To assess the presence of isocitrate dehydrogenase (IDH) 1 mutation in glioblastomas using real-time polymerase chain reaction (RT-PCR), which is the gold standard in the diagnosis of IDH1 mutation; by immunohistochemistry (IHC), which is available in most of the pathology laboratories; and by preoperative magnetic resonance imaging, which is a non-invasive method. We also investigated the relationship between these methods and their usability in routine practice. RT-PCR was performed to evaluate the presence of IDH1-R132H mutation on the blocks of 70 patients diagnosed with glioblastoma, and IDH1 stain was applied to the same blocks as IHC. Radiologically, preoperative magnetic resonance images of the patients were reviewed in terms of tumor size, localization, and presence of non-contrast-enhancing solid tumor component. Evaluation by RT-PCR revealed that 15 (21.4%) patients were IDH-mutant, whereas IHC examination revealed 13 (18.6%) and radiological evaluation revealed 11 (15.7%) patients were IDH-mutant. There was a statistically significant difference between the IDH1 mutation detected by RT-PCR and by IHC or radiological methods (p=0.034 and p=0.000, respectively). The sensitivity and specificity of IHC method in detecting IDH1 mutation were 86.6% and 100%, respectively, whereas those of radiological methods were 33.3% and 89%, respectively. Conclusively, radiological and IHC methods can be used in cases where RT-PCR cannot be applied for detecting IDH1 mutation. However, the results need to be confirmed by RT-PCR when necessary as these methods may sometimes overlook some IDH-mutant patients.

The aim of the systematic review was to assess recently published studies on diagnostic test accuracy of glioblastoma treatment response monitoring biomarkers in adults, developed through machine learning (ML). Articles published 09/2018–09/2020 were searched for using MEDLINE, EMBASE, and the Cochrane Register. Included study participants were adult patients with high grade glioma who had undergone standard treatment (maximal resection, radiotherapy with concomitant and adjuvant temozolomide) and subsequently underwent follow-up imaging to determine treatment response status (specifically, distinguishing progression/recurrence from progression/recurrence mimics - the target condition). Risk of bias and applicability was assessed with QUADAS 2 methodology. Contingency tables were created for hold-out test sets and recall, specificity, precision, F1-score, balanced accuracy calculated. Fifteen studies were included with 1038 patients in training sets and 233 in test sets. To determine whether there was progression or a mimic, the reference standard combination of follow-up imaging and histopathology at re-operation was applied in 67% (10/15) of studies. External hold-out test sets were used in 27% (4/15) to give ranges of diagnostic accuracy measures: recall = 0.70–1.00; specificity = 0.67–0.90; precision = 0.78–0.88; F1 score = 0.74–0.94; balanced accuracy = 0.74–0.83; AUC = 0.80–0.85. The small numbers of patient included in studies, the high risk of bias and concerns of applicability in the study designs (particularly in relation to the reference standard and patient selection due to confounding), and the low level of evidence, suggest that limited conclusions can be drawn from the data. There is likely good diagnostic performance of machine learning models that use MRI features to distinguish between progression and mimics. The diagnostic performance of ML using implicit features did not appear to be superior to ML using explicit features. There are a range of ML-based solutions poised to become treatment response monitoring biomarkers for glioblastoma. To achieve this, the development and validation of ML models require large, well-annotated datasets where the potential for confounding in the study design has been carefully considered. Therefore, multidisciplinary efforts and multicentre collaborations are necessary.

Background and Objective: Radiologically isolated syndrome (RIS) is characterized by magnetic resonance imaging (MRI) findings suggestive of multiple sclerosis (MS) in individuals without a history of clinical demyelinating events. Increasing evidence indicates that structural brain abnormalities may be present in RIS before the development of overt neurological symptoms. However, most previous studies have employed cross-sectional designs, limiting the ability to characterize the temporal evolution of neurodegenerative changes. This study aimed to investigate longitudinal changes in global, regional, and lobar brain volumes in individuals with RIS using serial MRI examinations obtained at least six months apart. Methods: Patients with RIS who had undergone two 3 Tesla (3T) brain MRI examinations separated by an interval of at least six months were retrospectively identified. Three-dimensional (3D) T1-weighted MRI scans obtained at both time points were analyzed using the volBrain pipeline of the volBrain platform to quantify global, regional, and lobar brain volumes. Longitudinal volumetric changes were expressed as annualized percentage volume changes. The number of T2 lesions and total T2 lesion volume were additionally assessed, and their associations with longitudinal brain volume changes were investigated. Results: A total of 28 patients with RIS were included. After correction for multiple comparisons using the Benjamini–Hochberg false discovery rate (FDR) procedure, significant annual volume loss was observed in total brain volume (p = 0.043), total cerebrum volume (p = 0.033), right cerebrum volume (p = 0.033), left cerebrum volume (p = 0.044), total thalamic volume (p = 0.033), right thalamic volume (p = 0.047), and left thalamic volume (p = 0.043). No significant associations were observed between annualized brain volume changes and follow-up duration, total or regional T2 lesion counts, or total T2 lesion volume. Similarly, annualized brain volume changes did not differ significantly between subgroups defined according to MRI progression, the presence of gadolinium-enhancing lesions, or the presence of spinal cord lesions. Conclusions: Longitudinal volumetric analysis demonstrated progressive brain volume loss involving both global and regional structures in patients with RIS. These findings provide further evidence of measurable longitudinal structural brain changes in RIS before the development of clinically overt demyelinating disease. Larger longitudinal studies incorporating appropriate control groups are warranted to further characterize the temporal trajectory of volumetric changes in RIS.

