Recursive Search Research Articles

Methods Used in Co-Creation Within the Health CASCADE Co-Creation Database and Gray Literature: Systematic Methods Overview

Background Co-creation is increasingly recognized for its potential to generate innovative solutions, particularly in addressing complex and wicked problems in public health. Despite this growing recognition, there are no standards or recommendations for method use in co-creation, leading to confusion and inconsistency. While some studies have examined specific methods, a comprehensive overview is lacking, limiting the collective understanding and ability to make informed decisions about the most appropriate methods for different contexts and research objectives. Objective This study aimed to systematically compile and analyze methods used in co-creation to enhance transparency and deepen understanding of how co-creation is practiced. Methods To enhance transparency and deepen understanding of how co-creation is practiced, this study systematically inventoried and analyzed methods used in co-creation. We conducted a systematic methods overview, applying 2 parallel processes: one within the peer-reviewed Health CASCADE Co-Creation Database and another within gray literature. An artificial intelligence–assisted recursive search strategy, coupled with a 2-step screening process, ensured that we captured relevant methods. We then extracted method names and conducted textual, comparative, and bibliometric analyses to assess the content, relationship between methods, fields of research, and the methodological underpinnings of the included sources. Results We examined a total of 2627 academic papers and gray literature sources, with the literature primarily drawn from health sciences, medical research, and health services research. The dominant methodologies identified were co-creation, co-design, coproduction, participatory research methodologies, and public and patient involvement. From these sources, we extracted and analyzed 956 co-creation methods, noting that only 10% (n=97) of the methods overlap between academic and gray literature. Notably, 91.3% (230/252) of the methods in academic literature co-occurred, often involving combinations of multiple qualitative methods. The most frequently used methods in academic literature included surveys, focus groups, photo voice, and group discussion, whereas gray literature highlighted methods such as world café, focus groups, role-playing, and persona. Conclusions This study presents the first systematic overview of co-creation methods, providing a clear understanding of the diverse methods currently in use. Our findings reveal a significant methodological gap between researchers and practitioners, offering insights into the relative prevalence and combinations of methods. By shedding light on these methods, this study helps bridge the gap and supports researchers in making informed decisions about which methods to apply in their work. Additionally, it offers a foundation for further investigation into method use in co-creation. This systematic investigation is a valuable resource for anyone engaging in co-creation or similar participatory methodologies, helping to navigate the diverse landscape of methods.

Bridging the gap: mapping co-creation methods for public health impact and innovation

Abstract Background Co-creation offers a promising approach to public health innovation, particularly for tackling complex issues. However, the lack of a standardized methodology limits its reliability and reproducibility, especially across diverse contexts. This study explores co-creation methods used by researchers and practitioners, aiming to shed light on their implications for public health. Understanding this methodological landscape can enhance the effective implementation of co-creation processes to address complex challenges. Methods Utilizing the Systematic Methods Overview approach, we examined co-creation methods. Data were gathered from the Health CASCADE Co-Creation Database and grey literature using an artificial intelligence-assisted recursive search strategy and a two-step screening process. We conducted textual, comparative, and bibliometric analyses to evaluate methodological underpinnings and relationships between methods. Results Analysis of 2,627 academic articles and grey literature sources identified over 900 co-creation methods. While surveys, focus groups, and group discussions were prevalent, participatory methods were underutilized. Notably, 91.3% of methods co-occurred, with a predominant combination of qualitative approaches. Furthermore, a significant discrepancy between research and practice was evident, with only 10.2% convergence. Conclusions This study revealed a critical gap between co-creation in research and practice, emphasizing the need to bridge this methodological divide for advancing co-creation as a reliable approach in public health. The over-reliance on qualitative methods raises concerns about tokenistic engagement and undermines meaningful public participation in co-creation. Addressing this trend is vital to preserve the essence of true co-creation and maximize its potential for addressing complex public health challenges. Key messages • Mapping the diverse methods of co-creation in public health reveals both richness and disparities in approach, from surveys to participatory methods. • Addressing the discrepancy between co-creation in research and practice is vital for establishing a reliable and effective approach to tackling public health challenges.

