1
|
Bloch L, Friedrich CM. Systematic comparison of 3D Deep learning and classical machine learning explanations for Alzheimer's Disease detection. Comput Biol Med 2024; 170:108029. [PMID: 38308870 DOI: 10.1016/j.compbiomed.2024.108029] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/09/2023] [Revised: 01/25/2024] [Accepted: 01/25/2024] [Indexed: 02/05/2024]
Abstract
Black-box deep learning (DL) models trained for the early detection of Alzheimer's Disease (AD) often lack systematic model interpretation. This work computes the activated brain regions during DL and compares those with classical Machine Learning (ML) explanations. The architectures used for DL were 3D DenseNets, EfficientNets, and Squeeze-and-Excitation (SE) networks. The classical models include Random Forests (RFs), Support Vector Machines (SVMs), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting (LightGBM), Decision Trees (DTs), and Logistic Regression (LR). For explanations, SHapley Additive exPlanations (SHAP) values, Local Interpretable Model-agnostic Explanations (LIME), Gradient-weighted Class Activation Mapping (GradCAM), GradCAM++ and permutation-based feature importance were implemented. During interpretation, correlated features were consolidated into aspects. All models were trained on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. The validation includes internal and external validation on the Australian Imaging and Lifestyle flagship study of Ageing (AIBL) and the Open Access Series of Imaging Studies (OASIS). DL and ML models reached similar classification performances. Regarding the brain regions, both types focus on different regions. The ML models focus on the inferior and middle temporal gyri, and the hippocampus, and amygdala regions previously associated with AD. The DL models focus on a wider range of regions including the optical chiasm, the entorhinal cortices, the left and right vessels, and the 4th ventricle which were partially associated with AD. One explanation for the differences is the input features (textures vs. volumes). Both types show reasonable similarity to a ground truth Voxel-Based Morphometry (VBM) analysis. Slightly higher similarities were measured for ML models.
Collapse
Affiliation(s)
- Louise Bloch
- Department of Computer Science, University of Applied Sciences and Arts Dortmund (FH Dortmund), Emil-Figge-Straße 42, Dortmund, 44227, North Rhine-Westphalia, Germany; Institute for Medical Informatics, Biometry and Epidemiology (IMIBE), University Hospital Essen, Hufelandstraße 55, Essen, 45122, North Rhine-Westphalia, Germany; Institute for Artificial Intelligence in Medicine (IKIM), University Hospital Essen, Hufelandstraße 55, Essen, 45122, North Rhine-Westphalia, Germany.
| | - Christoph M Friedrich
- Department of Computer Science, University of Applied Sciences and Arts Dortmund (FH Dortmund), Emil-Figge-Straße 42, Dortmund, 44227, North Rhine-Westphalia, Germany; Institute for Medical Informatics, Biometry and Epidemiology (IMIBE), University Hospital Essen, Hufelandstraße 55, Essen, 45122, North Rhine-Westphalia, Germany.
| |
Collapse
|
2
|
Sheng J, Zhang Q, Zhang Q, Wang L, Yang Z, Xin Y, Wang B. A hybrid multimodal machine learning model for Detecting Alzheimer's disease. Comput Biol Med 2024; 170:108035. [PMID: 38325214 DOI: 10.1016/j.compbiomed.2024.108035] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/14/2023] [Revised: 01/03/2024] [Accepted: 01/26/2024] [Indexed: 02/09/2024]
Abstract
Alzheimer's disease (AD) diagnosis utilizing single modality neuroimaging data has limitations. Multimodal fusion of complementary biomarkers may improve diagnostic performance. This study proposes a multimodal machine learning framework integrating magnetic resonance imaging (MRI), positron emission tomography (PET) and cerebrospinal fluid (CSF) assays for enhanced AD characterization. The model incorporates a hybrid algorithm combining enhanced Harris Hawks Optimization (HHO) algorithm referred to as ILHHO, with Kernel Extreme Learning Machine (KELM) classifier for simultaneous feature selection and classification. ILHHO enhances HHO's search efficiency by integrating iterative mapping (IM) to improve population diversity and local escaping operator (LEO) to balance exploration-exploitation. Comparative analysis with other improved HHO algorithms, classic meta-heuristic algorithms (MHAs), and state-of-the-art MHAs on IEEE CEC2014 benchmark functions indicates that ILHHO achieves superior optimization performance compared to other comparative algorithms. The synergistic ILHHO-KELM model is evaluated on 202 AD Neuroimaging Initiative (ADNI) subjects. Results demonstrate superior multimodal classification accuracy over single modalities, validating the importance of fusing heterogeneous biomarkers. MRI + PET + CSF achieves 99.2 % accuracy for AD vs. normal control (NC), outperforming conventional and proposed methods. Discriminative feature analysis provides further insights into differential AD-related neurodegeneration patterns detected by MRI and PET. The differential PET and MRI features demonstrate how the two modalities provide complementary biomarkers. The neuroanatomical relevance of selected features supports ILHHO-KELM's potential for extracting sensitive AD imaging signatures. Overall, the study showcases the advantages of capitalizing on complementary multimodal data through advanced feature learning techniques for improving AD diagnosis.
