Browsing by Author "Üner, A."

Now showing 1 - 6 of 6

Open Access
Automated cancer stem cell recognition in H and E stained tissue using convolutional neural networks and color deconvolution
(SPIE, 2017) Aichinger, W.; Krappe, S.; Çetin, A. Enis; Çetin-Atalay, R.; Üner, A.; Benz, M.; Wittenberg, T.; Stamminger, M.; Münzenmayer, C.
The analysis and interpretation of histopathological samples and images is an important discipline in the diagnosis of various diseases, especially cancer. An important factor in prognosis and treatment with the aim of a precision medicine is the determination of so-called cancer stem cells (CSC) which are known for their resistance to chemotherapeutic treatment and involvement in tumor recurrence. Using immunohistochemistry with CSC markers like CD13, CD133 and others is one way to identify CSC. In our work we aim at identifying CSC presence on ubiquitous Hematoxilyn and Eosin (HE) staining as an inexpensive tool for routine histopathology based on their distinct morphological features. We present initial results of a new method based on color deconvolution (CD) and convolutional neural networks (CNN). This method performs favorably (accuracy 0.936) in comparison with a state-of-the-art method based on 1DSIFT and eigen-analysis feature sets evaluated on the same image database. We also show that accuracy of the CNN is improved by the CD pre-processing.
Open Access
Çarpmasız yapay sinir ağı
(IEEE, 2015-05) Akbaş, Cem Emre; Bozkurt, Alican; Çetin, A. Enis; Çetin-Atalay, R.; Üner, A.
Bu bildiride çarpma işlemi kullanmadan oluşturulan bir Yapay Sinir Ağı (YSA) sunulmaktadır. Girdi vektörleri ve YSA katsayılarının iç çarpımları çarpmasız bir vektör işlemiyle hesaplanmıştır. Yapay sinir ağının eğitimi sign-LMS algoritması ile yapılmıştır. Önerilen YSA sistemi, hesap gücü kısıtlı olan veya düşük enerji tüketimine ihtiyaç duyulan mikroişlemcilerde kullanılabilir.
Open Access
Graph convolutional networks for region of interest classification in breast histopathology
(S P I E - International Society for Optical Engineering, 2021) Aygüneş, Bulut; Aksoy, Selim; Cinbiş, R.G.; Kösemehmetoğlu, K.; Önder, S.; Üner, A.
Deep learning-based approaches have shown highly successful performance in the categorization of digitized biopsy samples. The commonly used setting in these approaches is to employ convolutional neural networks for classification of data sets consisting of images all having the same size. However, the clinical practice in breast histopathology necessitates multi-class categorization of regions of interest (ROI) in biopsy samples where these regions can have arbitrary shapes and sizes. The typical solution to this problem is to aggregate the classification results of fixed-sized patches cropped from these images to obtain image-level classification scores. Another limitation of these approaches is the independent processing of individual patches where the rich contextual information in the complex tissue structures has not yet been sufficiently exploited. We propose a generic methodology to incorporate local inter-patch context through a graph convolution network (GCN) that admits a graph-based ROI representation. The proposed GCN model aims to propagate information over neighboring patches in a progressive manner towards classifying the whole ROI into a diagnostic class. The experiments using a challenging data set for a 4-class ROI-level classification task and comparisons with several baseline approaches show that the proposed model that incorporates the spatial context by using graph convolutional layers performs better than commonly used fusion rules.
Open Access
Mixture of learners for cancer stem cell detection using CD13 and H and E stained images
(SPIE, 2016) Oğuz, Oğuzhan; Akbaş, Cem Emre; Mallah, Maen; Taşdemir, K.; Akhan-Güzelcan, E.; Muenzenmayer, C.; Wittenberg, T.; Üner, A.; Çetin, A. Enis; Çetin-Atalay, R.
In this article, algorithms for cancer stem cell (CSC) detection in liver cancer tissue images are developed. Conventionally, a pathologist examines of cancer cell morphologies under microscope. Computer aided diagnosis systems (CAD) aims to help pathologists in this tedious and repetitive work. The first algorithm locates CSCs in CD13 stained liver tissue images. The method has also an online learning algorithm to improve the accuracy of detection. The second family of algorithms classify the cancer tissues stained with H and E which is clinically routine and cost effective than immunohistochemistry (IHC) procedure. The algorithms utilize 1D-SIFT and Eigen-Analysis based feature sets as descriptors. Normal and cancerous tissues can be classified with 92.1% accuracy in H and E stained images. Classification accuracy of low and high-grade cancerous tissue images is 70.4%. Therefore, this study paves the way for diagnosing the cancerous tissue and grading the level of it using H and E stained microscopic tissue images.