Dynamic Analysis Techniques for Road Bridges Under Large Traffic Flows

This work focuses on the dynamic analysis of road bridges under large microsimulated traffic flows, contributing an efficient dynamic analysis framework for future fatigue, comfort or driving safety studies in which realistic interactions between vehicles are considered. Different strategies to define the loading vector in the governing dynamic equations are presented for bridges discretized with beam and shell elements. The proposed interpolation techniques for beam-like models include a recursive binary search and a novel vectorized strategy based on shape functions expanded to the entire length of the deck. These are applied to a short simply supported bridge and a very long cable-stayed bridge under traffic flows generated with a cellular automata microsimulation model. The strategies presented here reduce the computational time required in conventional load interpolation algorithms, particularly in long bridges subject to dense traffic flows. Finally, the binary search algorithm is applied to the study of bridge decks discretized with shell elements under microsimulated traffic, demonstrating that high vehicle densities increase the contribution of the transverse vibration of the slab and the potential discomfort of pedestrians.

Polymorphous Low-Grade Neuroepithelial Tumor of the Young (PLNTY): Scoping Review of Case Reports and Case Series.

Polymorphous low-grade neuroepithelial tumor of the young (PLNTY) is considered one of the low-grade neuroepithelial tumors, as per the World Health Organization 2021 classification of brain tumors. First described in 2016, these morphologically variable tumors are characterized by oligodendroglioma-like cellular components, infiltrative growth patterns, and cluster of differentiation 34 immunopositivity. A literature search of the PubMed/MEDLINE, SCOPUS, ScienceDirect, and COCHRANE databases (from inception to 20th June 2022) was carried out to identify relevant studies. To identify additional studies, we performed a recursive search of the bibliographies of the selected articles and published systematic reviews on this topic. The search yielded a total of 64 results. After removing duplicates, 26 articles were eligible for the review. The diagnostic criteria for these glioneuronal variants, representing a broad neuropathological spectrum, are not distinct and hence impede proper diagnosis and prognosis. Frequent genetic abnormalities involving mitogen-activated protein kinase pathway constituents, such as B-Raf proto-oncogene or fibroblast growth receptor 2/3, are harbored by PLNTYs. Recent advances in molecular diagnostics have resulted in more accurate tumor classification systems, based on gene expression profiles and DNA methylation patterns. Gross total resection seems curative, with a low recurrence rate. Malignant transformation is rare; however, adjuvant radiation therapy and chemotherapy may be beneficial in selected cases.

Computing a compact local Smith–McMillan form

We define a compact local Smith–McMillan form of a rational matrix R ( λ ) as the diagonal matrix whose diagonal elements are the nonzero entries of a local Smith-McMillan form of R ( λ ) . We show that a recursive rank search procedure, applied to a block-Toeplitz matrix built on the Laurent expansion of R ( λ ) around an arbitrary complex point λ 0 , allows us to compute a compact local Smith-McMillan form of that rational matrix R ( λ ) at the point λ 0 , provided we keep track of the transformation matrices used in the rank search. It also allows us to recover the root polynomials of a polynomial matrix and root vectors of a rational matrix, at an expansion point λ 0 . Numerical tests illustrate the promising performance of the resulting algorithm.

A recursive framework for improving the performance of multi-objective differential evolution algorithms for gene selection

Gene selection is a pivotal process in machine-learning-driven medical diagnostics, where the goal is to identify a subset of genes from microarray expression profiles that can enhance the predictive accuracy of classifiers for disease diagnosis. The two key objectives of gene selection are to reduce the dimensionality of the data and to improve the accuracy of disease diagnosis, which is typically a multi-objective optimization problem. In recent years, multi-objective evolutionary algorithms (MOEAs) have gained wide attention in feature selection research, and several related algorithms have been produced. However, most algorithms tend to get stuck in local optimality when searching for solutions from a high-dimensional space. To solve the gene selection problem effectively, this study introduces a recursive multi-objective differential evolution algorithm with elite recursive strategy (RMODE-E) and a recursive multi-objective differential evolution algorithm with Pareto front recursive strategy (RMODE-P). RMODE-E amalgamates the features selected by the top E elite individuals, RMODE-P consolidates the features selected by the Pareto front set, and the combined features then serve as the foundation for subsequent recursive rounds of searching. The proposed feature subspace combination strategy not only reduces the recursive search space but also improves the capacity to find globally optimal feature subsets. Extensive experiments were conducted to compare our proposed algorithms with eight state-of-the-art evolutionary algorithms to validate their effectiveness. Experimental results demonstrate that RMODE-P has better global search capability as it achieves better best classification accuracy, mean classification accuracy, and minimal gene subset size.

