PubMed İndeksli Yayınlar Koleksiyonu / PubMed Indexed Publications Collection
Permanent URI for this collectionhttps://hdl.handle.net/11147/7645
Browse
3 results
Search Results
Now showing 1 - 3 of 3
Article Citation - WoS: 55Citation - Scopus: 56Evaluation of an Artificial Intelligence System for Diagnosing Scaphoid Fracture on Direct Radiography(Springer Verlag, 2020) Özkaya, Emre; Topal, Fatih Esad; Bulut, Tuğrul; Gürsoy, Merve; Özuysal, Mustafa; Karakaya, ZeynepPurpose The aim of this study is to determine the diagnostic performance of artificial intelligence with the use of convolutional neural networks (CNN) for detecting scaphoid fractures on anteroposterior wrist radiographs. The performance of the deep learning algorithm was also compared with that of the emergency department (ED) physician and two orthopaedic specialists (less experienced and experienced in the hand surgery). Methods A total 390 patients with AP wrist radiographs were included in the study. The presence/absence of the fracture on radiographs was confirmed via CT. The diagnostic performance of the CNN, ED physician and two orthopaedic specialists (less experienced and experienced) as measured by AUC, sensitivity, specificity, F-Score and Youden index, to detect scaphoid fractures was evaluated and compared between the groups. Results The CNN had 76% sensitivity and 92% specificity, 0.840 AUC, 0.680 Youden index and 0.826Fscore values in identifying scaphoid fractures. The experienced orthopaedic specialist had the best diagnostic performance according to AUC. While CNN's performance was similar to a less experienced orthopaedic specialist, it was better than the ED physician. Conclusion The deep learning algorithm has the potential to be used for diagnosing scaphoid fractures on radiographs. Artificial intelligence can be useful for scaphoid fracture diagnosis particularly in the absence of an experienced orthopedist or hand surgeon.Article Citation - WoS: 12Citation - Scopus: 15The Role of Loco-Regional Treatment in Long-Term Quality of Life in De Novo Stage Iv Breast Cancer Patients: Protocol Mf07-01q(Springer Verlag, 2021) Soran, Atilla; Soyder, Aykut; Özbaş, Serdar; Özmen, Vahit; Karanlık, Hasan; İğci, Abdullah; Sezgin, EfeBackground/objective Since more solid evidence has emerged supporting the effectiveness of loco-regional treatment (LRT), clinicians consider LRT a treatment option for selected de novo stage IV breast cancer (BC) patients. This is the first report on long-term quality of life (QoL) in a cohort of patients who were randomized to receive either LRT and then systemic treatment (ST) or ST alone in the protocol MF07-01. We aimed to evaluate QoL in patients living at least 3 years since randomization using scores from the SF-12 health survey. Methods SF-12 (V2) forms were completed during visits of patients who were living 36 months after the randomization. We first calculated PCS-12 (Physical Health Composite Scale) and MCS-12 (Mental Health Composite Scale) scores from de novo stage IV BC patients and compared them with the scores of patients diagnosed with stage I-III BC who lived more than 3 years. Further, PCS-12 and MCS-12 scores were compared between the LRT and ST groups with de novo stage IV BC. Additionally, general health, physical functioning, role functioning, bodily pain, vitality, mental health, and social functioning were evaluated and compared between the groups. Considering age-related changes in QoL, we also compared PCS-12 and MCS-12 scores of patients below or above 55 and 65 years of age. Responses to four additional questions (compare your physical health, mental health, daily activities, and energy currently vs. at diagnosis of BC) were recorded, considering cultural differences. Results There were 81 patients in this analysis; 68% of patients (n = 55) had LRT, and 32% (n = 26) received ST. General health was good or very good in 62% (n = 34) in the LRT group and 66% (n = 17) in the ST-only group (p = 0.63). Mean PCS-12 score was 40.8 + 1.6, and mean MCS-12 score was 43.4 + 2.0 (p = 0.34 and p = 0.54, respectively). PCS-12 and MCS-12 score difference was lower than that of the general Turkish population (PCS-12 = 49.3 + 12.8 and MCS-12 = 46.8 + 13.0) and stage I-III BC patients (PCS-12 = 51.1 +/- 0.5, MCS-12 = 45.7 +/- 0.6). PCS-12 and MCS-12 scores were similar between the LRT and ST-only groups in patients younger and older than 55 and 65, but QoL scores were much better in stage I-III BC patients younger than 65 when compared to the scores of those with de novo stage IV BC. Although treatment with or without LRT did not affect physical health, mental health, daily activities, and energy at 3 years vs. at diagnosis of BC in de novo stage IV BC patients (p > 0.05), these variables were significantly better in stage I-III BC patients (p < 0.001). Conclusion The current MF07-01Q study demonstrates that patient who had LRT has similar physical and mental health outcomes compared to ST only in a cohort of patients who lived longer than 3 years. Trial registration This study is registered on clinicaltrials.gov with identifier number NCT00557986.Article Citation - WoS: 10Citation - Scopus: 13Analysis of European Hazelnut (corylus Avellana) Reveals Loci for Cultivar Improvement and the Effects of Domestication and Selection on Nut and Kernel Traits(Springer Verlag, 2019) Frary, Amy; Öztürk, Süleyman Can; Balık, Hüseyin İrfan; Kayalak Balık, Selda; Kızılcı, Gökhan; Doğanlar, Sami; Frary, AnneTurkey is a rich source of European hazelnut (Corylus avellana) germplasm with nearly 400 accessions in the national collection. This genetic material encompasses cultivars, landraces and wild genotypes which were characterized for 12 nut and 13 kernel traits over 2years in the 1990s. Analysis of these attributes revealed both the positive and negative impacts that human selection and breeding have had on hazelnut. Thus, while selection has resulted in larger nuts and kernels, cultivars have fewer nuts per cluster and kernels with larger internal cavities. Breeding has also resulted in a propensity for cultivars to have higher proportions of double kernels and empty nuts, two traits which reduce quality and yield. In addition, it is clear that while selection has successfully increased hazelnut fat content it has not impacted overall flavor, a much more complex trait. The nut and kernel phenotypic data were combined with genotypic data from 406 simple sequence repeat marker alleles for association mapping of the quantitative trait loci (QTL) for the traits. A total of 78 loci were detected in the population with the highest proportions for nut (24%) and kernel (26%) appearance parameters followed by quality (19%), shell thickness (16%) and yield-related (15%) traits. It is hoped that some of the identified QTL will be useful for future breeding of hazelnut for improved nut and kernel yield and quality.