Amaç: Bu çalışmanın amacı meme kitlelerindeki vaskülariteyi değerlendirmek için, Renkli Doppler Görüntüleme (RDG), Power Doppler Görüntüleme (PDG) ve Süperb Mikrovasküler Görüntüleme (SMG) yaparak elde ettiğimiz bulguları, histopatolojik sonuçlar ile karşılaştırıp tanısal etkinliklerini değerlendirmektir. Yöntem: Belirtilen tarih aralığında, dahil etme kriterlerine uygun 46 hastaya gri skala USG, RDG, PDG ve SMG yapıldı. Her bir kitlenin vaskülarite özellikleri kaydedildi ve karşılaştırıldı. Bulgular: Kırk altı meme lezyonundan 26’sı benign ve 20’si maligndi. 26 benign lezyondan 19 tanesi fibroadenom, 5 tanesi granülomatöz mastit ve 2 tanesi sklerozan adenozis iken 20 malign lezyonun tamamı invaziv duktal karsinomdu. Benign lezyonlarda ortalama boyut 21,5± 9,6mm, malign lezyonlarda 12,4± 5,8mm idi. Renkli SMG (rSMG) ile malign lezyonlarda benign lezyonlardan daha fazla penetran damar (PV) tespit edilmiş olup malign lezyonlar için rSMG tarafınca tespit edilen PV sayısı, RDG ve PDG tarafınca tespit edilenden daha fazladır. Benign lezyonlarda rSMG ve RDG ile, PDG’ye kıyasla daha az PV tespit edildi. PV varlığı malignite tanısı için bir kriter olarak kullanıldığında, rSMG’nin duyarlılık, özgüllük, pozitif prediktif değer (PPD), negatif prediktif değer (NPD) ve tanısal doğruluğu, RDG ve PDG'ninkinden daha yüksekti. RDG ve PDG ile karşılaştırıldığında, rSMG ile daha fazla akım sinyali ve damar morfolojisi tespit edildi. Meme lezyonlarındaki vasküler paternlere göre farklı görüntüleme modalitelerinin tanısal etkinliğinde rSMG’nin duyarlılık, PPD, NPD ve tanısal doğruluğu, RDG ve PDG’ninkinden daha yüksekken, özgüllüğü daha düşüktü. Sonuç: rSMG, RDG ve PDG’den daha fazla vasküler dallanma detayı göstermekte olup benign-malign meme lezyonunun ayırt edilmesinde uygulanabilir bir yöntemdir.

Artificial intelligence (AI) is transforming oncologic imaging by enabling automated, accurate, and scalable diagnostic decision support. However, successful clinical translation requires not only strong predictive performance but also interpretability, transparency, and clinician confidence. We present an attention-guided deep learning framework for automated slice-level classification of parathyroid adenoma (PTA) from arterial-phase computed tomography (CT) images. The framework incorporates lightweight hierarchical attention mechanisms within residual neural networks to enhance feature representation and contextual understanding while maintaining computational efficiency for real-world deployment. Three novel architectures were developed: the Residual Block Light Attention Network (Res-BLANet), Residual Stage Light Attention Network (Res-SLANet), and Residual Layer Light Attention Network (Res-LLANet). Models were trained and evaluated on a rigorously curated, expert-annotated dataset of 63 patients, including 39 pathologically confirmed PTA cases, with 350–450 slices per patient. Training employed Hounsfield unit normalization, extensive data augmentation, and 10-fold cross-validation to ensure robust performance assessment. A key innovation is the integration of hierarchical attention modules that generate attention maps at multiple network levels, enabling qualitative visualization of diagnostically relevant regions and providing exploratory insight into the model’s decision-making process. Experimental evaluation showed strong diagnostic performance. Res-SLANet achieved the highest overall accuracy (87.37%), precision (92.56%), and F1-score (86.69%), while Res-LLANet attained the highest sensitivity (96.0%) on independent testing. These results should be interpreted as preliminary, given the limited size of the independent patient-level test cohort (n = 3). Res-BLANet delivered substantially faster inference with minimal computational overhead. These findings demonstrate that hierarchical attention-guided deep learning can achieve accurate and computationally efficient PTA detection from CT imaging while providing qualitative visual insights into the model’s decision-making process. The proposed framework represents a promising approach toward more interpretable AI-assisted diagnostic systems for oncologic imaging.