A Novel Approach to Sensor Placement: Recursive Exhaustive Search and Stochastic Optimization Parameter Impact Analysis

This study presents a comprehensive approach for single sensor placement optimization in two-dimensional and three-dimensional spaces. A traditional exhaustive search technique and a novel method called recursive exhaustive search are used to place a sensor in a way that maximizes the area coverage metric. Exhaustive search provides a baseline by methodically evaluating all potential placements, while recursive exhaustive search innovates by segmenting the search process into more manageable, recursive steps. Our findings highlight the significant impact of two key parameters, the number of evaluations and the rasterization value, on the achieved coverage and computation time. The results show that the right choice of parameters can significantly reduce the computational effort without compromising the quality of the solution. This underlines the critical need for a balanced approach that considers both computational complexity and placement efficacy. We show that exhaustive search is not feasible for three-dimensional environment models and propose to establish a modified exhaustive search method as a ground truth for the single sensor placement problem. We then explore nature-inspired genetic algorithms and the impact of the number of evaluations of the optimization function for these algorithms on both accuracy and computational cost.

Lattice layout and optimizer effect analysis for generating optimal transcranial electrical stimulation (tES) montages through the metaheuristic L1L1 method.

This study focuses on broadening the applicability of the metaheuristic L1-norm fitted and penalized (L1L1) optimization method in finding a current pattern for multichannel transcranial electrical stimulation (tES). The metaheuristic L1L1 optimization framework defines the tES montage via linear programming by maximizing or minimizing an objective function with respect to a pair of hyperparameters. In this study, we explore the computational performance and reliability of different optimization packages, algorithms, and search methods in combination with the L1L1 method. The solvers from Matlab R2020b, MOSEK 9.0, Gurobi Optimizer, CVX's SeDuMi 1.3.5, and SDPT3 4.0 were employed to produce feasible results through different linear programming techniques, including Interior-Point (IP), Primal-Simplex (PS), and Dual-Simplex (DS) methods. To solve the metaheuristic optimization task of L1L1, we implement an exhaustive and recursive search along with a well-known heuristic direct search as a reference algorithm. Based on our results, and the given optimization task, Gurobi's IP was, overall, the preferable choice among Interior-Point while MOSEK's PS and DS packages were in the case of Simplex methods. These methods provided substantial computational time efficiency for solving the L1L1 method regardless of the applied search method. While the best-performing solvers show that the L1L1 method is suitable for maximizing either focality and intensity, a few of these solvers could not find a bipolar configuration. Part of the discrepancies between these methods can be explained by a different sensitivity with respect to parameter variation or the resolution of the lattice provided.

Safety and Efficacy of Nemonoxacin vs Levofloxacin in Patients With Community-Acquired Pneumonia: A Systematic Review and Meta-Analysis of Randomized Control Trials.

Community-acquired pneumonia is a leading cause of morbidity and mortality throughout the world, which incurs significant healthcare costs. The aim of his meta-analysis is to assess the clinical efficacy and safety of a novel non-fluorinated quinolone, nemonoxacin, compared with levofloxacin in treating community-acquired pneumonia (CAP). A recursive literature search was conducted using PubMed, Google Scholar, andScopusup to August 2022. All randomized clinical trials comparing nemonoxacin to levofloxacin for community-acquired pneumonia were included.The patients selected for this study had mild to moderate CAP. Each individual received treatment with either nemonoxacin (500 mg or 750 mg) or levofloxacin (500 mg) for a duration of 3-10 days. Four randomized control trials with a total of 1955 patients were included. Nemonoxacin and levofloxacin were found to have similar clinical cure rates in the treatment of CAP. There were no significant differences reported in the treatment-emergent adverse events between the two drugs (RR=0.95, 95% CI: 0.86, 1.08, I2=0%). However, the most frequent symptoms exhibited were gastrointestinal system-related. Both the dosages (500 mg and 750 mg) of nemonoxacin were found to have similar efficacy as that of levofloxacin.Our meta-analysis indicates that nemonoxacin is a well-tolerated and effective antibiotic therapy for the treatment of community-acquired pneumonia (CAP), with clinical success rates comparable to those of levofloxacin. Furthermore, the adverse effects associated with nemonoxacin are generally mild. Therefore, both the 500 mg and 750 mg dosages of nemonoxacin can be recommended as appropriate antibiotic therapy regimens for the treatment of CAP.