Collapse
Affiliation(s)
- Jinhua Sheng
- School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China.
| | - Qian Zhang
- School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China; School of Data Science and Artificial Intelligence, Wenzhou University of Technology, Wenzhou, Zhejiang, 325035, China
| | - Qiao Zhang
- Beijing Hospital, Beijing, 100730, China; National Center of Gerontology, Beijing, 100730, China; Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, 100730, China
| | - Luyun Wang
- School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China
| | - Ze Yang
- School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China
| | - Yu Xin
- School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China
| | - Binbing Wang
- School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China
| |
Collapse
|
3
|
Zhang Q, Sheng J, Zhang Q, Wang L, Yang Z, Xin Y. Enhanced Harris hawks optimization-based fuzzy k-nearest neighbor algorithm for diagnosis of Alzheimer's disease. Comput Biol Med 2023; 165:107392. [PMID: 37669585 DOI: 10.1016/j.compbiomed.2023.107392] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/23/2023] [Revised: 07/30/2023] [Accepted: 08/25/2023] [Indexed: 09/07/2023]
Abstract
In order to stop deterioration and give patients with Alzheimer's disease (AD) early therapy, it is crucial to correctly diagnose AD and its early stage, mild cognitive impairment (MCI). A framework for diagnosing AD is presented in this paper, which includes magnetic resonance imaging (MRI) image preprocessing, feature extraction, and the Fuzzy k-nearest neighbor algorithm (FKNN) model. In particular, the framework's novelty lies in the use of an improved Harris Hawks Optimization (HHO) algorithm named SSFSHHO, which integrates the Sobol sequence and Stochastic Fractal Search (SFS) mechanisms for optimizing the parameters of FKNN. The HHO method improves the quality of the initial population overall by incorporating the Sobol sequence, and the SFS mechanism increases the algorithm's capacity to get out of the local optimum solution. Comparisons with other classical meta-heuristic algorithms, state-of-the-art HHO variants in low and high dimensions, and enhanced meta-heuristic algorithms on 30 typical IEEE CEC2014 benchmark test problems show that the overall performance of SSFSHHO is significantly better than other comparative algorithms. Moreover, the created framework based on the SSFSHHO-FKNN model is employed to classify AD and MCI using MRI scans from the ADNI dataset, achieving high classification performance for 6 representative cases. Experimental findings indicate that the proposed algorithm performs better than a number of high-performance optimization algorithms and classical machine learning algorithms, thus offering a promising approach for AD classification. Additionally, the proposed strategy can successfully identify relevant features and enhance classification performance for AD diagnosis.
Collapse
Affiliation(s)
- Qian Zhang
- College of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China; School of Data Science and Artificial Intelligence, Wenzhou University of Technology, Wenzhou, Zhejiang, 325035, China
| | - Jinhua Sheng
- College of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China.
| | - Qiao Zhang
- Beijing Hospital, Beijing, 100730, China; National Center of Gerontology, Beijing, 100730, China; Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, 100730, China
| | - Luyun Wang
- College of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China
| | - Ze Yang
- College of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China
| | - Yu Xin
- College of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou, Zhejiang, 310018, China; Key Laboratory of Intelligent Image Analysis for Sensory and Cognitive Health, Ministry of Industry and Information Technology of China, Hangzhou, Zhejiang, 310018, China
| |
Collapse
|
4
|
Mirkin S, Albensi BC. Should artificial intelligence be used in conjunction with Neuroimaging in the diagnosis of Alzheimer's disease? Front Aging Neurosci 2023; 15:1094233. [PMID: 37187577 PMCID: PMC10177660 DOI: 10.3389/fnagi.2023.1094233] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/09/2022] [Accepted: 03/27/2023] [Indexed: 05/17/2023] Open
Abstract
Alzheimer's disease (AD) is a progressive, neurodegenerative disorder that affects memory, thinking, behavior, and other cognitive functions. Although there is no cure, detecting AD early is important for the development of a therapeutic plan and a care plan that may preserve cognitive function and prevent irreversible damage. Neuroimaging, such as magnetic resonance imaging (MRI), computed tomography (CT), and positron emission tomography (PET), has served as a critical tool in establishing diagnostic indicators of AD during the preclinical stage. However, as neuroimaging technology quickly advances, there is a challenge in analyzing and interpreting vast amounts of brain imaging data. Given these limitations, there is great interest in using artificial Intelligence (AI) to assist in this process. AI introduces limitless possibilities in the future diagnosis of AD, yet there is still resistance from the healthcare community to incorporate AI in the clinical setting. The goal of this review is to answer the question of whether AI should be used in conjunction with neuroimaging in the diagnosis of AD. To answer the question, the possible benefits and disadvantages of AI are discussed. The main advantages of AI are its potential to improve diagnostic accuracy, improve the efficiency in analyzing radiographic data, reduce physician burnout, and advance precision medicine. The disadvantages include generalization and data shortage, lack of in vivo gold standard, skepticism in the medical community, potential for physician bias, and concerns over patient information, privacy, and safety. Although the challenges present fundamental concerns and must be addressed when the time comes, it would be unethical not to use AI if it can improve patient health and outcome.
