Author name cluster

Malek Adjouadi

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

1 author row

AIIM Journal 2023 Journal Article

A unique color-coded visualization system with multimodal information fusion and deep learning in a longitudinal study of Alzheimer's disease

Mohammad Eslami
Solale Tabarestani
Malek Adjouadi

Purpose Automated diagnosis and prognosis of Alzheimer's Disease remain a challenging problem that machine learning (ML) techniques have attempted to resolve in the last decade. This study introduces a first-of-its-kind color-coded visualization mechanism driven by an integrated ML model to predict disease trajectory in a 2-year longitudinal study. The main aim of this study is to help capture visually in 2D and 3D renderings the diagnosis and prognosis of AD, therefore augmenting our understanding of the processes of multiclass classification and regression analysis. Method The proposed method, Machine Learning for Visualizing AD (ML4VisAD), is designed to predict disease progression through a visual output. This newly developed model takes baseline measurements as input to generate a color-coded visual image that reflects disease progression at different time points. The architecture of the network relies on convolutional neural networks. With 1123 subjects selected from the ADNI QT-PAD dataset, we use a 10-fold cross-validation process to evaluate the method. Multimodal inputs* include neuroimaging data (MRI, PET), neuropsychological test scores (excluding MMSE, CDR-SB, and ADAS to avoid bias), cerebrospinal fluid (CSF) biomarkers with measures of amyloid beta (ABETA), phosphorylated tau protein (PTAU), total tau protein (TAU), and risk factors that include age, gender, years of education, and ApoE4 gene. Findings/results Based on subjective scores reached by three raters, the results showed an accuracy of 0. 82 ± 0. 03 for a 3-way classification and 0. 68 ± 0. 05 for a 5-way classification. The visual renderings were generated in 0. 08 msec for a 23 × 23 output image and in 0. 17 ms for a 45 × 45 output image. Through visualization, this study (1) demonstrates that the ML visual output augments the prospects for a more accurate diagnosis and (2) highlights why multiclass classification and regression analysis are incredibly challenging. An online survey was conducted to gauge this visualization platform's merits and obtain valuable feedback from users. All implementation codes are shared online on GitHub. Conclusion This approach makes it possible to visualize the many nuances that lead to a specific classification or prediction in the disease trajectory, all in context to multimodal measurements taken at baseline. This ML model can serve as a multiclass classification and prediction model while reinforcing the diagnosis and prognosis capabilities by including a visualization platform.

Details DOI

AIIM Journal 2023 Journal Article

Graph neural networks in EEG spike detection

Ahmed Hossam Mohammed
Mercedes Cabrerizo
Alberto Pinzon
Ilker Yaylali
Prasanna Jayakar
Malek Adjouadi

Objective: This study develops new machine learning architectures that are more adept at detecting interictal epileptiform discharges (IEDs) in scalp EEG. A comparison of results using the average precision (AP) metric is made with the proposed models on two datasets obtained from Baptist Hospital of Miami and Temple University Hospital. Methods: Applying graph neural networks (GNNs) on functional connectivity (FC) maps of different frequency sub-bands to yield a novel architecture we call FC-GNN. Attention mechanism is applied on a complete graph to let the neural network select its important edges, hence bypassing the extraction of features, a model we refer to as CA-GNN. Results: On the Baptist Hospital dataset, the results were as follows: Vanilla Self-Attention → 0. 9029 ± 0. 0431, Hierarchical Attention → 0. 8546 ± 0. 0587, Vanilla Visual Geometry Group (VGG) → 0. 92 ± 0. 0618, Satelight → 0. 9219 ± 0. 046, FC-GNN → 0. 9731 ± 0. 0187, and CA-GNN → 0. 9788 ± 0. 0125. In the same order, the results on the Temple University Hospital dataset are 0. 9692, 0. 9113, 0. 97, 0. 9575, 0. 963, and 0. 9879. Conclusion: Based on the good results they yield, GNNs prove to have a strong potential in detecting epileptogenic activity. Significance: This study opens the door for the discovery of the powerful role played by GNNs in capturing IEDs, which is an essential step for identifying the epileptogenic networks of the affected brain and hence improving the prospects for more accurate 3D source localization.

Details DOI

AIIM Journal 2021 Journal Article

Real-time frequency-independent single-Lead and single-beat myocardial infarction detection

Harold Martin
Ulyana Morar
Walter Izquierdo
Mercedes Cabrerizo
Anastasio Cabrera
Malek Adjouadi

This study proposes a novel real-time frequency-independent myocardial infarction detector for Lead II electrocardiograms. The underlying Deep-LSTM network is trained using the PTB-XL database, the largest to date publicly available electrocardiography dataset, and is tested over the same and the older PTB database. By testing the model over distinct datasets, collected under different conditions and from different patients, a more realistic measure of the performance can be gauged from the deployed system. The detector is trained over 3589 myocardial infarction (MI) patients and 7115 healthy controls (HC) while it is evaluated on 1076 MIs and 1840 HCs. The proposed algorithm, achieved an accuracy of 77. 12%, recall/sensitivity of 75. 85%, and a specificity of 83. 02% over the entire PTB database; 85. 07%, 81. 54%, 87. 31% over the PTB-XL validation set (fold 9), and 84. 17%, 78. 37%, 87. 55% over the PTB-XL test set (fold 10). The model also achieves stable performance metrics over the frequency range of 202 Hz to 2. 8 kHz. The processing time is dependent on the sampling frequency, ranging from 130 ms at 202 Hz to 1. 8 s at 2. 8 kHz. Such outcome is within the time required for real-time processing (less than 300 ms for fast heartbeats), between 202 Hz and 500 Hz making the algorithm practically real-time. Therefore, the proposed MI detector could be readily deployed onto existing wearable and/or portable devices and test instruments; potentially having significant societal and clinical impact in the lives of patients at risk for myocardial infarction.

