Bildiri Koleksiyonu

Bu koleksiyon için kalıcı URI

Güncel Gönderiler

Listeleniyor 1 - 20 / 20
  • Öğe
    Generatect: text-conditional generation of 3d chest ct volumes
    (2025) Hamamcı, İbrahim Ethem; Er, Sezgin; Tezcan, Alperen; Şimşek, Ayşe Gülnihan; Esirgün, Şevval Nil; Almas, Furkan; Doğan, İrem; Daşdelen, Muhammed Furkan; Menze, Bjoern
    Text-conditional medical image generation is vital for radiology, augmenting small datasets, preserving data privacy, and enabling patient-specific modeling. However, its applications in 3D medical imaging, such as CT and MRI, which are crucial for critical care, remain unexplored. In this paper, we introduce GenerateCT, the first approach to generating 3D medical imaging conditioned on free-form medical text prompts. GenerateCT incorporates a text encoder and three key components: a novel causal vision transformer for encoding 3D CT volumes, a text-image transformer for aligning CT and text tokens, and a text-conditional super-resolution diffusion model. Without directly comparable methods in 3D medical imaging, we benchmarked GenerateCT against cutting-edge methods, demonstrating its superiority across all key metrics. Importantly, we explored GenerateCT’s clinical applications by evaluating its utility in a multi-abnormality classification task. First, we established a baseline by training a multi-abnormality classifier on our real dataset. To further assess the model’s generalization to external datasets and its performance with unseen prompts in a zero-shot scenario, we employed an external dataset to train the classifier, setting an additional benchmark. We conducted two experiments in which we doubled the training datasets by synthesizing an equal number of volumes for each set using GenerateCT. The first experiment demonstrated an 11% improvement in the AP score when training the classifier jointly on real and generated volumes. The second experiment showed a 7% improvement when training on both real and generated volumes based on unseen prompts. Moreover, GenerateCT enables the scaling of synthetic training datasets to arbitrary sizes. As an example, we generated 100,000 3D CT volumes, fivefold the number in our real dataset, and trained the classifier exclusively on these synthetic volumes. Impressively, this classifier surpassed the performance of the one trained on all available real data by a margin of 8%. Lastly, domain experts evaluated the generated volumes, confirming a high degree of alignment with the text prompts. Access our code, model weights, training data, and generated data at https://github.com/ibrahimethemhamamci/GenerateCT.
  • Öğe
    Ct2rep: automated radiology report generation for 3d medical imaging
    (2024) Hamamcı, İbrahim Ethem; Er, Sezgin; Menze, Bjoern
    Medical imaging plays a crucial role in diagnosis, with radiology reports serving as vital documentation. Automating report generation has emerged as a critical need to alleviate the workload of radiologists. While machine learning has facilitated report generation for 2D medical imaging, extending this to 3D has been unexplored due to computational complexity and data scarcity. We introduce the first method to generate radiology reports for 3D medical imaging, specifically targeting chest CT volumes. Given the absence of comparable methods, we establish a baseline using an advanced 3D vision encoder in medical imaging to demonstrate our method’s effectiveness, which leverages a novel auto-regressive causal transformer. Furthermore, recognizing the benefits of leveraging information from previous visits, we augment CT2Rep with a cross-attention-based multi-modal fusion module and hierarchical memory, enabling the incorporation of longitudinal multimodal data. Access our code at https://github.com/ibrahimethemhamamci/CT2Rep.
  • Öğe
    Redox and lipidomic analysis of human brain microvascular endothelial cells (hcmec/d3) adapted to physiological oxygen levels
    (Elsevier Science Inc, 2024) Yang, Fan; Simon, Daniel; Eroğlu, Emrah; Aarsland, Dag; Siow, Richard C. M.; Mann, Giovanni E.
    The majority of studies with endothelial and other cell types are conducted in standard culture incubators gassed with 5% CO2 and room air (18 kPa O2), well known to induce oxidative stress.
