Right hippocampal volume and attenuated psychotic symptoms distinguish subgroups of youth with common patterns of temporoparietal effective connectivity.

Deep learning algorithms, particularly convolutional neural networks, have emerged as a dominant force in medical image analysis. Their superior performance in tasks like image classification, segmentation, and object detection has significantly advanced the field. Medical image classification (MIC), a fundamental task involving the categorization of medical images into predefined classes (e.g., health status, tumor type), is a prime application area for these models. Accurate MIC can support clinical decision-making and improve diagnostic efficiency. Given the critical implications of MIC outcomes, such as influencing patient prognosis, and potentially leading to misdiagnosis if inaccurate, rigorous evaluation and validation are essential to ensure reliability and clinical applicability. The demand for products and services designed to improve skin health and appearance continues to grow. Digital tools for skin analysis are playing an increasingly important role in this field. These technologies enable the assessment of facial features and skin conditions, facilitating the development of personalized skincare regimens. However, a significant challenge is the lack of robust, standardized methods for automated skin profiling through image analysis. While major commercial entities (e.g., L’Oréal, Haut.AI, Perfect Corp, and Canfield) have successfully deployed multi-attribute facial assessment systems at scale, their underlying datasets, annotation protocols, and modeling methodologies remain strictly proprietary. Consequently, the academic community lacks a standardized, open-source benchmark to evaluate and advance selfie-based cosmetic profiling. This work directly addresses this gap. To prevent conceptual misinterpretation, we explicitly clarify the intended use and regulatory framing of this system: this approach is strictly designed as an appearance-based profiling tool for cosmetic skin feature assessment and personalized skincare guidance. It is explicitly non-diagnostic and is not intended to replace professional clinical dermatological evaluation or to function as a medical device.