Energy-Efficient Resource Allocation in Multiple UAVs-Assisted Energy Harvesting-Powered Two-Hop Cognitive Radio Network

The unmanned aerial vehicle (UAV)-assisted energy harvesting (EH)-based cognitive wireless networks can solve the problems of spectrum scarcity, energy shortage, and challenging layout of communication facilities faced by traditional wireless networks, envisioned as a vital frontier communication technology in the 5G/6G era. In this article, we consider a UAVs-assisted two-hop cognitive radio network (CRN) with EH (UAVs-EH-DHCRN), where multiple UAVs are employed to act as air access points (AAPs) and relay nodes (RNs). Since all EH-powered secondary users (SUs) can simultaneously associate with AAPs by sharing the licensed spectrum of primary users (PUs), intercell interference (ICI) occurs among AAPs. Our goal is to maximize the end-to-end energy efficiency (EE) of the UAVs-EH-DHCRN while considering constraints like ICI, energy, minimum data rate, and interference to PUs. However, the formulated end-to-end EE maximization problem is a typical nonconvex optimization problem with complex constraints that off-the-shelf optimization strategies cannot solve. Therefore, we propose a resource allocation algorithm called approximate convex policy concerning associations, power allocation, and placements (ACP-APP). Specifically, in ACP-APP, we introduce additional parameters to convert the fractional objective function into a polynomial function. Then, the optimal AAPs’ placements, optimal associations between SUs and AAPs, and power allocation are obtained alternately by leveraging the proposed circular-based recursive random search (CRRS), 1-D linear programming, and Frank–Wolfe (FW) policy. Simulation results show that our proposed strategy can converge quickly to obtain the approximate optimal solution. Moreover, the ACP-APP can obtain significant end-to-end EE gains compared with benchmark schemes.

A new implementation of the beam migration by using a model-driven variable grid strategy

Beam migration offers higher imaging accuracy, better flexibility, stability and adaptability than conventional ray migration, while the variable grid method achieves lower storage and higher efficiency under the same accuracy by means of different grid generation of the velocity model. However, the efficiency and accuracy of Gaussian beam migration are strongly affected by the algorithm used for Gaussian beam summation whose key step is to find the underground point at which the local beam contributes relevantly to the reconstruction of the Green's function. Conventionally, one uses Hale's recursive search method that firstly searches on a global coarse grid and then interpolated to a local fine one. Practical experience shows that this method can greatly improve the efficiency of the beam scanning, but causes some wavy discontinuities and imaging artifacts as well. To solve this problem, we present a new method called “model-driven variable grid strategy”. In this strategy, a key step is to establish a relationship between the vertical gradient of the velocity model and the local grid constants through local velocity gradient with respect to the global average velocity. Numerical tests on synthetic and field data show that the new method has a minimal need on computer resources and can largely improve the accuracy and efficiency of the beam migration in regions with complex geological structures.

A novel hybrid gene selection for tumor identification by combining multifilter integration and a recursive flower pollination search algorithm

Gene selection is crucial to tumor identification based on microarray expression data. The identification of genes with strong discriminative power has been a hot research topic and a great challenge in research related to machine learning-based tumor identification. This study proposes a novel hybrid gene selection strategy for tumor identification by combining multifilter integration and a recursive flower pollination algorithm (MFI-RFPA) to select a minimum number of genes to achieve optimal classification performance. MFI-RFPA selects genes that involve two stages. In the first stage, most irrelevant and noise genes are eliminated by using a combination of three filters based on different metrics. In this process, multiple filters based on different metrics are combined in the voting, which makes the filter process of MFI-RFPA more robust. In the second stage, MFI-RFPA searches for the optimal subset of the remaining genes by using a recursive flower pollination algorithm (RFPA). In this process, the flower pollination algorithm is performed recursively without reducing the classification accuracy, which ensures that a small and compact subset of genes with excellent classification performance is searched. In the experimental part, MFI-RFPA is compared with both its directly related methods (such as MFI, FPA, RFPA, MFI-FPA) and other state-of-the-art methods. The excellent performance of MFI-RFPA in reducing the number of genes and improving the classification accuracy can be reflected in the comparison of experiments.

