PubMed İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12573/397
Browse
2 results
Search Results
Article Citation - WoS: 17Citation - Scopus: 18Microfluidic Chip Based Direct Triple Antibody Immunoassay for Monitoring Patient Comparative Response to Leukemia Treatment(Springer, 2020-07-13) Icoz, Kutay; Akar, Unal; Unal, EkremWe report a time and cost-efficient microfluidic chip for screening the leukemia cells having three specific antigens. In this method, the target blast cells are double sorted with immunomagnetic beads and captured by the 3rd antibody immobilized on the gold surface in a microfluidic chip. The captured blast cells in the chip were imaged using a bright-field optical microscope and images were analyzed to quantify the cells. First sorting was performed with nano size immunomagnetic beads and followed by 2nd sorting where micron size immunomagnetic beads were used. The low-cost microfluidic platform is made of PMMA and glass including micro size gold pads. The developed microfluidic platform was optimized with cultured B type lymphoblast cells and tested with the samples of leukemia patients. The 8 bone marrow samples of 4 leukemia patients on the initial diagnosis and on the 15th day after the start of the chemotherapy treatment were tested both with the developed microfluidic platform and the flow cytometry. A 99% statistical agreement between the two methods shows that the microfluidic chip is able to monitor the decrease in the number of blast cells due to the chemotherapy. The experiments with the patient samples demonstrate that the developed system can perform relative measurements and have a potential to monitor the patient response to the applied therapy and to enable personalized dose adjustment.Article Citation - Scopus: 10Building a Challenging Medical Dataset for Comparative Evaluation of Classifier Capabilities(Elsevier Ltd, 2024-08) Bozkurt, Berat; Coskun, Kerem; Bakal, GokhanSince the 2000s, digitalization has been a crucial transformation in our lives. Nevertheless, digitalization brings a bulk of unstructured textual data to be processed, including articles, clinical records, web pages, and shared social media posts. As a critical analysis, the classification task classifies the given textual entities into correct categories. Categorizing documents from different domains is straightforward since the instances are unlikely to contain similar contexts. However, document classification in a single domain is more complicated due to sharing the same context. Thus, we aim to classify medical articles about four common cancer types (Leukemia, Non-Hodgkin Lymphoma, Bladder Cancer, and Thyroid Cancer) by constructing machine learning and deep learning models. We used 383,914 medical articles about four common cancer types collected by the PubMed API. To build classification models, we split the dataset into 70% as training, 20% as testing, and 10% as validation. We built widely used machine-learning (Logistic Regression, XGBoost, CatBoost, and Random Forest Classifiers) and modern deep-learning (convolutional neural networks - CNN, long short-term memory - LSTM, and gated recurrent unit - GRU) models. We computed the average classification performances (precision, recall, F-score) to evaluate the models over ten distinct dataset splits. The best-performing deep learning model(s) yielded a superior F1 score of 98%. However, traditional machine learning models also achieved reasonably high F1 scores, 95% for the worst-performing case. Ultimately, we constructed multiple models to classify articles, which compose a hard-to-classify dataset in the medical domain. © 2024 Elsevier B.V., All rights reserved.