances are classified based on the different assembly mechanisms, i.e., assembly driven by chemical reactions, physical interactions, and the synergy of multiple microscopic interactions. The design principles of representative fabrication strategies with an emphasis on their respective advantages, e.g., structural design flexibility, material compatibility, resolution, or applications are analyzed. In the summary and outlook, existing challenges, as well as possible paths to solutions for future development are reviewed. We believe that with recent advances in assembly-based nanofabrication strategies, 3D nanofabrication has achieved tremendous progress in resolution, material generality, and manufacturing cost, for it to make a greater impact in the near future.

Staged evolutionary features of the aromatic structure in high volatile A bituminous coal (hvAb) during gold tube pyrolysis experiments

Low-temperature pyrolysis of coal is a crucial step in the coal thermal conversion process and involves very complex physical and chemical reactions that can have different effects on the coal's structure. The thermal evolution behavior and transformation mechanism of the coal microstructure are not yet fully understood, which also limits the efficient utilization of coal to a certain extent. The aromatic structural features (including size, molecular ordering, nematic symmetry, stacking, and curvature) of the char produced from low-temperature pyrolysis of high volatile A bituminous coal (hvAb) from the Xutuan coal mine, China, were quantitatively assessed via high-resolution transmission electron microscopy (HRTEM) image processing and analytical techniques. The thermal transformation process and the mechanisms controlling it were explored. The results show that, except for the unexpected slight growth of aromatic sheets at 440 °C, the lower pyrolysis temperature (< 521 °C) contributed weakly to their size growth, whereas at higher temperatures (561–600 °C), it significantly increased their size. The aromatic molecular ordering tended to gradually change in three stages: increasing between 340 and 440 °C, decreasing between 440 and 521 °C and increasing again between 521 and 600 °C. The nematic symmetry strength of aromatic fringes also followed a similar pattern with temperature at different scales. Additionally, in addition to a very minor development trend at 440 °C, the stacking did not significantly change at temperatures below 521 °C but developed appreciably further with increasing temperatures at 561–600 °C; however, the average spacing of the stacks did not appear to be significantly reduced at all temperatures. The curvature of the aromatic sheets also varied in different temperature stages, i.e., initially slightly increasing (340–380 °C), then gradually decreasing (380–480 °C), later increasing again (480–521 °C), and eventually decreasing (521–600 °C). The properties of the chemical composition and structure of the initial coal play important roles in the thermal reaction behavior, and the physical and chemical reactions that dominate at the different temperature stages may be responsible for such wiggly trends in the evolution of the aromatic structure. Notably, the properties of the mesophase (approximately 440 °C) strongly influence the subsequent structural transformation. These findings could provide useful information for the microstructure–property relationships and preparation of coal-based carbon materials.

The mechanism of acetylation-mediated fusion of lysosomes with autophagosomes in neurons after ischemic stroke.

Ischemic stroke is a serious cerebrovascular disease that brings a significant threat to human health [5]. Considerable factors are involved in occurrence of cerebral ischemia. Among them, autophagy is an important intracellular process that is activated after ischemic stroke, which plays a crucial role in maintaining homeostasis and survival of neurons [10]. The fusion of lysosomes with autophagosomes is a key step in autophagic processes [1]. In recent decades, investigations have found that acetylation, a common post-translational modification of proteins, has an important regulatory effect on autophagy [7]. The present article focuses on elucidating mechanism and roles of acetylation in fusion of lysosomes with autophagosomes in neurons after ischemic stroke, to seek novel targets and strategies for deeper understanding of the pathogenesis of ischemic stroke [8]. This review is also to provide clues for clinical treatment of ischemic stroke [9].

Mesomycoplasma (Mycoplasma) ovipneumoniae dihydrolipoamide dehydrogenase is an immunogenic plasminogen binding protein and a putative adhesin

The interaction of Mesomycoplasma (Mycoplasma) ovipneumoniae (M. ovipneumoniae) with host cells is a pivotal step in the infection process, underlining the necessity to develop vaccines and therapeutic approaches targeting the pathogen's key invasion mechanisms. The bacterium's capacity for adherence, invasion, and subsequent evasion of the host immune response underpins its pathogenicity, rendering adherence genes feasible vaccine targets. This study focuses on pyruvate dehydrogenase complex component E3 (PdhD), a membrane-anchored surface protein implicated in these pathogenic processes. Bioinformatics analysis reveals the conservation of PdhD sequence within M. ovipneumoniae. Membrane protein extraction, immunoblotting and ELISA assay have confirmed the presence of PdhD on the M. ovipneumoniae surface and cytoplasm, suggesting its multifunctionality. Our research employed antibody inhibition assays to characterize the bacterial adhesion suppression by anti-PdhD antibodies, complemented by bactericidal complement assays, supporting its candidacy as a putative vaccine target. The ELISA binding assay substantiated that PdhD binded to plasminogen (Plg) in a dose-dependent manner. Notably, PdhD is also involved in biofilm formation. The inhibitory effect of anti-PdhD sera on biofilm formation is congruent with novel therapeutic strategies targeting related mycoplasmas. This study reports the characterization of the first virulence-associated protein PdhD of M. ovipneumoniae and suggests its potential as a vaccine target to combat M. ovipneumoniae infection.