Maximum weight perfect matching problem with additional disjunctive conflict constraints

AbstractWe focus on an extension of the maximum weight perfect matching problem with additional disjunctive conflict constraints in conjunction with the degree and binary restrictions. Given a simple graph with a nonnegative weight associated with each edge and a set of conflicting edges, the perfect matching problem with conflict constraints consists of finding a maximum weight perfect matching without any conflicting edge pair. Unlike the well‐known ordinary maximum weight perfect matching problem this one is strongly ‐hard. We propose two branch‐and‐bound algorithms for the exact solution of the problem. The first one is based on an equivalent maximum weight stable set formulation with an additional cardinality restriction obtained on the graph representing conflict relations and uses the information coming from its maximal stable sets. The second one is essentially a recursive depth first search scheme that benefits from simple upper bounds incorporated with a fast infeasibility detection procedure to prune the branch‐and‐bound tree. According to the extensive computational tests it is possible to say that they are both very efficient.

A recursive local search method of separators for Bayesian network decomposition structure learning algorithm

A recursive local search method of separators for Bayesian network decomposition structure learning algorithm

Heuristic algorithms for the special knapsack packing problem with defects arising in aircraft arrangement

This paper addresses a special knapsack packing problem in which identical pieces are packed into an irregular board with unusable regions to maximize the number of pieces. This problem arises from a real-world application in which the aircraft is placed on the carrier’s deck to maximize the number of aircraft. We propose the support polygon concept to solve one of the challenges of the problem, i.e., ensure that the support points inside the piece’s interior are located within the board when the piece protrudes from the board. Many strategies for determining the piece’s location and selecting the rotation angle of the piece are proposed and compared. Based on these techniques, we propose a deterministic heuristic algorithm to solve the problem. In addition, we develop an improved version of the algorithm, called the recursive search heuristic algorithm, to get higher-quality solutions. Many experiments were conducted to test the performance of different combinations of strategies and the proposed algorithms. The computational results show that the proposed algorithms have a strong capacity for obtaining good solutions, and updating many of the best results in the benchmark instances provided by the literature.

A group-based structure for perfect sequence covering arrays

An (n, k)-perfect sequence covering array with multiplicity $$\lambda $$ , denoted $$\mathrm{{PSCA}}(n,k,\lambda )$$ , is a multiset whose elements are permutations of the sequence $$(1,2, \dots , n)$$ and which collectively contain each ordered length k subsequence exactly $$\lambda $$ times. The primary objective is to determine for each pair (n, k) the smallest value of $$\lambda $$ , denoted g(n, k), for which a $$\mathrm{{PSCA}}(n,k,\lambda )$$ exists; and more generally, the complete set of values $$\lambda $$ for which a $$\mathrm{{PSCA}}(n,k,\lambda )$$ exists. Yuster recently determined the first known value of g(n, k) greater than 1, namely $$g(5,3)=2$$ , and suggested that finding other such values would be challenging. We show that $$g(6,3)=g(7,3)=2$$ , using a recursive search method inspired by an old algorithm due to Mathon. We then impose a group-based structure on a perfect sequence covering array by restricting it to be a union of distinct cosets of a prescribed nontrivial subgroup of the symmetric group $$S_n$$ . This allows us to determine the new results that $$g(7,4)=2$$ and $$g(7,5) \in \{2,3,4\}$$ and $$g(8,3) \in \{2,3\}$$ and $$g(9,3) \in \{2,3,4\}$$ . We also show that, for each $$(n,k) \in \{ (5,3), (6,3), (7,3), (7,4) \}$$ , there exists a $$\mathrm{{PSCA}}(n,k,\lambda )$$ if and only if $$\lambda \ge 2$$ ; and that there exists a $$\mathrm{{PSCA}}(8,3,\lambda )$$ if and only if $$\lambda \ge g(8,3)$$ .

A framework based on heterogeneous ensemble models for liquid steel temperature prediction in LF refining process

The precise control of liquid steel temperature in the ladle furnace (LF) refining process is vital for stabilizing and improving the quality of liquid steel, necessitating a capable prediction system. To achieve better predictive performance, an effective prediction framework based on heterogeneous ensemble models is proposed in this paper, which mainly consists of three parts: (1) utilizing single models and tree-based ensemble models to constitute the heterogeneous ensemble model; (2) proposing Recursive Feature Increase (RFI) to facilitate the construction of the heterogeneous ensemble model, including Stacking and Majority Voting; (3) proposing a new optimization algorithm, namely Recursive Search Optimization (RSO), to optimize the hyper-parameters of the heterogeneous ensemble model. Through the verification of the collected industrial production data, it is found that the proposed framework in this paper possesses higher fitting and generalization ability, which is of great significance for engineering applications such as liquid steel temperature prediction in the LF refining process.