  • Öğe
    Pioneering precision: unraveling cellular bioenergetics with ph-control
    (2024) Ghaffari Zaki, Asal; Miri, Seyed Mohammad; Çimen, Şeyma; Akgül Çağlar, Tuba; Yiğit, Esra Nur; Aydın, Mehmet Şerif; Öztürk, Gürkan; Eroğlu, Emrah
    [Abstract Not Available]
  • Öğe
    Development of a non-animal testing method using genetically engineered biosensors for assessing skin sensitizers
    (2024) Seçilmiş, Melike; Katı, Ahmet; Eroğlu, Emrah
    [Abstract Not Available]
  • Öğe
    Hyper7 biosensor associate with proteomic shifts related to oxidative stress in endothelial cells
    (2024) Barakat, Sarah; Çimen, Şeyma; Miri, Seyed Mohammad; Vatandaşlar, Emre; Kök, Kıvanç; Öztürk, Gürkan; Eroğlu, Emrah
    [Abstract Not Available]
  • Öğe
    Revisiting the role of h2o2 in nrf2 activation using an oxygen-independent biosensor
    (2024) Miri, Seyed Mohammad; Eroğlu, Emrah
    [Abstract Not Available]
  • Öğe
    The relationship between mitochondrial oxidative stress and huntington's disease progression
    (2024) Yiğit, Esra Nur; Aydın, Mehmet Şerif; Eroğlu, Emrah; Öztürk, Gürkan
    [Abstract Not Available]
  • Öğe
    Revisiting the role of h2o2 in nrf2 activation using an oxygenindependent biosensor
    (2024) Miri, Seyed Mohammad; Eroğlu, Emrah
    [Abstract Not Available]
  • Öğe
    Decoding the implications of subcellular h2o2 dynamics in a stroke-on-a-dish model using cerebral microvascular endothelial cells
    (2024) Kalkan Yazıcı, Merve; Kekeç, Enes Görkem; Üvak, Melike Tuğba; Eroğlu, Emrah
    [Abstract Not Available]
  • Öğe
    Smart polymeric nanocarriers for miRNA delivery against triple negative breast cancer
    (2024) Yumuk, Kübra; Hamurcu, Zuhal; Ercan Ayra, Merve; Yüksel Durmaz, Yasemin; Aydın, Ömer
    The most typical malignancy in women is breast cancer. This translates to almost 2.3 million women who have been diagnosed with the illness, which, according to the World Health Organization (WHO), resulted in more than 685,000 fatalities globally in 2020. Additionally, 8 million women have received a breast cancer diagnosis in the past five years, making breast cancer the most prevalent cancer worldwide [1]. Additionally, triple-negative breast cancer (TNBC) is a subtype of breast cancer that lacks the human epidermal growth factor receptor 2, progesterone receptor, and oestrogen receptor. Furthermore, TNBC is known for its drug resistance, aggressiveness, and metastatic properties [2]. Even though radiotherapy and chemotherapy are two therapeutic modalities for the treatment of breast cancer, they have had limited effectiveness in clinics due to a number of limitations, such as the emergence of drug resistance throughout therapy. Gene therapy offers a promising avenue to overcome these challenges. MicroRNAs (miRNAs) are significant factors in the development, metastasis, and advancement of breast cancer [3]. The oncogenic transcription factor Forkhead Box M1 (FOXM1) is engaged in processes that are thought to be cancer hallmarks [4]. In a previous study Zuhal Hamurcu et.al claimed that the results of the miRNA expression profile in two distinct TNBC cells, miRNA was downregulated after FOXM1 was knocked down. Also, they conducted KEGG pathway analysis and GO enrichment analysis for miRNA, and these analyses revealed that this miRNA is connected to the cell cycle, AMPK, p53, and NF-kB signalling pathways, as well as the formation and progression of cancer [5]. However, a few barriers, including miRNA instability, unfavourable offtarget effects, and non-specific activation of the immune system's natural defences, prevent RNAi-based techniques from operating to their full potential. Delivery of miRNA into cells is further restricted by endothelial cell resistance, inability to achieve endosomal escape, which prevents the miRNA from reaching the location of cytosolic action. The development of nanocarriers addresses almost all these problems. To make complex and transport miRNA to breast cancer cells, we developed “smart” polymeric nanocarriers. Also, smart polymeric nanoparticles provide a protective shield, shielding miRNAs from enzymatic degradation and enhancing their bioavailability. Inhibitor miRNA and smart nanocarriers are complex at a 2/1 N/P ratio. Additionally, mimic miRNA and negative miRNA form an 8/1 complex with smart nanocarriers. Finally, with a 16/1 N/P ratio, FAM-labelled negative miRNA and smart nanocarriers are complex. Furthermore, the percentage of cells encapsulating miRNAs in cells analysed using flow cytometry was determined. Accordingly, the import percentages of BT-549 and MDA-MB-231 cells are over 35% and 55%, respectively. The fluorescence microscopy results obtained show that our FAM-labelled miRNAs reach the inside of the cell within 4 hours. As a result of this experiment, it was observed that miRNA could be transported to the cell without any enzymatic digestion. Our results demonstrated the potential of polymeric miRNA delivery systems showcasing promising outcomes in preclinical for triple-negative breast cancer treatment. These advances hold immense promise in improving the precision, efficacy, and safety of breast cancer treatments. To enable efficient delivery and controlled release of miRNA, the physicochemical characteristics of the smart nanocarriers, including size, charge on the surface, and stability, are optimized.
