Objective: Standardizing image acquisition protocols and image quality across cameras is an important need in imaging, in particular in multi-center clinical trials and the use of image analysis and machine learning algorithms. The objective of this study was to examine the effect of ordered subset expectation maximization (OSEM) reconstruction parameters on the quantitative image quality of cardiac perfusion SPECT images in different typical SPECT cameras and therefore assess the need to change the parameter values across cameras. Methods: The analysis was carried out by comparing the defect contrast-to-noise ratio (CNR) at 12 OSEM subset-iteration combinations. Eight frames were reconstructed using the SIMIND Monte Carlo Simulation package. An activity of 370 MBq (10mCi) and projection acquisition interval of 20 seconds per projection were used. Attenuation (AC) and scatter corrections (SC) were performed in this study for all images. Results: The 16-2 subset-iteration combination yielded the highest CNR and defect contrast values for both cameras. The difference between CNR values for two cameras was found to be close to 5%. Conclusions: Monte Carlo simulations can be useful to investigate how quantitative image quality behaves with respect to reconstruction parameters and correction algorithms in a controlled environment. In this study, the use of different camera brands did not seem to significantly affect the lesion detectability. Further simulations with more extended range of parameters and camera brands may be conducted in the future to quantify further the variability between different brands of cameras.

Instant and accurate evaluation of drug resistance in tumors before and during chemotherapy is important for patients with advanced colon cancer and is beneficial for prolonging their progression-free survival time. Here, the possible biomarkers that reflect the drug resistance of colon cancer were investigated using proton magnetic resonance spectroscopy (1H-MRS) in vivo. SW480[5-fluorouracil(5-FU)-responsive] and SW480/5-FU (5-FU-resistant) xenograft models were generated and subjected to in vivo 1H-MRS examinations when the maximum tumor diameter reached 1–1.5 cm. The areas under the peaks for metabolites, including choline (Cho), lactate (Lac), glutamine/glutamate (Glx), and myoinositol (Ins)/creatine (Cr) in the tumors, were analyzed between two groups. The resistancerelated protein expression, cell morphology, necrosis, apoptosis, and cell survival of these tumor specimens were assessed. The content for tCho, Lac, Glx, and Ins/Cr in the tumors of the SW480 group was significantly lower than that of the SW480/5-FU group (p < 0.05). While there was no significant difference in the degree of necrosis and apoptosis rate of tumor cells between the two groups (p > 0.05), the tumor cells of the SW480/5-FU showed a higher cell density and larger nuclei. The expression levels of resistance-related proteins (P-gp, MPR1, PKC) in the SW480 group were lower than those in the SW480/5-FU group (p < 0.01). The survival rate of 5-FU-resistant colon cancer cells was significantly higher than that of 5-FUresponsive ones at 5-FU concentrations greater than 2.5 μg/mL (p < 0.05). These results suggest that alterations in tCho, Lac, Glx1, Glx2, and Ins/Cr detected by 1H-MRS may be used for monitoring tumor resistance to 5-FU in vivo.

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.

Clinical/methodological problem: The identification of clinically significant prostate carcinomas while avoiding overdiagnosis of low-malignant tumors is a challenge in routine clinical practice. Standard radiologic procedures: Multiparametric magnetic resonance imaging (MRI) of the prostate acquired and interpreted according to PI-RADS (Prostate Imaging Reporting and Data System Guidelines) is accepted as a clinical standard among urologists and radiologists. Methodological innovations: The PI-RADS guidelines have been newly updated to version 2.1 and, in addition to more precise technical requirements, include individual changes in lesion assessment. Performance: The PI-RADS guidelines have become crucial in the standardization of multiparametric MRI of the prostate and provide templates for structured reporting, facilitating communication with the referring physician. Evaluation: The guidelines, now updated to version 2.1, represent a refinement of the widely used version 2.0. Many aspects of reporting have been clarified, but some previously known limitations remain and require further improvement of the guidelines in future versions.

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.