EKEla Kaplan

Ela Kaplan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Ela Kaplan has a total of 38 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Associations between pelvic MRI findings and symptom presence in uterine leiomyoma and adenomyosis: a quantitative analysis

    by Selçuk Kaplan, Ela Kaplan

    Objective: To evaluate the relationship between pelvic MRI findings and symptom presence in patients with leiomyoma and adenomyosis, and to determine whether these imaging parameters are independently associated with the presence of symptoms. Methods: This single-center retrospective study enrolled 150 patients who underwent pelvic MRI and were stratified into three groups based on pathological diagnosis: leiomyoma only ( n = 75), adenomyosis only ( n = 36), and concurrent pathology ( n = 39). Leiomyoma assessment included International Federation of Gynecology and Obstetrics (FIGO) classification, maximum diameter, cavity distortion, and apparent diffusion coefficient (ADC) values; for adenomyosis, junctional zone (JZ) thickness and involvement pattern served as primary parameters. Composite MRI burden scores were constructed for each pathology. Imaging features independently associated with symptom presence were identified by binary logistic regression, and model discrimination was assessed with internal validation by bootstrap resampling. Results: Symptom prevalence differed substantially across groups — 64.0% in the leiomyoma group, 72.2% in the adenomyosis group, and 97.4% in patients with concurrent pathology ( p < 0.001). Each 1 mm increase in JZ thickness raised the odds ratio (OR) of symptoms 1.31-fold in the adenomyosis group (OR = 1.31; 95% confidence interval [CI]: 1.08–1.58; p = 0.006), with JZ thickness correlating with pelvic pain at r = 0.61. FIGO type 0–2 location was the imaging feature most strongly associated with abnormal uterine bleeding (AUB) among leiomyoma patients (OR = 6.82; 95% CI: 2.87–16.21; p < 0.001). ADC values were markedly lower in symptomatic leiomyomas (0.91 vs. 1.14 × 10⁻³ mm²/s; p < 0.001). The apparent area under the curve (AUC) was 0.809 for the symptom model and 0.821 for the AUB model, with optimism-corrected values of 0.785 and 0.796, respectively, on bootstrap internal validation. Conclusion: Pelvic MRI findings in leiomyoma and adenomyosis are independently associated with the presence of symptoms, with JZ thickness, leiomyoma localization relative to the endometrial cavity, lesion size, and cavity distortion showing the strongest associations. Because these associations were derived and tested within a single cross-sectional sample, they should be regarded as hypothesis-generating and require prospective, externally validated confirmation before they can guide treatment decisions.

  • Open Access

    How Does ACR BI-RADS® v2025 Change the Radiologist’s Approach? A Practical Guide Across Mammography, Ultrasound, and MRI: A Narrative Review

    by Ela Kaplan, Ahmet Burak Aydemir

    Twelve years after the 2013 fifth edition, the American College of Radiology has released BI-RADS® v2025, with substantial revisions across mammography, ultrasonography, MRI, and the newly independent contrast-enhanced mammography (CEM) section. This review compares the two editions and reads the main changes against the available evidence on diagnostic performance and reader agreement, rather than only cataloguing them. Examples featuring digital breast tomosynthesis, synthetic mammography, and automated whole-breast ultrasonography now appear throughout the modality sections. “Lobulated” has been added as a shape descriptor across all modalities, while “microlobulated” was dropped from the mammography margin list. Calcification terms shifted from etiology to morphology: “milk of calcium” became “layering,” “punctate” was folded into “round,” and “dystrophic” moved under “coarse.” Ultrasonography gained two new entries, “non-mass lesion” and “echogenic rind,” and on MRI, “initial phase” was renamed “early phase.” Category 6 was rewritten; therefore, surgical excision is recognized as one of several definitive local treatments, and an “uncoupled” principle now separates assessment from management. The auditing section folds Category 3 follow-up into basic auditing and adds a Method of Detection data field. Most supporting data, however, predate v2025, and reader agreement remains lower for newer descriptors such as non-mass enhancement; whether the revisions measurably improve reproducibility is still unproven. Integrating every high-sensitivity tool also brings more false positives, overdiagnosis, and cost; artificial intelligence and radiomics may help close the reproducibility gap, but resource-stratified, equitable implementation will be essential. With v2025, BI-RADS becomes a multimodality framework rather than a reporting lexicon alone.

