58 research outputs found
Ensemble Learning for Three-dimensional Medical Image Segmentation of Organ at Risk in Brachytherapy Using Double U-Net, Bi-directional ConvLSTM U-Net, and Transformer Network
No full textAim:
This article presents a novel approach to automate the segmentation of organ at risk (OAR) for high-dose-rate brachytherapy patients using three deep learning models combined with ensemble learning techniques. It aims to improve the accuracy and efficiency of segmentation.
Materials and Methods:
The dataset comprised computed tomography (CT) scans of 60 patients obtained from our own institutional image bank and 10 patients from the other institute, all in Digital Imaging and Communications in Medicine format. Experienced radiation oncologists manually segmented four OARs for each scan. Each scan was preprocessed and three models, Double U-Net (DUN), Bi-directional ConvLSTM U-Net (BCUN), and Transformer Networks (TN), were trained on reduced CT scans (240 × 240 × 128) due to memory limitations. Ensemble learning techniques were employed to enhance accuracy and segmentation metrics. Testing and validation were conducted on 12 patients from our institute (OID) and 10 patients from another institute (DID).
Results:
For DID test dataset, using the ensemble learning technique combining Transformer Network (TN) and BCUN, i.e., TN + BCUN, the average Dice similarity coefficient (DSC) ranged from 0.992 to 0.998, and for DUN and BCUN (DUN + BCUN) combination, the average DSC ranged from 0.990 to 0.993, which reflecting high segmentation accuracy. The 95% Hausdorff distance (HD) ranged from 0.9 to 1.2 mm for TN + BCUN and 1.1 to 1.4 mm for DUN + BCUN, demonstrating precise segmentation boundaries.
Conclusion:
The proposed method leverages the strengths of each network architecture. The DUN setup excels in sequential processing, the BCUN captures spatiotemporal dependencies, and transformer networks provide a robust understanding of global context. This combination enables efficient and accurate segmentation, surpassing human expert performance in both time and accuracy
Current Debates in Clinical Decision Support Systems Research in the Field of Medical Informatics: What Text Mining Tell Us?
No full textCurrent Debates in Clinical Decision Support SystemsResearch in the Field of Medical Informatics: What Text Mining Tell Us?Soniya Norouzpoura, ÖmerFaruk ŞAYLANb, Muhammet DAMARa a Dokuz Eylul University, Faculty ofScience, Department of Computer Science, Tinaztepe Campus, Buca, Izmir, Turkiyeb Ege University, Faculty of Engineering,Department of Computer Engineering, Ege University Central Gate, Bornova,Izmir, Turkiye email: [email protected] AbstractMedical informaticsholds significant importance in today’s healthcare systems. At its core, it aims toenhance the quality of healthcare services by ensuring the accurate, effective,and secure utilization of health data [1]. Additionally, this field is highlyinterdisciplinary and plays a critical role in the software industry. GivenTurkey's strong position in the health sciences, fostering deeper collaborationwith computer sciences could significantly strengthen its role in this domain[2,3,4,5].This study underscores the importance of medical informatics andseeks to identify prominent discussions in clinical decision support systemsresearch using text mining methodologies.To achieve this, we analyzed researcharticles published between 2000and 2024 in the Web of Science(WoS) database, collected on March5, 2025. The study aims to highlight the key topics emerging in thefield of clinical decision support systems.This project was developed as an outcome of the TÜBİTAK 2209-A research initiative. Forthe literature search, the keywords “clinic decisionsupport”*and “clinical decision support”* were used to filterthe relevant articles. The study employed R Bibliometrix Biblioshiny, VOSviewer, and Pythonprogramming language with libraries such as Scikit-learn, Gensim, and Wordcloud for data analysis.The findingsindicate a growing interest in clinicaldecision support tools in the healthcare literature over the years.While only one article waspublished in 2000, the numberincreased to 170 articles in 2024,totaling 1,431 publications inthis period. Turkey contributed 15articles, ranking 23rd inthe global literature.The analysis highlights that artificialintelligenceand machine learning have emerged as critical tools in clinical decisionsupport systems in recent years. Key topics identified in the field includeprimary care, quality improvement, pediatrics, emergency medicine, precisionmedicine, clinical guidelines, and computerized physician order entry systems.Furthermore, business intelligence tools proved highly useful in implementingand processing the data in this study [6,7]. Medical informatics is an areawhere Turkey should aim to strengthen its position significantly [1,4,5].Similar to the advancements in the defense industry in recent years, it isimperative to extend this progress to the health sector. To achieve this, fostering stronger collaborations with fields such as materials science, fundamental sciences,electrical-electronics engineering, and computer sciences is of criticalimportance. Additionally, developing nationalscientific policies in this area is essential. Such initiatives will notonly advance Turkey’s healthcare system and establish a robustdomestic market, but also enhance the country’s competitiveness inthe international market with proven and innovative medical technologies.Keywords:Medical Informatics; Health Informatics; ClinicalDecision Support Systems; Decision Support; Text Mining. References [1] Damar, M., Küme, T., Yüksel, İ., Çetinkol, A. E., Pal, J.K., & Erenay, F. S. (2024). Medical Informatics as a Concept andField-Based Medical Informatics Research: The Case of Turkey. Duzce MedicalJournal, 26(1), 44-55.[2] Damar, M., & Özdağoğlu, G. (2021). Yazılımsektörü veuluslararasılaşma, politika önerileri. Editör, Ömer Aydın & ÇağdaşCengiz. Teknoloji ve Uluslararası İlişkiler. Nobel Yayıncılık: Ankara.[3] Damar, M. (2022). Dijital Dünyanın Dünü, Bugünü Ve Yarını: Bilişim Sektörünün Gelişimi Üzerine Değerlendirme.Nevşehir Hacı Bektaş Veli Üniversitesi Sbe Dergisi,12(Dijitalleşme), 51-76.[4] Damar, M., Özdağoğlu, G., & Özveri, O.(2020). ÜniversitelerdeDö<span style="font-size: 9pt; font-family: "Times New Roman", serif; color: rgb(34, 34, 34); background-image: initial; background-position: initial; background-size: initial; background-repeat: initial; background-attachment: initial; background-ori
