The development of foundation models for brain MRI depends critically on the scale, diversity, and consistency of available data, yet systematic assessments of these factors remain scarce. In this study, we analyze 54 publicly accessible brain MRI datasets encompassing over 538,031 scans to provide a structured, multi-level overview tailored to foundation model development. At the dataset level, we characterize modality composition, disease coverage, and dataset scale, revealing strong imbalances between large healthy cohorts and smaller clinical populations. At the image level, we quantify voxel spacing, orientation, and intensity distributions across 14 representative datasets, demonstrating substantial heterogeneity that can influence representation learning. We then perform a quantitative evaluation of preprocessing variability, examining how intensity normalization, bias field correction, skull stripping, spatial registration, and interpolation alter voxel statistics and geometry. While these steps improve within-dataset consistency, residual differences persist between datasets. Finally, a feature-space case study using a 3D DenseNet121 shows measurable residual covariate shift after standardized preprocessing, confirming that harmonization alone cannot eliminate inter-dataset bias. Together, these analyses provide a unified characterization of variability in public brain MRI resources and emphasize the need for preprocessing-aware and domain-adaptive strategies in the design of generalizable brain MRI foundation models.

This research delves into leveraging artificial intelligence for tuberculosis (TB) screening via chest radiography, addressing a persistent global health threat that impacts millions each year. The study's primary goal is to innovate remote diagnostic techniques that apply computer vision to facilitate early disease detection. This investigation critically evaluates convolutional neural networks, particularly ResNet50 and EfficientNet-B7, for their effectiveness in localizing and identifying pulmonary lesions. The findings reveal that ResNet50 excels in delineating lung structures, while EfficientNet-B7 outperforms in pathology identification. Both models, however, demonstrate pronounced sensitivity to image quality, often encountering difficulty in differentiating pathologies in complex anatomical regions. Achieving an impressive accuracy of approximately 99.6% on publicly accessible datasets, the models experienced a performance reduction to 79.3% on private clinical datasets, underscoring the variability in model efficacy across different data sources. The study underscores the critical role of explainable AI (XAI) methods, including Grad-CAM, Captum, and SHAP, in enhancing interpretability and aiding clinicians in comprehending the decision-making pathways of AI models. In sum, this research highlights AI's potential in advancing TB screening while addressing inherent challenges related to image fidelity and model generalizability.