  • Open Access

    Diagnostic Accuracy of Post-Treatment ADC and MRI Response Assessment for Predicting Pathological Complete Response in Breast Cancer Patients Receiving Neoadjuvant Chemotherapy

    by Ela Kaplan, Hüseyin Alakus, Selcuk Kaplan

    Background/Objectives: This study aimed to evaluate the diagnostic performance of post-neoadjuvant chemotherapy (NAC) contrast-enhanced MRI in breast cancer treatment response assessment and to determine whether apparent diffusion coefficient (ADC) parameters contribute to predicting and confirming pathologic complete response (pCR). Methods: This retrospective cohort study enrolled patients with histopathologically confirmed invasive breast cancer who underwent breast MRI before and after NAC, followed by surgical resection. Post-NAC MRI response was classified into four categories and subsequently dichotomised into complete response versus residual disease. On diffusion-weighted imaging (DWI), pre-treatment minimum and maximum ADC values, post-treatment ADC, and the percentage change in ADC (ΔADC) were calculated. Diagnostic performance was assessed by sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), receiver operating characteristic (ROC) curve analysis, and Cohen’s kappa coefficient. Results: Of 188 patients, 19.7% achieved pCR. Post-NAC MRI complete response predicted pCR with 100% sensitivity and 90.1% specificity, with a Cohen’s kappa of 0.781 and 15 false-positive cases. Post-treatment ADC achieved the highest predictive performance, with an area under the ROC curve (AUC) of 0.967 (optimism-corrected 0.958); at the derived threshold, sensitivity was 100.0% and specificity 96.7%. ΔADC was likewise a statistically significant predictor of pCR, and post-treatment ADC remained independent after adjustment. Pre-treatment ADC parameters carried no meaningful predictive value in the full-cohort analysis. Conclusions: Post-treatment ADC and ΔADC help identify true complete responders among cases that contrast-enhanced MRI alone misclassifies as positive. Incorporating quantitative diffusion parameters into standard post-NAC MRI assessment may support pCR confirmation and warrants prospective external validation.

  • Open Access

    Morphologic and Diffusion-Weighted MRI Characteristics of Axillary Lymph Nodes for Predicting Metastasis in Breast Cancer Patients: A Quantitative Analysis

    by Mehmet Sirik, Melahat Poyraz, Ela Kaplan

    Introduction: Accurate preoperative assessment of axillary lymph node metastasis is essential for treatment planning in breast cancer patients. This study aimed to evaluate the diagnostic performance of quantitative MRI parameters, including lymph node morphology and diffusion-weighted imagingderived Apparent Diffusion Coefficient (ADC) values, in predicting axillary lymph node metastasis. Methods: This retrospective study included 216 female patients with histopathologically confirmed breast cancer who underwent preoperative breast MRI. Lymph node long and short axis diameters, cortical thickness, ADC values, and derived ratios were measured. Receiver Operating Characteristic (ROC) analysis and multivariate logistic regression were performed to identify independent predictors of lymph node metastasis. Results: Metastatic lymph nodes demonstrated significantly greater cortical thickness and lower ADC values compared with non-metastatic nodes (p < 0.001). Among all parameters, the ADC/cortical thickness ratio showed the highest diagnostic performance (AUC = 0.978; 95% CI: 0.959-0.997), with 96.2% sensitivity and 92.8% specificity. The ADC/cortical thickness ratio successfully detected 128 out of 133 metastatic lymph nodes and 77 out of 83 non-metastatic lymph nodes, with total accuracies of 96.2% and 92.8%, respectively. The test demonstrated a positive predictive value of 95.5% and a negative predictive value of 94.2%. Cortical thickness (>3.5 mm), ADC value (≤900 ×10−3 mm2/s), and Ki-67 index (≥20%) were identified as independent predictors of lymph node metastasis. Discussion: The present study demonstrates that lymph node ADC values and ADC-based ratios provide significant diagnostic performance in differentiating metastatic from non-metastatic axillary lymph nodes. The identified cut-off values yielded clinically meaningful sensitivity and specificity, supporting the role of diffusion-weighted imaging as a non-invasive adjunct tool in preoperative axillary evaluation. These findings may contribute to improved patient stratification and may help reduce unnecessary invasive procedures in selected cases. Conclusion: Quantitative MRI parameters combining morphologic and diffusion features demonstrate high diagnostic accuracy for predicting axillary lymph node metastasis in breast cancer patients. The ADC/cortical thickness ratio is a simple, reproducible imaging biomarker that may support preoperative axillary evaluation and clinical decision-making in routine breast MRI practice.

