Interpreting artificial intelligence models: a systematic review on the application of LIME and SHAP in Alzheimer’s disease detection

Brain Informatics, Journal Year: 2024, Volume and Issue: 11(1)

Published: April 5, 2024

Abstract Explainable artificial intelligence (XAI) has gained much interest in recent years for its ability to explain the complex decision-making process of machine learning (ML) and deep (DL) models. The Local Interpretable Model-agnostic Explanations (LIME) Shaply Additive exPlanation (SHAP) frameworks have grown as popular interpretive tools ML DL This article provides a systematic review application LIME SHAP interpreting detection Alzheimer’s disease (AD). Adhering PRISMA Kitchenham’s guidelines, we identified 23 relevant articles investigated these frameworks’ prospective capabilities, benefits, challenges depth. results emphasise XAI’s crucial role strengthening trustworthiness AI-based AD predictions. aims provide fundamental capabilities XAI enhancing fidelity within clinical decision support systems prognosis.

Language: Английский

---

Robustness in deep learning models for medical diagnostics: security and adversarial challenges towards robust AI applications

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 58(1)

Published: Nov. 8, 2024

The current study investigates the robustness of deep learning models for accurate medical diagnosis systems with a specific focus on their ability to maintain performance in presence adversarial or noisy inputs. We examine factors that may influence model reliability, including complexity, training data quality, and hyperparameters; we also security concerns related attacks aim deceive along privacy seek extract sensitive information. Researchers have discussed various defenses these enhance robustness, such as input preprocessing, mechanisms like augmentation uncertainty estimation. Tools packages extend reliability features frameworks TensorFlow PyTorch are being explored evaluated. Existing evaluation metrics additionally This paper concludes by discussing limitations existing literature possible future research directions continue enhancing status this topic, particularly domain, ensuring AI trustworthy, reliable, stable.

Language: Английский

---

An Interpretable PyCaret Approach for Alzheimer's Disease Prediction

International Journal of Computational and Experimental Science and Engineering, Journal Year: 2024, Volume and Issue: 10(4)

Published: Nov. 29, 2024

Alzheimer's Disease (AD) is a major global health concern. The research focuses on early and accurate diagnosis of AD for its effective treatment management. This study presents novel Machine Learning (ML) approach utilizing PyCaret SHAP interpretable prediction. employs span classification algorithms the identifies best model. value determines contribution individual features final prediction thereby enhancing model’s interpretability. feature selection using improves overall performance proposed XAI framework clinical decision making patient care by providing reliable transparent method detection.

Language: Английский

---

A Comparative Study of Pretrained Deep Neural Networks for Classifying Alzheimer's and Parkinson's Disease

2021 IEEE Symposium Series on Computational Intelligence (SSCI), Journal Year: 2023, Volume and Issue: unknown, P. 1334 - 1339

Published: Dec. 5, 2023

Early detection of neurodegenerative diseases can be challenging, where Deep Learning (DL) techniques have shown promise. Most DL provide a robust and accurate classification performance. However, due to the complex architectures models, results are difficult interpret, causing challenges for their adoption in healthcare industry. To facilitate this, current work proposes an effective interpretable analysis pipeline that compares performances pre-trained models early Alzheimer's Disease (AD) Parkinson's (PD). The proposed allows tuning hyperparameters, such as batch size, number epochs, learning rates, achieve more classification. Additionally, validation predictions using heatmaps drawn from GradCAM also provided.

Language: Английский

---

Enhancing Alzheimer’s Disease Detection through Ensemble Learning of Fine-Tuned Pre-Trained Neural Networks

Electronics, Journal Year: 2024, Volume and Issue: 13(17), P. 3452 - 3452

Published: Aug. 30, 2024

Alzheimer’s Disease, a progressive brain disorder that impairs memory, thinking, and behavior, has started to benefit from advancements in deep learning. However, the application of learning medicine faces challenge limited data resources for training models. Transfer offers solution by leveraging pre-trained models similar tasks, reducing computational requirements achieve high performance. Additionally, augmentation techniques, such as rotation scaling, help increase dataset size. In this study, we worked with magnetic resonance imaging (MRI) datasets applied various pre-processing techniques including include intensity normalization, affine registration, skull stripping, entropy-based slicing, flipping, zooming, shifting, rotating clean expand dataset. We transfer high-performing models—ResNet-50, DenseNet-201, Xception, EfficientNetB0, Inception V3, originally trained on ImageNet. fine-tuned these using feature extraction technique augmented data. Furthermore, implemented ensemble stacking boosting, enhance final prediction The novel methodology achieved precision (95%), recall (94%), F1 score accuracy (95%) disease detection. Overall, study establishes robust framework applying machine diagnose MRI scans. combination learning, via neural networks processed dataset, proven highly effective, marking significant advancement medical diagnostics.

Language: Английский

---

Deep Learning-based Classification of MRI Images for Early Detection and Staging of Alzheimer's Disease

International Journal of Advanced Computer Science and Applications, Journal Year: 2024, Volume and Issue: 15(5)

Published: Jan. 1, 2024

Alzheimer's disease (AD) poses a significant challenge to modern healthcare, as effective treatment remains elusive. Drugs may slow down the progress of disease, but there is currently no cure for it. Early AD identification crucial providing required medications before brain damage occurs. In this course research, we studied various deep learning techniques address early detection by utilizing structural MRI (sMRI) images biomarkers. Deep are pivotal in accurately analyzing vast amounts data identify and anticipate its progression. A balanced image dataset 12,936 was used study extract sufficient features distinguishing stages, due similarities characteristics necessitating more than previous studies. The GoogLeNet model utilized our investigation derive from each scan image. These were then inputted into feed-forward neural network (FFNN) stage prediction. FFNN model, features, underwent rigorous training over multiple epochs using small batch size ensure robust performance on unseen achieved 98.37% accuracy, 98.39% sensitivity, 98.50% precision, 99.45% specificity. Most remarkably, results show that detected with an amazing average accuracy rate 99.01%.

Language: Английский

---

Impacts of Environmental Factors on Wellbeing: Machine Learning-Based Benchmarking of Spatial and Temporal Properties

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 631 - 642

Published: Jan. 1, 2024

Language: Английский

---

Performance Analysis of a Single-Input Thermal Image Classifier with Patient Information for the Detection of Breast Cancer

Communications in computer and information science, Journal Year: 2024, Volume and Issue: unknown, P. 148 - 168

Published: Jan. 1, 2024

Language: Английский

---

Optimizing Medical Imaging Quality: An In-Depth Examination of Preprocessing Methods for Brain MRIs

Communications in computer and information science, Journal Year: 2024, Volume and Issue: unknown, P. 65 - 81

Published: Jan. 1, 2024

Language: Английский

---

Transfer Learning-Based Ensemble of Deep Neural Architectures for Alzheimer’s and Parkinson’s Disease Classification

Communications in computer and information science, Journal Year: 2024, Volume and Issue: unknown, P. 186 - 204

Published: Jan. 1, 2024

Language: Английский

---