Hasil untuk "Computer Science"

A. M. Mutawa, Khalid Sabti, Bibin Shalini Sundaram Thankaleela et al.

Diabetic macular edema (DME) is a primary contributor to visual impairment in diabetic patients, necessitating precise and prompt analysis for optimal treatment. Recent breakthroughs in deep learning (DL) and machine learning (ML) have yielded promising outcomes in ophthalmic image analysis. However, researchers often overlook the significance of optimization algorithms in enhancing the efficacy of their models for DME-related tasks. This review aims to consolidate, seek, discover, assess, and integrate existing work on the application of DL and ML, with emphasis on the integration and impact of optimization algorithms in enhancing their efficacy, robustness, and performance for DME in the fields of computer science and engineering. The population, intervention, comparison, and outcome framework was employed in this study to facilitate a clear and comprehensive analysis. The procedural superiority of the included investigations was evaluated using the Joanna Briggs Institute Critical Appraisal Tools for assessing methodological quality. The Auto-Metric Graph Neural Network achieved the greatest accuracy of 99.57% for combined diabetic retinopathy-DME grading, illustrating the higher efficacy of hybrid architectures augmented by meta-heuristic optimizers, such as Jaya and ant colony optimization. Successful deployment, however, depends on overcoming hurdles, such as the low mean average precision in lesion identification (0.1540) in YOLO-based models in the test set performance, and improved clinical interpretability to foster clinician trust. A Sankey diagram visually analyzes the flow of quantities between different entities of the survey.Systematic review registrationB. (2025, November 2). A Review of Optimization Strategies for Deep and Machine Learning in DME. Retrieved from osf.io/qsh4j.

Jakub Fusiak, Andreas Wolkenstein, Verena S. Hoffmann

BackgroundPatient preferences are a critical component of shared decision-making (SDM), particularly when choosing between treatment options with differing risks and outcomes. Many methods exist to elicit these preferences, but their complexity, usability, and acceptance vary.ObjectiveWe aim to gain insight into the acceptance, effort and preferences of participants regarding five different methods of preference assessment. Additionally, we investigate the influence of health status, experiences within the health system and of demographic factors on the results.MethodsWe conducted a cross-sectional online survey including five preference elicitation Methods: best-worst scaling, direct weighting, PAPRIKA (Potentially All Pairwise Rankings of all Possible Alternatives), time trade-off, and standard gamble. The questionnaire was distributed via academic and patient advocacy mailing lists, reaching both healthy individuals and those with acute or chronic illnesses. Participants rated each method using six standardized statements on a 5-point Likert scale. Additional items assessed general acceptance of algorithm-assisted preference assessments and the clarity of the questionnaire.ResultsOf 258 initiated questionnaires, 123 (48%) were completed and included in the analysis. Participants were diverse in age, gender, and health status, but predominantly highly educated and digitally literate. Across all measures, the PAPRIKA method received the highest ratings for clarity, usability, and perceived ability to express preferences. Simpler methods (best-worst scaling, direct weighting) were rated as less useful for capturing nuanced preferences, while abstract utility-based methods (standard gamble, time trade-off) were seen as cognitively demanding. Subgroup analyses showed minimal variation across demographic groups. Most participants (82%) could imagine using at least one of the presented methods in real clinical settings, but also emphasized the importance of physician involvement in interpreting results.ConclusionThe interactive PAPRIKA method best balanced cognitive demand and expressiveness and was preferred by most participants. Structured methods for preference elicitation may enhance SDM when integrated into clinical workflows and supported by healthcare professionals. Further research is needed to evaluate their use in real-world decisions and among more diverse patient populations.

Jianjun Chen, Liyuan Zheng, Huilai Zou et al.

Precise segmentation of individual teeth from digital three-dimensional (3D) tooth models is critical in computer-assisted orthodontic surgery. This study explores the application of Point Multi-Layer Perceptron (PointMLP) in processing 3D tooth models and introduces an innovative integration of the Graph Attentional Convolution (GAC) Layer with a graph attention mechanism. By incorporating the GAC Layer into PointMLP, the model can focus on key local regions in the 3D tooth model and dynamically adjust the attention applied to these areas. This enhanced attention mechanism allows the model to better capture subtle surface structures, facilitating the accurate extraction of valuable local features. Compared to other traditional segmentation algorithms, the proposed method shows improvements of 1.1, 2.04, 1.06, 2.2, and 1.8 percentage points in Overall Accuracy (OA), Sensitivity (SEN), Positive Predictive Value (PPV), and Intersection Over Union (IoU), respectively. At the same number of training epochs, our method outperforms both GAC and PointMLP in segmentation performance.

Zanyu Tang, Mingzhi Mao, Yunong Zhang et al.