Efficacy Of Beta-Carotene In Modulating Vitamin-A Deficiency Induced Biochemical Changes In Liver Of Rat
No full textThis Dissertation / Report is the outcome of investigation carried out by the creator(s) / author(s) at the department/division of Central Food Technological Research Institute (CFTRI), Mysore mentioned below in this page
Industrial Park Role as a Catalyst for Regional Development: Zooming on Middle East Countries
No full textThe development of the industrial park has been one of the priorities of the plans of different countries and has created a magnitude challenge concerning regional disparities. Globally, the Middle Eastern countries have demonstrated a more remarkable dedication to industrial park development, given its general importance since the 1970s. Due to this importance, this paper is divided into three sections due to the critical role of industrial park development in the case of Middle Eastern countries. First, this study highlighted the relevant literature using Scinotometric analysis. In the second step, following the investigation of the relationship between selected critical variables and the development of industrial parks towards regional development in the Middle Eastern countries from 2000 to 2018. In this regard, panel data were used to determine the association between the selected variables and industrial park performance. According to the findings, the author suggests policy implementation for industrial park development in three categories: economic growth, environmental issues, and reduction in regional disparities. Finally, this study can serve as a foundation for future research, such as comparing the first batch of industrial parks with their upgraded counterparts in the Middle East and studying the competitive advantages issues
A Combined Loss-driven Framework for Automated Parotid Segmentation in Head-and-Neck Computed Tomography
No full textPurpose:
This study presents a deep learning framework for automatic parotid segmentation using three-dimensional (3D) U-Net and attention-augmented 3D U-Net architectures trained with a novel combined loss function tailored for anatomical accuracy and class imbalance.
Materials and Methods:
A curated dataset of 379 noncontrast head-and-neck computed tomography scans with expert-verified contours was used. Two architectures a residual 3D U-Net and its attention-enhanced variant were implemented using TensorFlow. The networks were trained with both categorical cross-entropy and a proposed combined loss integrating modified Dice Score Coefficient (mDSC) and focal loss (FL) with weights 0.7 and 0.3. The models were evaluated using dice similarity coefficient (DSC), Intersection over Union (IoU), and categorical accuracy. A custom checkpointing strategy was designed to preserve model weights corresponding to both peak DSC and minimum validation loss. The code and pretrained models are hosted on a publicly available GitHub repository at: https://github.com/1aryantyagi/Segmentation-Paper.
Results:
The 3D U-Net trained with the combined loss achieved a median Dice score of 0.8835 (left parotid) and 0.8709 (right), with mean IoU values of 0.7672 and 0.7358, indicating strong segmentation accuracy. The U-Net produced comparable results, supporting the combined loss’s consistency. Bland–Altman analysis confirmed reduced variability and improved agreement.
Conclusion:
The integration of mDSC and FL within a 3D U-Net architecture significantly improves segmentation performance, robustness, and spatial precision. These findings support the clinical feasibility of the proposed framework for automated, reproducible parotid delineation in radiotherapy planning
Industrial organization and trade liberalization : evidence from Korea
No full textDrawing on evidence about industrial organization and market structure, the authors develop a computable general equilibrium model in selected industrial sectors with increasing returns to scale. They use this model to estimate the welfare gains Korea would realize from abolishing the import restraints prevailing in 1982. Under constant returns to scale, they estimate welfare gains to be 1 percent of GDP. With increasing returns to scale in three industrial sectors, they estimate welfare gains ranging from -0.5 percent to 10 percent of 1982 GDP, depending on assumptions about the pricing behavior (markup pricing or Cournot competition) and profit levels that existed under protection.Economic Theory&Research,Environmental Economics&Policies,Markets and Market Access,Access to Markets,TF054105-DONOR FUNDED OPERATION ADMINISTRATION FEE INCOME AND EXPENSE ACCOUNT
Advancing the COO Construct From an Affective Dimension : The Application of Projective Technique
No full textMaster thesis, Master's degree program Marketing, Economic School Linnaeus University, Växjö Sweden, Spring semester 2015. Author: Andersson Anthon, Guntell Robin Tutor: Soniya Billore Examiner: Anders Pehrsson Title: Advancing the COO Construct From an Affective Dimension: The Application of Projective Technique Purpose: The purpose in this article is to break from traditional research and its accompanying cognitive research methods in order to advance the COO field from a more accurate perspective that also involves an affective dimension as well. Design/methodology/approach: Drawing from prior research in the COO field, the methodology accounted for assumptions that were tested in collage technique and ad copy technique. Findings: The results shows that some people only seems to be susceptible to COO influence when communicating emotional CSAs nonverbally, whilst some people only reveal rational CSAs when being cognitively asked about COO influence in a directed manner. As a result, the present findings might suggest that prior research in the academic field might suffer from bias. Practical implications: In the light of COO, managers should bear in mind that some people cannot be targeted with solely rely on a cognitive marketing communication strategy. More specifically, the ad copy technique provides guidelines for appropriate design of advertisements when one consider to serving the brand’s origin as salient cue in consumers’ minds. Originality/value: Advancing the COO construct with using collage technique, this study is to the best our knowledge the second to account for an affective dimension as well.