  • Open Access

    Pre-Treatment Breast MRI Features and ADC Values as Predictors of Pathologic Complete Response in Breast Cancer: A Molecular Subtype-Based Analysis

    by Ela Kaplan, Hüseyin Alakus, Selcuk Kaplan

    Background/Objectives: The role of pre-treatment breast magnetic resonance imaging (MRI) findings and apparent diffusion coefficient (ADC) values in predicting pathologic complete response (pCR) in breast cancer patients receiving neoadjuvant chemotherapy (NAC) has not yet been sufficiently clarified, especially in the context of molecular subtype differences. In this study, we questioned whether these imaging parameters were independent predictors of pCR. Methods: This study retrospectively explored MRI characteristics of 188 patients who underwent NAC from 2015 to 2023. The patients were divided into the pCR-positive and pCR-negative groups—the latter comprising patients with partial response (n = 61) and stable disease (n = 90)—and were classified into four molecular subtypes: Luminal A/B, HER2-enriched, and triple-negative breast cancer (TNBC). The MRI parameters included pre-chemotherapy T2-weighted signal characteristics, shape features, contrast kinetics, peritumoral edema, and ADC MIN/ADC MAX. Post-treatment ADC and ΔADC were the post-chemotherapy MRI parameters. Independent predictors were evaluated by logistic regression and discriminant performance by ROC analysis. Results: The overall pCR rate was 19.7%. In multivariate analysis, T2-weighted isointense signal (OR = 4.50), uniform tumor shape (OR = 12.83), HER2-enriched subtype (OR = 6.03), TNBC (OR = 5.15), ADC MIN (OR = 1.41), tumor size (OR = 1.28), and kinetic Type 3 pattern (OR = 3.21) were identified as independent predictors. Pre-treatment ADC MIN yielded an AUC of 0.724, while post-treatment ADC achieved 100% sensitivity and 96.7% specificity (AUC = 0.967). Conclusions: MRI morphology and ADC values may make a meaningful contribution to the prediction of pCR when evaluated in the context of molecular subtype. Post-treatment ADC demonstrated particularly strong discriminatory performance; however, external validation in multicenter cohorts is required before clinical implementation.

  • Open Access

    Differentiating types of breast cancer from digital mammography images with artificial intelligence methods

    by Ela Kaplan, Orhan Yaman, Hacı Taner Bulut, Mehmet Şirik, Türker Tuncer, Şengül Doğan

    Objectives: Breast cancer (BCA) is one of the world’s most prevalent cancer and the top cause of mortality. For many decades, mammography has been used routinely for screening of early breast cancer and diagnosing symptomatic patients. The main purpose of this work is to investigate the usefulness of machine learning techniques using mammography images. Methods: A total of 194 patients who underwent ultrasound examination after observing suspicious lesions on mammography images and were diagnosed with BCA by ultrasound-guided core needle biopsy were included in the study. A set of mammography images with complete cancer subtypes was used. A transfer learning-based computer vision method was adopted in this study. AlexNet was to extract the features and select the most significant features using a feature selection function. Our deep learning-based model attained more than 80% accuracy in classifying malignant and benign cancers. However, the employed deep learning model cannot classify subtypes accurately. Results: Per the results, the commonly used image classification model is highly accurate in distinguishing malignant and benign changes, however unable to classify cancer subtypes. Conclusions: In conclusion, machine learning can still not simulate conventional immunohistochemistry subtyping using tissue biopsy.

  • Open Access

    Güncel Radyoloji Çalışmaları IV

    by İhSan Kaplan, Sema Kaya, Burçin Avcı, Murat Kılıç, İsmail Faruk Durmuş, Ela Kaplan, M. Fatih Erbay, Ayşegül Sağır Kahraman, Müge Otlu Karadağ, Ersoy Kekilli, et al.