Two innovative acceleration-layer configuration amendment (CA) schemes are proposed to achieve the CA of constrained redundant robot arms. Specifically, by applying the Zhang neurodynamics equivalency (ZNE) method, an acceleration-layer CA performance indicator is derived theoretically. To obtain a unified-layer inequality constraint by transforming from angle-layer and velocity-layer constraints to acceleration-layer constraints, five theorems and three corollaries are theoretically derived and rigorously proved. Then, together with the unified acceleration-layer bound constraint, an enhanced acceleration-layer CA scheme specially considering three-layer time-variant physical limits is proposed, and a simplified acceleration-layer CA scheme considering three-layer time-invariant physical limits is also proposed. The proposed CA schemes are finally formulated in the form of standard quadratic programming and are solved by a projection neurodynamics solver. Moreover, comparative simulative experiments based on a four-link planar arm and a UR3 spatial arm are performed to verify the efficacy and superiority of the proposed CA schemes. At last, physical experiments are conducted on a real Kinova Jaco2 arm to substantiate the practicability of the proposed CA schemes.

Yiming Bao, Yongbiao Xue

Anirban Chaudhuri, Graham Pash, David A. Hormuth et al.

We develop a methodology to create data-driven predictive digital twins for optimal risk-aware clinical decision-making. We illustrate the methodology as an enabler for an anticipatory personalized treatment that accounts for uncertainties in the underlying tumor biology in high-grade gliomas, where heterogeneity in the response to standard-of-care (SOC) radiotherapy contributes to sub-optimal patient outcomes. The digital twin is initialized through prior distributions derived from population-level clinical data in the literature for a mechanistic model's parameters. Then the digital twin is personalized using Bayesian model calibration for assimilating patient-specific magnetic resonance imaging data. The calibrated digital twin is used to propose optimal radiotherapy treatment regimens by solving a multi-objective risk-based optimization under uncertainty problem. The solution leads to a suite of patient-specific optimal radiotherapy treatment regimens exhibiting varying levels of trade-off between the two competing clinical objectives: (i) maximizing tumor control (characterized by minimizing the risk of tumor volume growth) and (ii) minimizing the toxicity from radiotherapy. The proposed digital twin framework is illustrated by generating an in silico cohort of 100 patients with high-grade glioma growth and response properties typically observed in the literature. For the same total radiation dose as the SOC, the personalized treatment regimens lead to median increase in tumor time to progression of around six days. Alternatively, for the same level of tumor control as the SOC, the digital twin provides optimal treatment options that lead to a median reduction in radiation dose by 16.7% (10 Gy) compared to SOC total dose of 60 Gy. The range of optimal solutions also provide options with increased doses for patients with aggressive cancer, where SOC does not lead to sufficient tumor control.