Analysis of Industry 4.0 and the Building Inspection System in the Republic of Türkiye's Applications

Since the enactment of Law No. 4708 on Building Inspection, developments have taken place in the construction sector in Türkiye in terms of material production, their usage in construction sites, construction stages, project management, and implementation, which have led to the need for changes in the inspection field. In this study, the Building Inspection process was classified under seven headings within the scope of the Regulation. The fundamental problems in these headings were analyzed with research methods in the literature. Based on the information, which digital transformation technologies are included in the current practices and which ones are missing have been identified. Collaborations with which technology can solve existing problems have been interpreted, and solutions have been proposed to complete the shortcomings that may be experienced in the inspection field. It has been determined that Industry 4.0 Technologies should be adapted to every step of the inspection process for a healthy inspection system.

Scar-free tag removal by CASPON® enzyme with broad physicochemical stability in biomanufacturing – A case study of five proteins

Manufacturing of recombinant proteins in microbial systems, in particular in E. coli, generally requires thorough process development due to the absence of a platform process. The caspase-based fusion process (CASPON®) offers a platform manufacturing process using special protease cleavable fusion-tags. These tags allow the implementation of His-tag based affinity chromatography for facile target protein capture and offer expression and solubility enhancing capabilities. The tags are intended as N-terminal fusion motifs that can be fully cleaved using a modified caspase-2 protease, the CASPON® enzyme. Here, we systematically explore the influence of various physicochemical parameters on its enzymatic activity. This characterization was performed in parallel in two independent research laboratories using different assays, i.e. a Förster resonance energy transfer-based assay using small peptide substrates and a reversed phase high performance liquid chromatography method using model proteins. Both assays demonstrated great agreement and reveal that CASPON® enzyme is highly active at a wide range of temperatures, pH and is resistant to a variety of chemical substances that are commonly employed in bioprocessing, e.g. NaCl, kosmotropes, chaotropes and others. The presence of imidazole at concentrations commonly found in the elution fraction of immobilized metal affinity chromatography does not seem to affect the activity of CASPON® enzyme, enabling its use directly after the capture step in downstream processing (DSP). A case study of five biopharmaceuticals is presented. The platform process exhibits consistently high performance in up- and downstream processing, achieving high soluble titers, high yield and purity. The presence of host cell proteins during the DSP in particular was investigated in-depth using process proteomics, revealing that a core of host cell proteins is process dependent and can commonly be expected to be found in the platform DSP. The information presented here can serve as a guide on how to implement the CASPON® platform process for the production of various recombinant proteins.

Assessing cacao beans fermentation degree with improved YOLOv8 instance segmentation

Fermentation is a critical step in cacao bean processing, significantly influencing the quality and flavor of the final chocolate product. Meeting international market standards requires well-fermented cacao beans, emphasizing the importance of precisely determining their fermentation degree before export. Traditionally, human graders rely on a cut test to assess the degree of fermentation, but this method is prone to individual variability and is time consuming. To address this problem, the YOLO-CoLa model has been developed within the YOLOv8 framework, focusing on accurately detecting the fermentation degree of cacao beans. This new model, an extension of the YOLOv8s model, incorporates the innovative Large Selective Kernel Block (LSKBlock) within the network’s backbone, replacing the C2f module to improve detection accuracy. Data augmentation was applied to mitigate limitations related to the availability of training images. The results showed the effectiveness of YOLOv8-CoLa, achieving a mAP0.5 of 70.4 %, a notable improvement of 9.3 % over YOLOv8. These findings highlight the significance of integrating LSKBlock and the value of tailored model adaptations in accurately identifying fermentation degrees in cacao beans. The advancements presented in this research offer a viable solution to the challenges faced in determining cacao bean fermentation levels, contributing to the optimization of cacao bean processing.

An introduction to CIABATTA: Increasing accessibility to corpus creation for non-experts

This methodological report introduces CIABATTA (Corpus In A Box: Automated Tools, Tutorials, & Advising), a suite of tools that provides novice corpus compilers with a starting point to develop new corpora. These tools are a distillation of guiding principles and templates for corpus building developed by the Corpus & Repository of Writing (CROW) team during the compilation of the CROW corpus. We start by discussing the best practices and ethical issues that need to be considered when compiling a corpus. We then focus on the steps of corpus data processing, introducing two tools that support corpus building. The first tool, the Corpus Text Processor, is used to convert, encode, and standardize texts in the corpus. The second, the deidentifying tool, replaces identifying information with special annotation tags in corpus files, such as [name] and [place]. Both can be downloaded from the CIABATTA wiki and do not require programming experience. In the final part of this paper, we focus on customizable Python scripts developed for CIABATTA that aid in the processing of metadata. We believe that CIABATTA can aid those without corpus building expertise to build corpora for research purposes, thus expanding the use of corpora to a wider group of researchers.