Collapse
Affiliation(s)
- Sophia Mirkin
- Dr. Kiran C. Patel College of Osteopathic Medicine, Nova Southeastern University, Fort Lauderdale, FL, United States
| | - Benedict C. Albensi
- Barry and Judy Silverman College of Pharmacy, Nova Southeastern University, Fort Lauderdale, FL, United States
- St. Boniface Hospital Research, Winnipeg, MB, Canada
- University of Manitoba, Winnipeg, MB, Canada
| |
Collapse
|
5
|
Carcagnì P, Leo M, Del Coco M, Distante C, De Salve A. Convolution Neural Networks and Self-Attention Learners for Alzheimer Dementia Diagnosis from Brain MRI. SENSORS (BASEL, SWITZERLAND) 2023; 23:1694. [PMID: 36772733 PMCID: PMC9919436 DOI: 10.3390/s23031694] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 12/05/2022] [Revised: 12/28/2022] [Accepted: 01/30/2023] [Indexed: 06/18/2023]
Abstract
Alzheimer's disease (AD) is the most common form of dementia. Computer-aided diagnosis (CAD) can help in the early detection of associated cognitive impairment. The aim of this work is to improve the automatic detection of dementia in MRI brain data. For this purpose, we used an established pipeline that includes the registration, slicing, and classification steps. The contribution of this research was to investigate for the first time, to our knowledge, three current and promising deep convolutional models (ResNet, DenseNet, and EfficientNet) and two transformer-based architectures (MAE and DeiT) for mapping input images to clinical diagnosis. To allow a fair comparison, the experiments were performed on two publicly available datasets (ADNI and OASIS) using multiple benchmarks obtained by changing the number of slices per subject extracted from the available 3D voxels. The experiments showed that very deep ResNet and DenseNet models performed better than the shallow ResNet and VGG versions tested in the literature. It was also found that transformer architectures, and DeiT in particular, produced the best classification results and were more robust to the noise added by increasing the number of slices. A significant improvement in accuracy (up to 7%) was achieved compared to the leading state-of-the-art approaches, paving the way for the use of CAD approaches in real-world applications.
Collapse
|
6
|
An Enhanced Machine Learning Approach for Brain MRI Classification. Diagnostics (Basel) 2022; 12:diagnostics12112791. [PMID: 36428850 PMCID: PMC9689115 DOI: 10.3390/diagnostics12112791] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/22/2022] [Revised: 11/07/2022] [Accepted: 11/10/2022] [Indexed: 11/16/2022] Open
Abstract
Magnetic Resonance Imaging (MRI) is a noninvasive technique used in medical imaging to diagnose a variety of disorders. The majority of previous systems performed well on MRI datasets with a small number of images, but their performance deteriorated when applied to large MRI datasets. Therefore, the objective is to develop a quick and trustworthy classification system that can sustain the best performance over a comprehensive MRI dataset. This paper presents a robust approach that has the ability to analyze and classify different types of brain diseases using MRI images. In this paper, global histogram equalization is utilized to remove unwanted details from the MRI images. After the picture has been enhanced, a symlet wavelet transform-based technique has been suggested that can extract the best features from the MRI images for feature extraction. On gray scale images, the suggested feature extraction approach is a compactly supported wavelet with the lowest asymmetry and the most vanishing moments for a given support width. Because the symlet wavelet can accommodate the orthogonal, biorthogonal, and reverse biorthogonal features of gray scale images, it delivers higher classification results. Following the extraction of the best feature, the linear discriminant analysis (LDA) is employed to minimize the feature space's dimensions. The model was trained and evaluated using logistic regression, and it correctly classified several types of brain illnesses based on MRI pictures. To illustrate the importance of the proposed strategy, a standard dataset from Harvard Medical School and the Open Access Series of Imaging Studies (OASIS), which encompasses 24 different brain disorders (including normal), is used. The proposed technique achieved the best classification accuracy of 96.6% when measured against current cutting-edge systems.
Collapse
|