Open Access
Reactivation of cAMP pathway by PDE4D inhibition represents a novel druggable axis for overcoming tamoxifen resistance in ER-positive breast cancer
(American Association for Cancer Research, 2018) Mishra, Rasmi R.; Belder, Nevin; Ansari, Suhail A.; Kayhan, Merve; Bal, Hilal; Raza, Umar; Ersan, Pelin G.; Tokat, Ünal M.; Eyüpoğlu, Erol; Saatçi, Özge; Jandaghi, P.; Wiemann, S.; Üner, A.; Çekiç, Çağlar; Riazalhosseini, Y.; Şahin, Özgür
Purpose: Tamoxifen remains an important hormonal therapy for ER-positive breast cancer; however, development of resistance is a major obstacle in clinics. Here, we aimed to identify novel mechanisms of tamoxifen resistance and provide actionable drug targets overcoming resistance. Experimental Design: Whole-transcriptome sequencing, downstream pathway analysis, and drug repositioning approaches were used to identify novel modulators [here: phosphodiesterase 4D (PDE4D)] of tamoxifen resistance. Clinical data involving tamoxifen-treated patients with ER-positive breast cancer were used to assess the impact of PDE4D in tamoxifen resistance. Tamoxifen sensitization role of PDE4D was tested in vitro and in vivo. Cytobiology, biochemistry, and functional genomics tools were used to elucidate the mechanisms of PDE4D-mediated tamoxifen resistance. Results: PDE4D, which hydrolyzes cyclic AMP (cAMP), was significantly overexpressed in both MCF-7 and T47D tamoxifen-resistant (TamR) cells. Higher PDE4D expression predicted worse survival in tamoxifen-treated patients with breast cancer (n ¼ 469, P ¼ 0.0036 for DMFS; n ¼ 561, P ¼ 0.0229 for RFS) and remained an independent prognostic factor for RFS in multivariate analysis (n ¼ 132, P ¼ 0.049). Inhibition of PDE4D by either siRNAs or pharmacologic inhibitors (dipyridamole and Gebr-7b) restored tamoxifen sensitivity. Sensitization to tamoxifen is achieved via cAMP-mediated induction of unfolded protein response/ER stress pathway leading to activation of p38/JNK signaling and apoptosis. Remarkably, acetylsalicylic acid (aspirin) was predicted to be a tamoxifen sensitizer using a drug repositioning approach and was shown to reverse resistance by targeting PDE4D/ cAMP/ER stress axis. Finally, combining PDE4D inhibitors and tamoxifen suppressed tumor growth better than individual groups in vivo. Conclusions: PDE4D plays a pivotal role in acquired tamoxifen resistance via blocking cAMP/ER stress/p38-JNK signaling and apoptosis.
Open Access
Targeting lysyl oxidase (LOX) overcomes chemotherapy resistance in triple negative breast cancer
(Nature Research, 2020) Saatçi, Ö.; Kaymak, A.; Raza, Umar; Ersan, Pelin G.; Akbulut, Özge; Banister, C. E.; Sikirzhytski, V.; Tokat, Ünal Metin; Aykut, Gamze; Ansari, Suhail A.; Tatlı-Doğan, H.; Doğan, M.; Jandaghi, P.; Işık, A.; Gündoğdu, F.; Kösemehmetoğlu, K.; Dizdar, Ö.; Aksoy, S.; Akyol, A.; Üner, A.; Buckhaults, P. J.; Riazalhosseini, Y.; Şahin, Özgür
Chemoresistance is a major obstacle in triple negative breast cancer (TNBC), the most aggressive breast cancer subtype. Here we identify hypoxia-induced ECM re-modeler, lysyl oxidase (LOX) as a key inducer of chemoresistance by developing chemoresistant TNBC tumors in vivo and characterizing their transcriptomes by RNA-sequencing. Inhibiting LOX reduces collagen cross-linking and fibronectin assembly, increases drug penetration, and downregulates ITGA5/FN1 expression, resulting in inhibition of FAK/Src signaling, induction of apoptosis and re-sensitization to chemotherapy. Similarly, inhibiting FAK/Src results in chemosensitization. These effects are observed in 3D-cultured cell lines, tumor organoids, chemoresistant xenografts, syngeneic tumors and PDX models. Re-expressing the hypoxia-repressed miR-142-3p, which targets HIF1A, LOX and ITGA5, causes further suppression of the HIF-1α/LOX/ITGA5/FN1 axis. Notably, higher LOX, ITGA5, or FN1, or lower miR-142-3p levels are associated with shorter survival in chemotherapy-treated TNBC patients. These results provide strong pre-clinical rationale for developing and testing LOX inhibitors to overcome chemoresistance in TNBC patients.