  • Öğe
    Importance of physiological oxygen levels for endothelial redox signaling using Hyper 7
    (Elsevier Science Inc, 2023) Altun, Hamza Yusuf; Yang, Fan; Mann, Giovanni; Eroğlu, Emrah
    Cellular redox status is tightly linked to ambient oxygen levels. Endothelial cells (EC) in vivo are exposed to O2 levels between 3- 13 kPa, yet the majority in vitro studies maintain EC under atmospheric O2 levels (~18 O2). Under 18 kPa O2, EC are exposed to hyperoxia, leading to persistent upregulation of NRF2 regulated antioxidant defences. Exposure of EC to physiological O2 levels for at least 5 days attenuates NRF2-regulated antioxidant gene expression in the absence of HIF-1a stabilisation.
  • Öğe
    Reactivity of posterior cortical electroencephalographic alpha rythms during eyes opening in cognitively intact older adults and patients with dementia due to Alzheimer's and lewy diseases
    (Elsevier, 2023) Lopez, S.; Babiloni, C.; Noce, G.; Tucci, F.; Lizio, R.; Ferri, R.; Soricelli, A.; del Percio, C.
    Introduction: Patients with Lewy Bodies (DLB) and Parkinson’s disease dementia (PDD) suffer from vigilance dysregulation reflected by hypersomnolence, cognitive fluctuations, hallucinations, and sleep disorders. Therefore, we hypothesized that such dysregulation may be reflected by altered reactivity of posterior resting-state electroencephalographic (rsEEG) alpha rhythms during the vigilance transition from eyes-closed to -open condition. More specifically, we tested whether the above mentioned reactivity of posterior rsEEG alpha rhythms may differ in patients with DLB, PDD and Alzheimer’s disease (ADD) dementia, as a functional probe of the dominant cortical neural synchronization mechanisms regulating the vigilance in posterior visual systems. Methods: In two different studies, we extracted clinical, demographical, and rsEEG datasets in (1) older adults (Healthy, N=28) seniors as well as matched DLB (N=42) and ADD (N=48) patients, and (2) 25 Healthy seniors as well as matched PDD (N=73) and ADD (N=35) patients, from an international Eurasian database. The eLORETA freeware was used to estimate cortical rsEEG sources at delta, theta, alpha, beta, and gamma frequency bands during eye-closed and eye-open conditions.
  • Öğe
    Resting state electroencephalographic alpha rhythms predict and are sensitive to Alzheimer's disease mild cognitive impairment progression at a 6-month follow-up
    (Elsevier, 2023) Babiloni, C.; Noce, G.; del Percio, Claudio; Tucci, F.; Lopez, S.; Ferri, R.; Soricelli, A.; Arnaldi, D.; Giubilei, F.; Stocchi, F.; Güntekin, Bahar; Yener, Görsev; Taylor, J. P.; Pantano, P.; Piervincenzi, Claudia; Nikolaos, Petsas; Frisoni, Giovanni B.; Jakhar, Deepak
    Introduction: Patients with Alzheimer’s disease and mild cognitive impairment (ADMCI) suffer not only from severe cognitive deficits and disabilities in daily living but also from vigilance dysregulation reflected by substantial hypersomnolence due to vigilance fluctuations. Therefore, we hypothesized that such vigilance dysregulation in ADMCI patients might be reflected by altered electroencephalographic (rsEEG) rhythms recorded in a resting-state eyes-closed condition in relation to the disease progression at 6-month follow-ups.
  • Öğe
    Event-related brain oscillatory responses topographically differentiated between visual and auditory emotional stimuli
    (Elsevier, 2023) Dalmızrak, Esra; Güntekin, Bahar
    Previous EEG studies using emotional stimuli have successfully demonstrated that EEG-brain oscillation responses are sensitive to the valance and arousal of the stimulus. While the valance is represented by increased beta and gamma responses, arousal is represented by increased delta and theta responses. Although the event-related oscillations during visual-emotional paradigms are relatively well-established, the differentiation between visual and auditory emotional stimulation is not fully understood yet.