Implementation of Back Tracking Algorithm in The Scheduling of Mathematics Study Program Faculty of MIPA Unsoed

The complicated case of scheduling courses at the Mathematics and Natural Sciences Faculty of Mathematics Study Program, Universitas Jenderal Soedirman, is a topic that is quite interesting to discuss and find a solution to using a mathematical method. In fact, manual course scheduling without a method is prone to scheduling errors such as class schedule clashes, clashes in the use of lecture rooms and so on, so a more efficient method of scheduling courses is needed. Scheduling lectures with the backtracking algorithm is a systematic and efficient method of scheduling lectures with influencing factors such as the number of courses, number of sessions, number of rooms, and lecture time. Algorithm backtracking is an algorithm based on Depth First Search to find solutions to problems more efficiently. The back tracking algorithm performs a systematic search for solutions to all possible solutions at each node based on recursive Depth First Search. Depth First Search is a search method that is carried out at one node in each level from the far left. If a solution has not been found, then the search continues on the right hand node. And so on until a solution is found or if you find a dead end it will backtrack to the previous position. If a solution is found, the search will stop even if there are nodes that have not been traced. The implementation of the backtracking algorithm pays close attention to the factors that become obstacles in scheduling courses. The course schedule function with the backtracking algorithm can meet every influencing factor such as the number of courses, rooms, classes, and lecture time so that scheduling lectures with this method is very helpful because the method used is more efficient and can avoid errors in scheduling.

Matrix metallopeptidase 9 and outcome prediction in patients with acute coronary syndrome

The aim of this review was to analyse the scientific literature data on matrix metallopeptidase 9 and to analyse the available information on its prognostic value as a marker of negative outcome in the short- and long-term prognosis in patients with acute coronary syndrome. Materials and methods. In our study was used a recursive literature search strategy at PubMed. The following criteria for inclusion in the analysis were defined – a prospective study in patients with acute coronary syndrome that had data on the effect of MMP-9 levels on short-term and/or long-term outcomes, including mortality and major adverse сardiovascular events. Review articles, clinical cases, animal studies, and studies with insignificant statistical data were not included in the analysis. The depth of the initial search was set at 15 years with a search for similar articles in citations. We selected 5 studies for meta-analysis. Meta-Essentials 1.5 was used for the analysis. Odds ratio and 95 % confidence interval were calculated using the Haenszel method. The association between MMP-9 levels and short-term and long-term outcomes (mortality and major adverse CV events) was determined. The statistically significant level was defined as P < 0.05. Results. The analysis showed no significant association between the level of MMP-9 and the outcome (OR = 1.39; 95 % CI = 0.25–7.79; P = 0.595; I2 = 78 %). Conclusions. Matrix metallopeptidase 9 is a promising marker for further investigation of its predictive strength of outcome. Despite the opposite results of single studies and no significant association of MMP-9 with outcome further research on this issue is a promising direction.

HG-FCN: Hierarchical Grid Fully Convolutional Network for Fast VVC Intra Coding

As one of the key technologies of Versatile Video Coding (VVC), a flexible quad-tree with a nested multi-type tree (QTMT) partition structure significantly improves the rate-distortion (RD) performance. However, this structure brings additional complexity due to the recursive search for the best partition type. Traditional fast partition methods in previous encoders, cannot adapt to this new complex structure, because it’s too complicated to predict each block size from one layer to another layer. Some indirect bottom-up designed methods are simple enough, but cannot predict specific split structures, making the acceleration capacity limited. Therefore, in this paper, we propose a learning-based approach to effectively predict the QTMT structure without having to heuristically explore the partitions of each layer. Firstly, we propose a hierarchy grid fully convolutional network (HG-FCN) framework, which concisely requires inference only once to obtain the entire partition information of the current CU and sub-CUs, and the inference is highly parallel. Secondly, we design a representation of complicated QTMT of CU partition in the form of hierarchy grid map (HGM), which can directly and effectively predict the specific hierarchical split structure. Lastly, a dual-threshold decision scheme is adopted to automatically control the trade-off between coding performance and complexity. Extensive experiments demonstrate the effectiveness of HG-FCN, which can reduce 51.15% <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim ~65.53$ </tex-math></inline-formula> % complexity of VVC intra coding with negligible 1.17% <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim ~2.19$ </tex-math></inline-formula> % BD-BR increase, superior to other state-of-the-art methods.