    Peptit Reseptör Radyonüklid Tedavi İhsan KAPLAN Çocuklarda Konik Işınlı Bilgisayarlı Tomografi Kullanım Endikasyonları Sema KAYA Burçin AVCI Radyodirençlilikte GSTP-1 İzoenziminin Olası Rolü: Radyoterapiyle İndüklenen Ros Artışına Yanıt Murat KILIÇ İsmail Faruk DURMUŞ İleri Evre Hepatosellüler Karsinom Tedavisinde Transarteryel Radyoembolizasyon Sonrası Bilgisayarlı Tomografi Perfüzyon Görüntüleme Parametreleri Değişimi Ela KAPLAN Fatih ERBAY Ayşegül SAĞIR KAHRAMAN Müge OTLU KARADAĞ Ersoy KEKİLLİ Ramazan KUTLU Covid-19 Kardiyak Tutulumunda Manyetik Rezonans Görüntüleme Safa ÖZYILMAZ Temporomandibular Eklem Anatomisi ve Görüntüleme Yöntemleri Menduh Sercan KAYA Mehmet Emin DOĞAN Pediatrik Hastalarda Diz Görüntüleme Eren ÇAMUR

  • Open Access

    Is there a relationship between vascular flow parameters of carotid arterial system and trigeminal neuralgia?

    by Ela Kaplan, Mehtap Koparal

    Aim: To investigate the flow velocity values and resistance parameters determined by carotid artery doppler ultrasonography (DUS) examination in patients with trigeminal neuralgia (TN). Materials and Methods: 79 patients diagnosed with TN and who underwent carotid DUS were included in the study. Spectral examination results were recorded for the TN side and the healthy side. Results: Peak systolic velocity (PSV) internal carotid artery (ICA), resistive index (RI) (ICA) and PSV (ICA)/PSV external carotid (ECA) values in patients were relatively higher in arterial structures on the TN clinical side compared to the healthy side. RI (ICA)/RI common carotid artery (CCA) value was also high in arterial structures on the TN diagnosed side. Conclusion: It was shown that especially ICA flow velocity and vascular resistance parameters increased in the carotid arterial system on the TN side. These findings indicate that flow parameters in the carotid artery system may also be effective in TN pathophysiology.

  • PFP-HOG: Pyramid and Fixed-Size Patch-Based HOG Technique for Automated Brain Abnormality Classification with MRI

    by Ela Kaplan, Wai Yee Chan, Hasan Baki Altinsoy, Mehmet Baygin, Prabal Datta Barua, Subrata Chakraborty, Sengul Dogan, Turker Tuncer, U. Rajendra Acharya

    Detecting neurological abnormalities such as brain tumors and Alzheimer’s disease (AD) using magnetic resonance imaging (MRI) images is an important research topic in the literature. Numerous machine learning models have been used to detect brain abnormalities accurately. This study addresses the problem of detecting neurological abnormalities in MRI. The motivation behind this problem lies in the need for accurate and efficient methods to assist neurologists in the diagnosis of these disorders. In addition, many deep learning techniques have been applied to MRI to develop accurate brain abnormality detection models, but these networks have high time complexity. Hence, a novel hand-modeled feature-based learning network is presented to reduce the time complexity and obtain high classification performance. The model proposed in this work uses a new feature generation architecture named pyramid and fixed-size patch (PFP). The main aim of the proposed PFP structure is to attain high classification performance using essential feature extractors with both multilevel and local features. Furthermore, the PFP feature extractor generates low- and high-level features using a handcrafted extractor. To obtain the high discriminative feature extraction ability of the PFP, we have used histogram-oriented gradients (HOG); hence, it is named PFP-HOG. Furthermore, the iterative Chi2 (IChi2) is utilized to choose the clinically significant features. Finally, the k-nearest neighbors (kNN) with tenfold cross-validation is used for automated classification. Four MRI neurological databases (AD dataset, brain tumor dataset 1, brain tumor dataset 2, and merged dataset) have been utilized to develop our model. PFP-HOG and IChi2-based models attained 100%, 94.98%, 98.19%, and 97.80% using the AD dataset, brain tumor dataset1, brain tumor dataset 2, and merged brain MRI dataset, respectively. These findings not only provide an accurate and robust classification of various neurological disorders using MRI but also hold the potential to assist neurologists in validating manual MRI brain abnormality screening.