To further evaluate the performance of the proposed model

The gap

Evidence profile

Sourced from the future-work section and inline gaps and synthesized of the source papers, classified as general, spanning 3 journals.

Supporting evidence — 3 representative gaps

• Multi-Class Brain Tumor Classification from MRI Using ResNet50V2 Transfer Learning with Two-Stage Augmentation, Grad-CAM Explainability, and Cross-Dataset Validation (2026) · International Journal for Research in Applied Science and Engineering Technology · doi

  To further evaluate the performance of the proposed model on larger and more diverse datasets. To explore the application of the proposed model to other medical imaging tasks. To develop more advanced explainability techniques that can provide deeper insights into the decisions made by the model.

  generalfuture-work section
  Keywords: further evaluate performance proposed model larger diverse datasets
• Improved CNN with BiLSTM model for early melanoma and skin lesion classification (2026) · Frontiers in Artificial Intelligence · doi

  Future studies are needed to confirm the framework the use of attention mechanisms and on various data sets, explainable AI tools such as Grad-CAM to improve interpretability, multimodal data such as patient clinical metadata, and prospective clinical trials to confirm the usefulness of diagnosing in clinics and patient outcomes. shown good performance in medical image classification tasks because they can capture long-range dependencies and contextual relationships between images in the image itself that are not limited to the local context of a single pixel area (Li et al.

  generalinline gapsevidence 5/5
  Keywords: patient clinical image future needed framework attention mechanisms various sets explainable tools grad improve interpretability
• From Black Box to Glass Box: A Survey of Explainable AI in Mammographic Screening (2026) · Sakarya University Journal of Computer and Information Sciences · doi

  Medical vision-language models lack interpretability and explainability mechanisms needed for clinical trust; while explainable AI methods exist for individual imaging tasks, there is no systematic integration of XAI techniques into multimodal medical vision-language models for transparent clinical decision-making.

  generalsynthesizedevidence 5/5
  Keywords: medical vision-language models lack interpretability explainability mechanisms needed

Questions about this gap