Details DOI

YNIMG Journal 2020 Journal Article

A distributed multitask multimodal approach for the prediction of Alzheimer’s disease in a longitudinal study

Solale Tabarestani
Maryamossadat Aghili
Mohammad Eslami
Mercedes Cabrerizo
Armando Barreto
Naphtali Rishe
Rosie E. Curiel
David Loewenstein

Predicting the progression of Alzheimer's Disease (AD) has been held back for decades due to the lack of sufficient longitudinal data required for the development of novel machine learning algorithms. This study proposes a novel machine learning algorithm for predicting the progression of Alzheimer's disease using a distributed multimodal, multitask learning method. More specifically, each individual task is defined as a regression model, which predicts cognitive scores at a single time point. Since the prediction tasks for multiple intervals are related to each other in chronological order, multitask regression models have been developed to track the relationship between subsequent tasks. Furthermore, since subjects have various combinations of recording modalities together with other genetic, neuropsychological and demographic risk factors, special attention is given to the fact that each modality may experience a specific sparsity pattern. The model is hence generalized by exploiting multiple individual multitask regression coefficient matrices for each modality. The outcome for each independent modality-specific learner is then integrated with complementary information, known as risk factor parameters, revealing the most prevalent trends of the multimodal data. This new feature space is then used as input to the gradient boosting kernel in search for a more accurate prediction. This proposed model not only captures the complex relationships between the different feature representations, but it also ignores any unrelated information which might skew the regression coefficients. Comparative assessments are made between the performance of the proposed method with several other well-established methods using different multimodal platforms. The results indicate that by capturing the interrelatedness between the different modalities and extracting only relevant information in the data, even in an incomplete longitudinal dataset, will yield minimized prediction errors.

Details DOI

YNICL Journal 2019 Journal Article

Free-water imaging of the hippocampus is a sensitive marker of Alzheimer's disease

Edward Ofori
Steven T. DeKosky
Marcelo Febo
Luis Colon-Perez
Paramita Chakrabarty
Ranjan Duara
Malek Adjouadi
Todd E. Golde

Validating sensitive markers of hippocampal degeneration is fundamental for understanding neurodegenerative conditions such as Alzheimer's disease. In this paper, we test the hypothesis that free-water in the hippocampus will be more sensitive to early stages of cognitive decline than hippocampal volume, and that free-water in hippocampus will increase across distinct clinical stages of Alzheimer's disease. We examined two separate cohorts (N = 126; N = 112) of cognitively normal controls, early and late mild cognitive impairment (MCI), and Alzheimer's disease. Demographic, clinical, diffusion-weighted and T1-weighted imaging, and positron emission tomography (PET) imaging were assessed. Results indicated elevated hippocampal free-water in early MCI individuals compared to controls across both cohorts. In contrast, there was no difference in volume of these regions between controls and early MCI. ADNI free-water values in the hippocampus was associated with low CSF AB1–42 levels and high global amyloid PET values. Free-water imaging of the hippocampus can serve as an early stage marker for AD and provides a complementary measure of AD neurodegeneration using non-invasive imaging.

Details DOI

YNIMG Journal 2013 Journal Article

Age association of language task induced deactivation induced in a pediatric population

Binjian Sun
Madison M. Berl
Thomas G. Burns
William D. Gaillard
Laura Hayes
Malek Adjouadi
Richard A. Jones

Task-induced deactivation (TID) potentially reflects the interactions between the default mode and task specific networks, which are assumed to be age dependent. The study of the age association of such interactions provides insight about the maturation of neural networks, and lays out the groundwork for evaluating abnormal development of neural networks in neurological disorders. The current study analyzed the deactivations induced by language tasks in 45 right-handed normal controls aging from 6 to 22years of age. Converging results from GLM, dual regression and ROI analyses showed a gradual reduction in both the spatial extent and the strength of the TID in the DMN cortices as the brain matured from kindergarten to early adulthood in the absence of any significant change in task performance. The results may be ascribed to maturation leading to either improved multi-tasking (i. e. reduced deactivation) or reduced cognitive demands due to greater experience (affects both control and active tasks but leads to reduced overall difference). However, other effects, such as changes in the DMN connectivity that were not included in this study may also have influenced the results. In light of this, researchers should be cautious when investigating the maturation of DMN using TID. With a GLM analysis using the concatenated fMRI data from several paradigms, this study additionally identified an age associated increase of TID in the STG (bilateral), possibly reflecting the role of this area in speech perception and phonological processing.

Details DOI