Eden Ozeri, Meirav Zehavi, Michal Ziv-Ukelson

Abstract We define two new computational problems in the domain of perfect genome rearrangements, and propose three algorithms to solve them. The rearrangement scenarios modeled by the problems consider Reversal and Block Interchange operations, and a PQ-tree is utilized to guide the allowed operations and to compute their weights. In the first problem, $$\mathsf {Constrained \ TreeToString \ Divergence}$$ Constrained TreeToString Divergence ( $$\textsf{CTTSD}{}$$ CTTSD ), we define the basic structure-informed rearrangement measure. Here, we assume that the gene order members of the gene cluster from which the PQ-tree is constructed are permutations. The PQ-tree representing the gene cluster is ordered such that the series of gene IDs spelled by its leaves is equivalent to that of the reference gene order. Then, a structure-informed genome rearrangement distance is computed between the ordered PQ-tree and the target gene order. The second problem, $$\mathsf {TreeToString \ Divergence}$$ TreeToString Divergence ( $$\textsf{TTSD}{}$$ TTSD ), generalizes $$\textsf{CTTSD}{}$$ CTTSD , where the gene order members are not necessarily permutations and the structure informed rearrangement measure is extended to also consider up to $$d_S$$ d S and $$d_T$$ d T gene insertion and deletion operations, respectively, when modelling the PQ-tree informed divergence process from the reference gene order to the target gene order. The first algorithm solves $$\textsf{CTTSD}{}$$ CTTSD in $$O(n \gamma ^2 \cdot (m_p \cdot 1.381^\gamma + m_q))$$ O ( n γ 2 · ( m p · 1 . 381 γ + m q ) ) time and $$O(n^2)$$ O ( n 2 ) space, where $$\gamma $$ γ is the maximum number of children of a node, n is the length of the string and the number of leaves in the tree, and $$m_p$$ m p and $$m_q$$ m q are the number of P-nodes and Q-nodes in the tree, respectively. If one of the penalties of $$\textsf{CTTSD}$$ CTTSD is 0, then the algorithm runs in $$O(n m \gamma ^2)$$ O ( n m γ 2 ) time and $$O(n^2)$$ O ( n 2 ) space. The second algorithm solves $$\textsf{TTSD}{}$$ TTSD in $$O(n^2 \gamma ^2 {d_T}^2 {d_S}^2\,m^2 (m_p \cdot 5^\gamma \gamma + m_q))$$ O ( n 2 γ 2 d T 2 d S 2 m 2 ( m p · 5 γ γ + m q ) ) time and $$O(d_T d_S m (m n + 5^\gamma ))$$ O ( d T d S m ( m n + 5 γ ) ) space, where $$\gamma $$ γ is the maximum number of children of a node, n is the length of the string, m is the number of leaves in the tree, $$m_p$$ m p and $$m_q$$ m q are the number of P-nodes and Q-nodes in the tree, respectively, and allowing up to $$d_T$$ d T deletions from the tree and up to $$d_S$$ d S deletions from the string. The third algorithm is intended to reduce the space complexity of the second algorithm. It solves a variant of the problem (where one of the penalties of $$\textsf{TTSD}$$ TTSD is 0) in $$O(n \gamma ^2 {d_T}^2 {d_S}^2\,m^2 (m_p \cdot 4^{\gamma }\gamma ^2n(d_T+d_S+m+n) + m_q))$$ O ( n γ 2 d T 2 d S 2 m 2 ( m p · 4 γ γ 2 n ( d T + d S + m + n ) + m q ) ) time and $$O(\gamma ^2 n m^2 d_T d_S (d_T+d_S+m+n))$$ O ( γ 2 n m 2 d T d S ( d T + d S + m + n ) ) space. The algorithm is implemented as a software tool, denoted MEM-Rearrange, and applied to the comparative and evolutionary analysis of 59 chromosomal gene clusters extracted from a dataset of 1487 prokaryotic genomes.

XIONG Zhongmin, ZENG Qi, LU Peng, WANG Zhenhua, ZHENG Zongsheng

Logical reasoning is the ability to perceive patterns and connections between visual elements. Endowing computers with human-like reasoning ability is a critical area of research；state-of-the-art deep neural networks have achieved superhuman performance in image processing and other fields.However，the concept of logical reasoning through images requires further research.To address the problems of insufficient feature extraction and generalization of Multi-scale Relation Network（MRNet），an improved logical reasoning method，called Residual Attention Multi-scale Relation Network（ResAMRNet），is proposed. In the backbone network，shallow features are integrated into the deep network training process by utilizing residual structures and combining jump and long jump. This reduces the loss of feature information and improves the feature extraction capability of the model. In the reasoning module，the channel attention mechanism and residuals are combined to detect the relationship features between each image line.It can differentiate the significance of each feature channel，learn the attention weight adaptively，and extract key features.In this study，a Double-pooled Efficient Channel Attention（DECA） mechanism is proposed to combine global maximum pooling to further obtain feature information regarding objects and to improve generalization.Experimental results on representative logical reasoning datasets，Relational and Analogical Visual rEasoNing（RAVEN） and Improved RAVEN（I-RAVEN），show that the accuracy of the proposed method using these datasets is higher by 8.3 and 18.1 percentage points，respectively，than that of MRNet. Therefore，it demonstrates strong logical reasoning capabilities.

Noor Hellyda Hermawati, Susy Rosyida

PT. Telkom Akses Pontianak memiliki sistem informasi Inventory yang selama ini digunakan, selama melakukan penelitian ditemukanlah beberarapa temuan, yaitu seperti informasi terkait ketersedian material, sistem yang kurang efektif terkait data pengeluaran barang yang berdampak pada laporan periodik perusahaan, dan kurangnya optimalisasi Sumber Daya Manusia yang ada. sehingga dengan permsalahan yang ada menjadi dasar untuk melakukan audit sistem informasi yang digunakan. Audit mengacu pada framework COBIT 5 dengan menggunakan Domain MEA ditemukanlah hasil dari tingkat kapabilitas masing-masing sub domain MEA itu sendiri dan juga Gap Analisisnya. Dengan nilai kapabilitas dari subdomain MEA 01 senilai 3,83, Subdomain MEA 02 senilai 3,60, dan Subdomain MEA 03 senilai 3,69, dengan nilai rata-rata yaitu 3,70 dengan keterangan Predictable Process yang berarti objek yang diteliti sudah mencapai proses yang ditetapkan berjalan dalam suatu batas yang ditentukan untuk mencapai tujuan prosesnya. Serta dengan perhitungan Gap Analisis yaitu pada subdomain MEA 01 senilai 1,2, Subdomain MEA 02 senilai 1,4, dan Subdomain MEA 03 senilai 1,3, dengan nilai rata-rata yaitu 1,3 yang berarti perusahaan masih perlu meningkatkan terkait sistem informasi Inventory yang digunakan agar dapat memperoleh hasil yang optimal bagi seluruh pemangku kepentingan.

