Background/Objectives: Accurate classification of brain tumors and reliable prediction of patient survival are essential in neuro-oncology, guiding clinical decisions and enabling precision treatment planning. However, conventional machine learning and deep learning methods often struggle with challenges such as data scarcity, class imbalance, limited model interpretability, and poor generalization across diverse clinical settings. This study presents GlioSurvQNet, a novel reinforcement learning-based framework designed to address these limitations for both glioma grading and survival prediction. Methods: GlioSurvQNet is built upon a DuelContextAttn Deep Q-Network (DQN) architecture, tailored for binary classification of low-grade vs. high-grade gliomas and multi-class survival prediction (short-, medium-, and long-term categories). Radiomics features were extracted from multimodal MRI scans, including FLAIR, T1CE, and T2 sequences. Feature optimization was performed using a hybrid ensemble of metaheuristic algorithms, including Harris Hawks Optimization (HHO), Modified Gorilla Troops Optimization (mGTO), and Zebra Optimization Algorithm (ZOA). Subsequently, SHAP-based feature selection was applied to enhance model interpretability and robustness. Results: The classification module achieved the highest accuracy of 99.27% using the FLAIR + T1CE modality pair, while the survival prediction model attained an accuracy of 93.82% with the FLAIR + T2 + T1CE fusion. Comparative evaluations against established machine learning and deep learning models demonstrated that GlioSurvQNet consistently outperformed existing approaches in both tasks. Conclusions: GlioSurvQNet offers a powerful and interpretable AI-driven solution for brain tumor analysis. Its high accuracy and robustness make it a promising tool for clinical decision support in glioma diagnosis and prognosis.
This study aimed to generate synthetic MR images from real CT images. CT# mean and standard deviation of a moving window across every pixel in the reconstructed CT images were mapped to their corresponding tissue-mimicking types. Identification of the tissue enabled remapping it to its corresponding intrinsic parameters: T1, T2, and proton density (ρ). Lastly, synthetic weighted MR images of a selected slice were generated by simulating a spin-echo sequence using the intrinsic parameters and proper contrast parameters (TE and TR). Experiments were performed on a 3D multimodality abdominal phantom and on human knees at different TE and TR parameters to confirm the clinical effectiveness of the approach. Results demonstrated the validity of the approach of generating synthetic MR images at different weightings using only CT images and the three predefined mapping functions. The slope of the fitting line and percentage root-mean-square difference (PRD) between real and synthetic image vector representations were (0.73, 10%), (0.9, 18%), and (0.2, 8.7%) for T1-, T2-, and ρ-weighted images of the phantom, respectively. The slope and PRD for human knee images, on average, were 0.89% and 18.8%, respectively. The generated MR images provide valuable guidance for physicians with regard to deciding whether acquiring real MR images is crucial.
12 December 2023 in Computers in Biology and Medicine
Conclusions in literature regarding the effect of Autism on the size of different brain structures are contradictory. The aim of this study is to determine the effect of Autism on the volumes of different brain subcortical structures, and the age stage at which those changes occur. 7 main brain structures were segmented and their volumes were obtained. Volumes and the ratio of the volume to total brain volume (SBR) were compared in Autism group to their corresponding values in Control group. Then, each group was divided into 4 subgroups based on age; the comparison was repeated for each subgroup. Independent t-test was used to determine if significant differences existed between compared groups. Significant reductions were observed in the SBR of Autistic Pallidum and Accumbens compared to Control group when considering the full range of ages (5–25 years). However, Amygdala’s volume was significantly smaller in Autism in the 5–8 year subgroup. In addition, the SBR of Putaman, Pallidum, Hippocampus, and Accumbens were reduced in the 18–25 year Autism subgroup. In conclusion, the alteration in the ratio of structure’s volume to total brain volume is a better indicator of Autism diagnosis than change in the absolute volume alone.
Gel dosimeters are commonly used to verify the dose distribution and consequently, ensure the accurate delivery of the dose as planned, which is very critical for the success of the radiotherapy. The current study aims to investigate the feasibility of using the magnetic resonance imaging (MRI)-based relaxation rate maps (R2 maps) of the novel acrylic acid with organic glucose polyvinyl alcohol (ACAGLPVA) hydrogel dosimeter for 3-D dose verification. The (ACAGLPVA) hydrogel was prepared and scanned with a 3-T MRI, prior to irradiation, using a spin-echo (SE) sequence. A series of 30 images with different echo times (TE) were acquired in order to generate R2 maps. The hydrogel was then irradiated with a maximum dose of 8 Gy using a CyberKnife system and the same MRI scans were acquired again for the irradiated gel phantom. The measured dose distributions obtained from R2-MRI maps were compared, using 2-D and 3-D Gamma analyses with 3% dose difference and 3-mm distance-to-agreement criteria, to those calculated from the treatment planning system (TPS). Results have shown high spatial agreement between the dose distributions calculated from TPS and the dose distributions derived from R2 maps, particularly in high-dose regions. The 3-D Gamma analysis achieved pass rates exceeding 90% across all planes for all evaluated dose levels—namely 90%, 70%, 50% and 20% of the maximum dose in the axial plane, as well as 90%, 70% and 50% of the maximum dose in the coronal and sagittal planes. In conclusion, based on the analysis of R2-MRI maps, this study has demonstrated the reliability of using the novel (ACAGLPVA) hydrogel dosimeter as an accurate tool for 3-D verification of radiotherapy treatment plans.
Artificial intelligence is increasingly used in connected medical-device systems for physiological-signal classification and continuous monitoring. This paper presents a simulation study of the learning and policy layers of a zero-trust federated-learning framework for Internet of Medical Things (IoMT) devices. The implemented prototype combines federated averaging, Gaussian perturbation of client updates, a device-level trust score, and gradient-magnitude screening of anomalous client updates. This is a simulation study of the federated-learning and zero-trust policy layers; cryptographic, encrypted-transport, and hardware-attestation components are modeled as deployment considerations or overheads and are not implemented as complete physical or protocol-level modules. Using the ECG classification pipeline described in the methodology, we compare centralized training, traditional federated learning, differentially private federated learning, and the proposed zero-trust federated configuration. The proposed ZT-FL configuration is reported at 0.928 accuracy, 0.920 F1-score, and 0.968 AUC. These results support feasibility within the stated simulation scope and do not establish clinical validity, formal end-to-end DP accounting, comprehensive adversarial robustness, regulatory compliance, or post-quantum readiness.