How do sensory cues and trust affect the customer experience? : A study on the relationships between sensory cues, trust and experience in the Swedish nightclub industry
No full textTitle How do sensory cues and trust affect the customer experience?, A study on the relationships between sensory cues, trust and experience in the Swedish nightclub industry Author Miralem Hasanovic Tutor Soniya Billore Examiner Sarah Philipson Course Marketing Master Programme, Advanced, Spring 2013, Master thesis, 30 ETCS Keywords Experience, Trust, Sensory Marketing, Sensory Cues in Offering Experiences, Nightclubs, Five Senses, Swedish restaurant industry. Purpose Purpose is to investigate the relationships between sensory marketing, trust and experience in nightclubs. Theory Sensory cues in offering experiences, Trust, Experiences. Method Mixed method approach/ Sequenced method 14 observations 7 interviews with companies 102 answers in a survey Findings Visual cues affect the experience mostly positive. Audio cues affect the experience both positive and negative, depending on other aspects as the possibility to escape the loud sound. Touch cues affect the experience mostly negative, which is possible to alter through interior and design. Scent cues are not affecting the industry as for the moment and there is a big gap to fill in here for the industry. Taste cues seem not to matter as much as the other cues. Trust is inflicting experience through expectation and fulfillment of promise. There is a weak (or least enough) correlation of trust and positive experience in nightclubs. Violence does not affect trust considerably towards a nightclu
The menu approach to developing country external debt : an analysis of commercial banks'choice behavior
No full textThis study provides evidence that bank characteristics are significant determinants of commercial-bank choice behavior when confronted with a menu of options. It develops a theoretical model of bank choice behavior and empirically tests its implications using data from the 1988 Brazilian financing package. The empirical results show that bank characteristics are capable of explaining over 80 percent of this choice. One of the main implications of the theoretical model is that under risk-neutrality assumption, financially stronger and more exposed banks prefer to exit. The findings have several important implications for the new debt reduction strategy. (i) First, larger debt reductions operated on a market basis are more costly, per unit of debt reduced. In order to increase debt reduction, weaker banks must be convinced to exit, increasing the needed exit price. (ii) Second, the exit price depends on the strength of the banking industry, and thus, the effectiveness of the present debt strategy is affected by changes in the world economy. In periods of booms, banks become stronger and exit prices are reduced. (iii) Third, regulators can affect the cost of debt reduction by altering the regulatory framework within which the banks operate. (iv) Fourth, LDC debt reductions are beneficial to the deposit insurance agencies of the major creditor nations.Financial Intermediation,Economic Theory&Research,Municipal Financial Management,Financial Crisis Management&Restructuring,Banks&Banking Reform
Advancing the COO Construct From an Affective Dimension : The Application of Projective Technique
No full textMaster thesis, Master's degree program Marketing, Economic School Linnaeus University, Växjö Sweden, Spring semester 2015. Author: Andersson Anthon, Guntell Robin Tutor: Soniya Billore Examiner: Anders Pehrsson Title: Advancing the COO Construct From an Affective Dimension: The Application of Projective Technique Purpose: The purpose in this article is to break from traditional research and its accompanying cognitive research methods in order to advance the COO field from a more accurate perspective that also involves an affective dimension as well. Design/methodology/approach: Drawing from prior research in the COO field, the methodology accounted for assumptions that were tested in collage technique and ad copy technique. Findings: The results shows that some people only seems to be susceptible to COO influence when communicating emotional CSAs nonverbally, whilst some people only reveal rational CSAs when being cognitively asked about COO influence in a directed manner. As a result, the present findings might suggest that prior research in the academic field might suffer from bias. Practical implications: In the light of COO, managers should bear in mind that some people cannot be targeted with solely rely on a cognitive marketing communication strategy. More specifically, the ad copy technique provides guidelines for appropriate design of advertisements when one consider to serving the brand’s origin as salient cue in consumers’ minds. Originality/value: Advancing the COO construct with using collage technique, this study is to the best our knowledge the second to account for an affective dimension as well.
- …
