Deep learning (DL) has revolutionized medical image analysis (MIA), enabling early anomaly detection, precise lesion segmentation, and automated disease classification. However, its clinical integration faces two major challenges: reliance on limited, narrowly annotated datasets that inadequately capture real-world patient diversity, and the inherent “black-box” nature of DL decision-making, which complicates physician scrutiny and accountability. Eye tracking (ET) technology offers a transformative solution by capturing radiologists’ gaze patterns to generate supervisory signals. These signals enhance DL models through two key mechanisms: providing weak supervision to improve feature recognition and diagnostic accuracy, particularly when labeled data are scarce, and enabling direct comparison between machine and human attention to bridge interpretability gaps and build clinician trust. This approach also extends effectively to multimodal learning models (MLMs) and vision–language models (VLMs), supporting the alignment of machine reasoning with clinical expertise by grounding visual observations in diagnostic context, refining attention mechanisms, and validating complex decision pathways. Conducted in accordance with the PRISMA statement and registered in PROSPERO (ID: CRD42024569630), this review synthesizes state-of-the-art strategies for ET-DL integration. We further propose a unified framework in which ET innovatively serves as a data efficiency optimizer, a model interpretability validator, and a multimodal alignment supervisor. This framework paves the way for clinician-centered AI systems that prioritize verifiable reasoning, seamless workflow integration, and intelligible performance, thereby addressing key implementation barriers and outlining a path for future clinical deployment.

Background: The Vesical Imaging-Reporting and Data System (VI-RADS) has limited diagnostic accuracy in distinguishing non-muscle-invasive bladder cancer (NMIBC) within VI-RADS categories 2 and 3, despite its value for overall NMIBC assessment. Dynamic contrast-enhanced MRI (DCE-MRI), which reflects tumor vascularity, holds promise for improving these challenging cases but remains underutilized due to unexploited spatiotemporal information. Methods: We developed a deep learning model to comprehensively quantify spatiotemporal features from multiphase DCE-MRI in 184 patients with VI-RADS 2 or 3 (training: n = 115, validation: n = 20, testing: n = 49). The model integrated multiscale feature extraction and contextual attention mechanisms to enhance diagnostic performance. Results: The model outperformed established benchmarks (e.g., VGG, ResNet) and the conventional VI-RADS ≤ 2 threshold (sensitivity: 0.67 for NMIBC), achieving a sensitivity of 0.90 (95% CI: 0.81–0.96) for NMIBC and an area under the curve (AUC) of 0.82 (95% CI: 0.75–0.89) for overall classification. Visualizations confirmed its ability to identify key spatiotemporal patterns linked to muscle invasion. Conclusions: By leveraging comprehensive spatiotemporal information from DCE-MRI, our deep learning model significantly improves NMIBC diagnosis in VI-RADS 2/3 cases, offering a clinically valuable tool to address the limitations of current VI-RADS assessment.

Fibroblast growth factor receptor 3 (FGFR3) is a key oncogenic driver in bladder cancer (BCa). Although erdafitinib, the pioneering FGFR3 inhibitor, is approved for clinic use, its oral administration frequently causes inadequate therapeutic effect and side effects. Here, we report a pH-responsive and erdafitinib-loaded liposomal nanosystem with surface modification of anti-FGFR3 antibodies (Phl-erda@Ab), which demonstrates improved antitumor effects in intravesical instillation therapy. In RT112 cells, we evaluated targeted uptake and antiproliferative activity. In vivo, we assessed intravesical retention, antitumor efficacy, and histopathological markers (TUNEL, Ki67) in a BCa xenograft BALB/C nude mouse model, alongside systemic safety profiling. The Phl-erda@Ab exhibited sharp pH sensitivity, releasing 49% of the drug payload within 1 h at pH 5.3 while maintaining minimal release at pH 7.4 and 6.5. Liposome retention after intravesical instillation was doubled compared with non-targeted Phl-erda in mice. The Phl-erda@Ab intravesical therapy demonstrated effective antitumor activity, including maximal necrosis, the highest TUNEL-positive apoptotic index, and the lowest Ki67 proliferation index, compared to other groups. No significant pathological damage to major organs was observed. In summary, Phl-erda@Ab enables FGFR3-targeted, pH-triggered intravesical delivery with enhanced intravesical retention, improved antitumor activity, and favorable local safety, providing preclinical evidence to support clinical translation in FGFR3-overexpressing BCa.