Adaptive Aquila Optimizer with Explainable Artificial Intelligence-Enabled Cancer Diagnosis on Medical Imaging.
Salem Alkhalaf1, Fahad Alturise1, Adel Aboud Bahaddad2
1Department of Computer, College of Science and Arts in Ar Rass, Qassim University, Ar Rass 58892, Saudi Arabia.
Cancers
|March 11, 2023
Summary
This study introduces an Explainable Artificial Intelligence (XAI) method for enhanced cancer diagnosis in medical imaging. The Adaptive Aquila Optimizer with XAI (AAOXAI-CD) improves accuracy and transparency in detecting colorectal and osteosarcoma cancers.
Area of Science:
- Artificial Intelligence in Medical Imaging
- Explainable AI (XAI) for Diagnostic Systems
- Deep Learning for Cancer Classification
Background:
- Explainable Artificial Intelligence (XAI) aims to make AI decisions understandable, crucial for medical applications like cancer diagnosis.
- Current AI diagnostic tools often lack transparency, hindering trust and clinical adoption.
- Need for interpretable AI models in medical imaging to highlight diagnostic indicators and decision processes.
Purpose of the Study:
- To develop an Adaptive Aquila Optimizer with Explainable Artificial Intelligence Enabled Cancer Diagnosis (AAOXAI-CD) technique for medical imaging.
- To achieve effective classification of colorectal and osteosarcoma cancers using an explainable AI approach.
- To enhance transparency and trust in AI-driven cancer diagnosis through clear explanations.
Main Methods:
- Feature vector generation using the Faster SqueezeNet model.
- Hyperparameter tuning of Faster SqueezeNet via the Adaptive Aquila Optimizer (AAO).
- Cancer classification using a majority weighted voting ensemble of Recurrent Neural Network (RNN), Gated Recurrent Unit (GRU), and Bidirectional Long Short-Term Memory (BiLSTM) classifiers.
- Integration of the LIME (Local Interpretable Model-agnostic Explanations) technique for XAI.
Main Results:
- The AAOXAI-CD technique demonstrated effective classification of colorectal and osteosarcoma cancers.
- The methodology provided clear explanations for diagnostic decisions, highlighting indicative image areas.
- Simulation results showed superior performance of AAOXAI-CD compared to existing approaches on medical cancer imaging datasets.
Conclusions:
- The developed AAOXAI-CD technique offers a transparent and accurate AI-driven solution for cancer diagnosis in medical imaging.
- Combining advanced DL models with XAI and optimization algorithms enhances diagnostic reliability and interpretability.
- The study underscores the potential of XAI to increase clinician and patient trust in AI-based medical diagnostics.
Related Concept Videos
02:53Adaptive Mechanisms in Cancer Cells
5.9K
Cancer cells accumulate genetic changes at an abnormally rapid rate due to the defects in the DNA repair mechanisms. From an evolutionary perspective, such genetic instability is advantageous for cancer development. Mutant cell lines accumulate a series of beneficial mutations that contribute to their progression into cancer.
Some of the advantages that cancer cells have on normal cells include - enhanced ability to divide without terminally differentiating, induce new blood vessel formation,...
Some of the advantages that cancer cells have on normal cells include - enhanced ability to divide without terminally differentiating, induce new blood vessel formation,...
5.9K
01:21Cancer Survival Analysis
407
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
407