Yazar "Aydin, Ozlem" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    Generating headlines for Turkish news texts with transformer architecture based deep learning method
    (Gazi Univ, Fac Engineering Architecture, 2024) Karaca, Abdulkadir; Aydin, Ozlem
    Nowadays, the Internet is a structure that people can access easily and at the same time produce content easily and without control. In parallel with this situation, the ability of extract information from the raw data that makes up big data has become more complex. The fact that the headlines of the contents contain uncontrolled and misleading elements makes it difficult to reach the right information. The headlines of the contents are important for people to reach the information they want in their limited time. In this study, it is aimed to produce headlines suitable for the content instead of headlines that may be misleading for news. For this purpose, an application that produces headlines for Turkish news with deep learning method has been developed. SuDer news corpus is used as dataset. For the training of the model, it is aimed to obtain more humanoid results in the production of news headlines by using the Transformer architecture, which is frequently preferred in natural language studies today and the abstract summarization method. In this study, in order to compare the performance of the Transformer model, models are prepared and trained with Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) architectures. At the end of 25 epochs of training with LSTM, GRU and Transformer architectures on the corpus, the values of loss are 1.03, 0.55 and 2.49 respectively. In the experiments performed on the validation data, measurements are made with ROUGE-1, ROUGE-2 and ROUGE-L metrics. As a result of the measurements, it is observed that the Transformer architecture is partially good, based on the metric values produced. In addition, when the headlines produced with these architectures are examined, it is observed that the headline obtained with the Transformer architecture produce headlines that are partially more suitable for the news content compared to other architectures.
  • Küçük Resim Yok
    Öğe
    Performance of the Bethesda System for Reporting Thyroid Cytology in Multi-Institutional Large Cohort of Pediatric Thyroid Nodules: A Detailed Analysis
    (Mdpi, 2022) Canberk, Sule; Barroca, Helena; Girao, Ines; Aydin, Ozlem; Uguz, Aysun; Erdogan, Kivilcim; Tastekin, Ebru
    Background: To evaluate the performance of TBSRTC through multi-institutional experience in the paediatric population and questioning the management recommendation of ATA Guidelines Task Force on Paediatric Thyroid Cancer; Methods: A retrospective search was conducted in 4 institutions to identify consecutive thyroid FNAC cases in paediatric population between 2000 and 2018. Following the 2nd TBSRTC, the risk of malignancy ratios (ROMs) was given in ranges and calculated by 2 different ways. Sensitivity, specificity, PPV, NPV and DA ratios were calculated using histologic diagnosis as the gold standard; Results: Among a total of 405 specimens, the distribution of cases for each category was, 44 (11%) for ND, 204 (50%) for B category, 40 (10%) for AUS/FLUS, 36 (9%) for FN/SFN, 24 (6%) for SFM and 57 (14%) for M categories. 153 cases have a histological diagnosis. The ratio of surgery was 23% in ND, 16% in the B, 45% for AUS/FLUS, 75% for SFN/FN and 92% for SFM and 75% in M categories; Conclusions: The data underlines the high ROM values in paediatric population which might be clinically meaningful. The high rate of malignancy of the cohort of operated patients (50%) also underlines the need of better preoperative indicators for stratification. Considering that more than half of the nodules in AUS/FLUS category were benign, direct surgery recommendation could be questionable as proposed in ATA 2015 guidelines.
  • Küçük Resim Yok
    Öğe
    What does the Data of 354,725 Patients from Turkey Tell Us About Cervical Smear Epithelial Cell Abnormalities? - The Epithelial Cell Abnormality Rate is Increasing - Quality Control Studies and Corrective Activity are Musts
    (Federation Turkish Pathology Soc, 2017) Turkmen, Ilknur Cetinaslan; Usubutun, Alp; Cakir, Asli; Aydin, Ozlem; Bolat, Filiz Aka; Akbulut, Metin; Altinay, Serdar
    Objective: There is no other screening program close to the success rate of PAP test. Cervical cytology constitutes a large workload so that quality control in cervical cytology is important for the quality assurance of pathology laboratories. Material and Method: In this study, we collected the cervical cytology results from all over Turkey and discussed the parameters influencing the quality of the PAP test. The study was conducted with Turkish gynaecopathology working group and 38 centers (totally 45 hospitals) agreed to contribute from 24 different cities. The study was designed to cover the cervical cytology results during 2013. The results were evaluated from the data based on an online questionnaire. Results: The total number of Epithelial Cell Abnormality was 18,020 and the global Epithelial Cell Abnormality rate was 5.08% in the total 354,725 smears and ranging between 0.3% to 16.64% among centers. The Atypical squamous cells/Squamous intraepithelial lesion ratios changed within the range of 0.21-13.94 with an average of 2.61. When the centers were asked whether they performed quality assurance studies, only 14 out of 28 centers, which shared the information, had such a control study and some quality parameters were better in these centers. Conclusion: There is an increase in the global Epithelial Cell Abnormality rate and there are great differences among centers. Quality control studies including the Atypical squamous cells/Squamous intraepithelial lesion ratio are important. Corrective and preventive action according to quality control parameters is a must. A cervical cytology subspecialist in every center can be utopic but a dedicated pathologist in the center is certainly needed.