Rattanakorn Wattanataweekul, Kobkoon Janngam

Abstract In this paper, we define a new concept of left and right coordinate affine of a directed graph and then employ it to introduce a new accelerated common fixed point algorithm for a countable family of G-nonexpansive mappings in a real Hilbert space with a graph. We prove, under certain conditions, weak convergence theorems for the proposed algorithm. As applications, we also apply our results to solve convex minimization and image restoration problems. Moreover, we show that our algorithm provides better convergence behavior than other methods in the literature.

Hanne Lefrère, Giuseppe Floris, Marjanka K. Schmidt et al.

Postpartum breast cancer (PPBC) - which according to new data, can extend to 5–10 years after the birth - are estimated to represent 35–55% of all cases of breast cancer in women younger than 45 years. Increasing clinical evidence indicates that PPBC represents a high-risk form of breast cancer in young women with an approximately 2-fold increased risk for metastasis and death. Yet, the exact mechanisms that underlay this poor prognosis are incompletely understood and, hence, it is unknown why postpartum breast cancer has an enhanced risk for metastasis or how it should be effectively targeted for improved survival. This article is an accompanying resource of the original article entitled “Breast cancer diagnosed in the post-weaning period is indicative for a poor outcome” and present epidemiological data that compare standard prognostic parameters, first site of metastatic disease and survival and metastatic rates in young women with primary invasive breast cancer diagnosed within two years postpartum (PP-BC), in young women diagnosed during pregnancy (Pr-BC) and nulliparous women (NP-BC). Via an international collaboration of 13 centres participating in the International Network on Cancer, Infertility and Pregnancy (INCIP), retrospective data of 1180 patients with primary invasive breast cancer, aged 25–40 years and diagnosed between January 1995 and December 2017 were collected. In particular, tumour-, patient, and therapy-related characteristics were collected. Furthermore, patient files were reviewed thoroughly to assess, for each parity, if and for how long breastfeeding was given. For PP-BC patients, breastfeeding history was used to differentiate breast cancers identified during lactation (PP-BCDL) from those diagnosed post-weaning (PP-BCPW). Primary exposures were prior childbirth or no childbirth, time between most recent childbirth and breast cancer diagnosis, time between cessation of lactation and breast cancer diagnosis and time between breast cancer diagnosis and metastasis or death. Distribution of standard prognostic parameters and first site of distant metastasis among study groups was determined applying fisher's exact, chi-squared, One-Way ANOVA or Kruskal-Wallis tests or logistic regression models, where applicable. The risks for metastasis and death were assessed using Cox proportional hazards models. A subgroup analysis was performed in PP-BCPW patients that never lactated (PP-BCPW/NL), lactated ≤3 months (PP-BCPW/Lshort) or lactated >3 months (PP-BCPW/Llong).

Felix Uba, Eric Osei Essandoh, Gilbert Ayine Akolgo et al.

This research presents a study on the heat transfer coefficient for an unglazed solar plate collector in an unsteady humid outdoor environment. The purpose for undertaking this research is to investigate the correlation between the heat transfer coefficient and air speed and also verify whether heat transfer from unglazed solar thermal collectors under outdoor conditions can be experimentally determined using a particular mathematical relationship for different locations. In estimating the heat transfer coefficient for an unglazed solar plate in an unsteady humid outdoor condition, an experiment was held using an outdoor setup that measured temperatures, wind speeds, and solar radiations from 11:00 A.M. to 2:00 P.M. The solar plate collector was placed on a flat bed of height 2.2 m and a collection area of 0743 m2. An average temperature of 45°C was recorded for a mild steel plate collector which was initially exposed to an ambient temperature which ranges from 25°C to 32°C. The interfacial temperature between the plate and an asbestos board ranges from 42°C to 52°C, and that of the asbestos and a plywood is 40°C to 46°C. The specific heat capacity of the mild steel plate and the asbestos board used for the construction of the experimental setup are 25.00 kJ/kg and 950.00 kJ/kg, respectively, while the thermal conductivity of these materials is 0.46 W/m·K and 0.25 W/m·K, respectively. The novelty of this work is the use of such a study to generate empirical equations for Ghana and to produce representative equations for determining the heat transfer coefficient for solar plate collectors in unsteady humid outdoor conditions in West Africa. This work is expected to contribute data alongside similar works done for different areas to help propose empirical equations for estimating global and not site-specific heat transfer coefficients.