  • Öğe
    Chemogenetic tools and genetic biosensors in redox biology: Probing Hyper7 signals in specific cellular locales
    (Elsevier Science Inc., 2023) Altun, Hamza Yusuf; Erdoğan, Yusuf Ceyhun; Seçilmiş, Melike; Eroğlu, Emrah
    H2O2 is one of the most studied reactive oxygen species (ROS) in redox biology because of its critical role in oxidant-dependent signaling pathways. Conventional approaches apply exogenous H2O2 on cells and tissues, which barely recapitulates intracellular redox signaling events. An alternative approach is D-amino acid oxidase (DAAO)-based chemogenetic tools that permit intracellular H2O2 generation by D-amino acid catalysis. Exploiting the targetability of the DAAOs to cellular ultralocales allow precise manipulation of the redox tone with high spatial and temporal resolution. DAAOs have been utilized in several redox studies, yet due to the incomplete characterization and the lack of fine-tuned protocol on how to employ DAAOs, the potential of this multiparametric imaging technology could not be fully exploited. Using the ultra-sensitive genetically encoded H2O2 biosensor (HyPer), we tested the performance of a modified version of DAAO (mDAAO) under hypoxic and hyperoxic conditions. Our multiparametric imaging approach unveiled that cell type, cellular locales, pericellular oxygen concentrations, D-amino acid type, and its concentrations are critical and modulatable factors that permit fine-tuned H2O2 generation with high precision. These findings and fine-tuned protocols provide practical approaches for redox scientists to spatiotemporally control oxidative stress-dependent signaling pathways in different cultured cell model systems.
  • Öğe
    An integrated top-stage incubator and lens free holographic imaging system for culture monitoring applications
    (SPIE, 2022) Özbilgin, Zeynep; Akşit, Sevdenur; Toy, Muhammed Fatih
    Lens free inline holographic microscopy was shown to be a potent approach for many applications relying on cellular imaging. Applications requiring a large field of view at moderate resolution are the ones that are most suitable for this platform. Besides, the simplicity of the overall imaging system, which requires only a light source and a camera, positions this approach as an easily accessible one. Acquired holograms from such a system are processed to recover phase images. As an additional advantage on top of the simplicity, phase imaging enables the imaging of otherwise transparent cell samples without any need for labeling or staining. Eventually, such a system can be used for long term imaging of live cell cultures with wide field of view. Up to now, two alternatives were explored for the imaging of live cell cultures for extended duration. In one approach, a portable imaging system was placed inside a standard incubator with cell cultures on top. A stage top incubator was used on a modified microscope in the other approach. In both approaches, the cost of the system grows due to commercial systems, and the overall footprint of the system with incubator is too large to be classified as portable. Here, an integrated portable system is presented that can maintain cell cultures at desired temperature in a 3D printed enclosure while imaging them in lens free inline holographic microscopy modality. Such a system is well suited for tissue culturing and monitoring at limited resources settings.
  • Öğe
    Maskless lithography with holographic feedback for the fabrication of optical elements
    (SPIE, 2022) Gürcan, Tolga; Toy, Muhammed Fatih
    Photolithography has become a powerful tool in the fabrication of micro-optical elements following the advancements in grayscale approaches. However, hitting tight design tolerance goals require precise control of all parameters such as temperature, resist nonlinearity or preventing vignetting. In this work we took an alternative route to these problems by combining maskless lithography with digital holography. Addition of digital holography enables the use of feedback by measuring the quantitative phase of specimen near real time and in situ nondestructively. After each near UV exposure, phase retardation map of exposed photopolymer is measured with digital holography part of the system. Any deviation from target phase is corrected by changing the pattern displayed on the mask. We showed that the proposed method reduces the standard deviation of resulting phase compared to traditional one-shot grayscale lithography. It also does not require any precalibration of photoresist and relaxes the constraints for uniform UV illumination in sample plane.
  • Öğe
    Dual function of HSP70 in cytoprotection of tumor cells and generation of immunosuppressive tumor microenvironment
    (American Association for Cancer Research, 2021) Korkaya, Hasan; Köksalar Alkan, Fulya; Ece Arslan, Z.; Altıntaş, Esra; Piranlıoğlu, Raziye; Lee, Eunmi; Wicha, Max S.
    [Abstract Not Available]
  • Öğe
    Prognostic significance of SERPINB1 expression in gliomas
    (Wiley, 2021) Uludağ, D.; Karaoğlan, B. S.; Genç, B.; Tahta, A.; Mohr, T.; Çakır, A.; Mayliba, H. B.; Karakaş, N.
    Glioblastoma (GB) is the most prevalent and aggressive brain tumor with high morbidity. Determination of tumor initiative and prognostic factors are ultimately needed for therapy options. The effects of Serine Protease Inhibitor (SERPIN) B1 on tumor progression have been reprted in several cancer types including breast and lung cancers. However, the relation between SERPINB1 expression levels and glioma progression is still to be elucidated.