Brain tumors are a primary factor causing cancer-related deaths globally, and their classification remains a significant research challenge due to the variability in tumor intensity, size, and shape, as well as the similar appearances of different tumor types. Accurate differentiation is further complicated by these factors, making diagnosis difficult even with advanced imaging techniques such as magnetic resonance imaging (MRI). Recent techniques in artificial intelligence (AI), in particular deep learning (DL), have improved the speed and accuracy of medical image analysis, but they still face challenges like overfitting and the need for large annotated datasets. This study addresses these challenges by presenting two approaches for brain tumor classification using MRI images. The first approach involves fine-tuning transfer learning cutting-edge models, including SEResNet, ConvNeXtBase, and ResNet101V2, with global average pooling 2D and dropout layers to minimize overfitting and reduce the need for extensive preprocessing. The second approach leverages the Vision Transformer (ViT), optimized with the AdamW optimizer and extensive data augmentation. Experiments on the BT-Large-4C dataset demonstrate that SEResNet achieves the highest accuracy of 97.96%, surpassing ViT’s 95.4%. These results suggest that fine-tuning and transfer learning models are more effective at addressing the challenges of overfitting and dataset limitations, ultimately outperforming the Vision Transformer and existing state-of-the-art techniques in brain tumor classification.

Objective: To describe magnetic resonance imaging (MRI) findings of the brain in patients younger than 65 years who were studied by transcranial Doppler (TCD) with microbubble contrast, with a history of cryptogenic cerebrovascular accident (CVA) and suspected patent foramen ovale (PFO).

Materials and methods: This retrospective cross-sectional study included patients of both sexes, younger than 65 years of age.

Results: Our sample (n = 47.47% male and 53% female, mean age is 42 years) presented high-intensity transient signals (HITS) positive in 61.7% and HITS-negative in 38.3%. In HITS-positive patients, lesions at the level of the subcortical U-brains, single or multiple with bilaterally symmetrical distribution, predominated. In patients with moderate HITS, lesions in the vascular territory of the posterior circulation predominated.

Conclusion: In patients younger than 65 years with cryptogenic stroke and subcortical, single or multiple U-shaped lesions with bilateral and symmetrical distribution, a PFO should be considered as a possible cause of these lesions.

A systemic and synthetic review of the anatomy of the temporomandibular joint in magnetic resonance imaging was developed for its evaluation. The temporomandibular joint is an anatomical structure composed of bones, muscles, ligaments and an articular disc that allows important physiological movements, such as mandibular opening, closing, protrusion, retrusion and lateralization. Magnetic resonance imaging is an imaging technique that does not use ionizing radiation and is more specific for the evaluation and interpretation of soft tissues, due to its high resolution, so it has an important role in the diagnosis of various maxillofacial pathologies, which is why the dentist should have knowledge of the structures and functions of the temporomandibular joint through magnetic resonance imaging. The review demonstrates the importance of magnetic resonance imaging in the study of the anatomy of the temporomandibular joint, in addition to mentioning the advantages provided by this imaging technique such as its good detail of the soft tissues in its different sequences and the non-use of ionizing radiation to obtain its images.

Objective: To evaluate the imaging features of spondyloarthritis on magnetic resonance imaging (MRI) of the sacroiliac (SI) joints in terms of topography (in thirds) and affected margin, since this aspect is rarely addressed in the literature. Methods: Cross-sectional study with MRI (1.5 T) evaluation of the SI in 16 patients with diagnosis of axial spondyloarthritis regarding the presence of acute (subchondral bone edema, enthesitis, synovitis and capsulitis) and chronic changes (erosions, subchondral bone sclerosis, bone bridging and fatty replacement), performed by two radiologists, blinded to clinical data. MRI findings were correlated with clinical data including age, disease duration, medications, HLA-B27, BASDAI, ASDAS-VHS and ASDAS-PCR, BASMI, BASFI, and mSASSS. Results: Bone edema pattern and erosions showed predominance in the upper third of SI (p = 0.050, p = 0.0014, respectively). There was a correlation between the time of disease and structural changes by affected third (p = 0.028-0.037), as well as the presence of bone bridges with BASMI (p = 0.028) and mSASSS (p = 0.014). Patients with osteitis of the lower third had higher ASDAS values (ESRV: p = 0.011 and CRP: p = 0.017). Conclusion: Chronic inflammatory changes and the pattern of bone edema predominated in the upper third of the SI, but there was also concomitant involvement of the middle or lower thirds of the joint. The localization of involvement in the upper third of the SI was insufficient to differentiate between degeneration and inflammation.

Colorectal cancer is the fourth leading cause of death worldwide and the fifth leading cause of cancer death in Colombia. Magnetic resonance imaging is the ideal modality for the evaluation of colorectal cancer, since it allows staging by determining invasion beyond the muscularis propria, extension towards adjacent organs, identification of patients who are candidates for chemotherapy or pre-surgical radiotherapy and planning of the surgical procedure. The key point is based on the differentiation between T2 and T3 stages through the use of sequences with high-resolution T2 information. In addition to this, it allows the assessment of the size and morphology of the lymph nodes, and considerably increases the specificity for the detection of lymph node involvement. MRI is a technique with high specificity and high reproducibility.

The human brain has been described as a complex system. Its study by means of neurophysiological signals has revealed the presence of linear and nonlinear interactions. In this context, entropy metrics have been used to uncover brain behavior in the presence and absence of neurological disturbances. Entropy mapping is of great interest for the study of progressive neurodegenerative diseases such as Alzheimer’s disease. The aim of this study was to characterize the dynamics of brain oscillations in such disease by means of entropy and amplitude of low frequency oscillations from Bold signals of the default network and the executive control network in Alzheimer’s patients and healthy individuals, using a database extracted from the Open Access Imaging Studies series. The results revealed higher discriminative power of entropy by permutations compared to low-frequency fluctuation amplitude and fractional amplitude of low-frequency fluctuations. Increased entropy by permutations was obtained in regions of the default network and the executive control network in patients. The posterior cingulate cortex and the precuneus showed differential characteristics when assessing entropy by permutations in both groups. There were no findings when correlating metrics with clinical scales. The results demonstrated that entropy by permutations allows characterizing brain function in Alzheimer’s patients, and also reveals information about nonlinear interactions complementary to the characteristics obtained by calculating the amplitude of low frequency oscillations.