Brain tumors pose a significant medical challenge, characterized by high incidence and mortality rates, which underscore the critical need for accurate and early diagnosis using minimally invasive techniques such as magnetic resonance imaging. In this context, Artificial Intelligence has emerged as a promising tool to enhance diagnostic precision and efficiency. However, its widespread adoption in clinical practice remains limited due to the opacity of Artificial Intelligence-driven decision-making processes. To address this challenge, we introduce BrAInVision , a hybrid and doubly explainable AI framework for brain tumor detection. The novelty of our approach is the integration of both deep learning and traditional machine learning techniques, combining deep-extracted features with hand-crafted features to create a more robust and interpretable classification system. In contrast to conventional single-explanation methods, our framework provides comprehensive explainability through a multi-level analytical approach, enhancing both interpretability and transparency. The first level employs Grad-CAM to visualize regions of interest identified by the deep feature extractor, while the second level utilizes Permutation Feature Importance and Partial Dependence Plots to understand and quantify the contribution of specific image characteristics to diagnostic decisions. The proposed framework achieved an F1-score of 97% on the four classes (Glioma/Meningioma/Pituitary/NoTumor) and an average 99% in binary classification (Glioma/NoTumor), outperforming current state-of-the-art methods. The proposed approach has been validated on both the original dataset and an independent dataset with radiologist-annotated tumor masks, demonstrating strong generalizability. Designed for seamless integration into radiologists’ workflows as a decision support system, BrAInVision ensures a high degree of explainability, thereby fostering greater trust in AI-assisted medical decision-making. • A new framework that offers high accuracy and multi-level explainability is proposed. • The framework distinguish three distinct types of brain tumors from a healthy class. • A visual explanations of deep features is given through Grad-CAM. • A global and model-agnostic feature-level explanation is given using PFI and PDP. • A new and scientifically reliable dataset is explored.

Non-small cell lung cancer (NSCLC) remains a major global health challenge, with high post-surgical recurrence rates underscoring the need for accurate pathological response predictions to guide personalized treatments. Although artificial intelligence models show promise in this domain, their clinical adoption is limited by the lack of medically grounded guidance during training, often resulting in non-explainable intrinsic predictions. To address this, we propose Doctor-in-the-Loop , a novel framework that integrates expert-driven domain knowledge with explainable artificial intelligence techniques, directing the model toward clinically relevant anatomical regions and improving both interpretability and trustworthiness. Our approach employs a gradual multi-view strategy, progressively refining the model’s focus from broad contextual features to finer, lesion-specific details. By incorporating domain insights at every stage, we enhance predictive accuracy while ensuring that the model’s decision-making process aligns more closely with clinical reasoning. Evaluated on a dataset of NSCLC patients, Doctor-in-the-Loop delivers promising predictive performance and provides transparent, justifiable outputs, representing a significant step toward clinically explainable artificial intelligence in oncology. • Integrating clinical expertise with XAI for NSCLC pathological response prediction. • Gradual multi-view strategy refines focus from context to lesion-specific regions. • Improved explainability guides the model to clinically relevant anatomical regions. • Outperforms state-of-the-art, achieving higher accuracy and clinical relevance.