Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
February 21, 2026AlgorithmsOpen Access

A Trust-Centered Explainable Deep-Learning Framework for Acute Lymphoblastic Leukemia Detection Using Multi-Model Fusion and Interpretability Scoring

View Full Paper
Ask AI
Bookmark
Share

Authors

KPKhadija ParwezSSSyed Irfan SohailMBMuhammad Bilal

Discussion

Loading...

Member takes

Overview

Demonstrates a trust-centered framework for accurate leukemia detection in children, highlighting its potential for clinical adoption.

Key Points

  • The study aims to develop an explainable deep-learning framework that enhances trust and transparency in leukemia diagnostics.
  • Evaluated 153 microscopic blood smear images for model training and testing.
  • Utilized multiple transfer-learning models with a focus on EfficientNetB4.
  • Introduced a unified interpretability score to measure model trustworthiness.
  • Conducted human-centric evaluations using clinician feedback on trust scales.
  • Achieved a diagnostic accuracy of 98.31% with the EfficientNetB4 model.
  • Introduced a novel quantitative trust formulation, integrating diagnostic performance and clinician feedback.
  • Enhanced transparency via visual and textual explanations and advanced fusion heatmaps.

Cite This Study

Parwez et al. (2026) studied this question.

synapsesocial.com/papers/69994c4b873532290d020b02https://doi.org/10.3390/a19020162
View Full Paper
Ask AI
Bookmark
Share