Changes for page Methodology

Last modified by manuelmenendez on 2025/03/14 08:31
From version 4.3
edited by manuelmenendez
on 2025/01/29 19:11
Change comment: There is no comment for this version
To version 6.1
edited by manuelmenendez
on 2025/02/01 11:57
Change comment: There is no comment for this version

Summary

Details

Page properties
Content
... ... @@ -1,6 +1,6 @@
1 -=== **Overview** ===
1 +==== **Overview** ====
2 2  
3 -This section describes the step-by-step process used in the **Neurodiagnoses** project to develop a novel diagnostic framework for neurological diseases. The methodology integrates artificial intelligence (AI), biomedical ontologies, and computational neuroscience to create a structured, interpretable, and scalable diagnostic system.
3 +This project develops a **tridimensional diagnostic framework** for **CNS diseases**, incorporating **AI-powered annotation tools** to improve **interpretability, standardization, and clinical utility**. The methodology integrates **multi-modal data**, including **genetic, neuroimaging, neurophysiological, and biomarker datasets**, and applies **machine learning models** to generate **structured, explainable diagnostic outputs**.
4 4  
5 5  ----
6 6  
... ... @@ -8,102 +8,166 @@
8 8  
9 9  ==== **Data Sources** ====
10 10  
11 -* **Biomedical Ontologies**:
12 -** Human Phenotype Ontology (HPO) for phenotypic abnormalities.
13 -** Gene Ontology (GO) for molecular and cellular processes.
14 -* **Neuroimaging Datasets**:
15 -** Example: Alzheimer’s Disease Neuroimaging Initiative (ADNI), OpenNeuro.
16 -* **Clinical and Biomarker Data**:
17 -** Anonymized clinical reports, molecular biomarkers, and test results.
11 +**Biomedical Ontologies & Databases:**
18 18  
13 +* **Human Phenotype Ontology (HPO)** for symptom annotation.
14 +* **Gene Ontology (GO)** for molecular and cellular processes.
19 19  
20 -==== **Data Preprocessing** ====
16 +**Dimensionality Reduction and Interpretability:**
21 21  
22 -1. **Standardization**: Ensure all data sources are normalized to a common format.
23 -1. **Feature Selection**: Identify relevant features for diagnosis (e.g., biomarkers, imaging scores).
24 -1. **Data Cleaning**: Handle missing values and remove duplicates.
18 +* **Evaluate interpretability** using metrics like the **Area Under the Interpretability Curve (AUIC)**.
19 +* **Leverage DEIBO (Data-driven Embedding Interpretation Based on Ontologies)** to connect model dimensions to ontology concepts.
25 25  
21 +**Neuroimaging & EEG/MEG Data:**
22 +
23 +* **MRI volumetric measures** for brain atrophy tracking.
24 +* **EEG functional connectivity patterns** (AI-Mind).
25 +
26 +**Clinical & Biomarker Data:**
27 +
28 +* **CSF biomarkers** (Amyloid-beta, Tau, Neurofilament Light).
29 +* **Sleep monitoring and actigraphy data** (ADIS).
30 +
31 +**Federated Learning Integration:**
32 +
33 +* **Secure multi-center data harmonization** (PROMINENT).
34 +
26 26  ----
27 27  
37 +==== **Annotation System for Multi-Modal Data** ====
38 +
39 +To ensure **structured integration of diverse datasets**, **Neurodiagnoses** will implement an **AI-driven annotation system**, which will:
40 +
41 +* **Assign standardized metadata tags** to diagnostic features.
42 +* **Provide contextual explanations** for AI-based classifications.
43 +* **Track temporal disease progression annotations** to identify long-term trends.
44 +
45 +----
46 +
28 28  === **2. AI-Based Analysis** ===
29 29  
30 -==== **Model Development** ====
49 +==== **Machine Learning & Deep Learning Models** ====
31 31  
32 -* **Embedding Models**: Use pre-trained models like BioBERT or BioLORD for text data.
33 -* **Classification Models**:
34 -** Algorithms: Random Forest, Support Vector Machines (SVM), or neural networks.
35 -** Purpose: Predict the likelihood of specific neurological conditions based on input data.
51 +**Risk Prediction Models:**
36 36  
37 -==== **Dimensionality Reduction and Interpretability** ====
53 +* **LETHE’s cognitive risk prediction model** integrated into the annotation framework.
38 38  
39 -* Leverage [[DEIBO>>https://drive.ebrains.eu/f/8d7157708cde4b258db0/]] (Data-driven Embedding Interpretation Based on Ontologies) to connect model dimensions to ontology concepts.
40 -* Evaluate interpretability using metrics like the Area Under the Interpretability Curve (AUIC).
55 +**Biomarker Classification & Probabilistic Imputation:**
41 41  
57 +* **KNN Imputer** and **Bayesian models** used for handling **missing biomarker data**.
58 +
59 +**Neuroimaging Feature Extraction:**
60 +
61 +* **MRI & EEG data** annotated with **neuroanatomical feature labels**.
62 +
63 +==== **AI-Powered Annotation System** ====
64 +
65 +* Uses **SHAP-based interpretability tools** to explain model decisions.
66 +* Generates **automated clinical annotations** in structured reports.
67 +* Links findings to **standardized medical ontologies** (e.g., **SNOMED, HPO**).
68 +
42 42  ----
43 43  
44 -=== **3. Diagnostic Framework** ===
71 +=== **3. Diagnostic Framework & Clinical Decision Support** ===
45 45  
46 -==== **Axes of Diagnosis** ====
73 +==== **Tridimensional Diagnostic Axes** ====
47 47  
48 -The framework organizes diagnostic data into three axes:
75 +**Axis 1: Etiology (Pathogenic Mechanisms)**
49 49  
50 -1. **Etiology**: Genetic and environmental risk factors.
51 -1. **Molecular Markers**: Biomarkers such as amyloid-beta, tau, and alpha-synuclein.
52 -1. **Neuroanatomical Correlations**: Results from neuroimaging (e.g., MRI, PET).
77 +* Classification based on **genetic markers, cellular pathways, and environmental risk factors**.
78 +* **AI-assisted annotation** provides **causal interpretations** for clinical use.
53 53  
54 -==== **Recommendation System** ====
80 +**Axis 2: Molecular Markers & Biomarkers**
55 55  
56 -* Suggests additional tests or biomarkers if gaps are detected in the data.
57 -* Prioritizes tests based on clinical impact and cost-effectiveness.
82 +* **Integration of CSF, blood, and neuroimaging biomarkers**.
83 +* **Structured annotation** highlights **biological pathways linked to diagnosis**.
58 58  
85 +**Axis 3: Neuroanatomoclinical Correlations**
86 +
87 +* **MRI and EEG data** provide anatomical and functional insights.
88 +* **AI-generated progression maps** annotate **brain structure-function relationships**.
89 +
59 59  ----
60 60  
61 -=== **4. Computational Workflow** ===
92 +=== **4. Computational Workflow & Annotation Pipelines** ===
62 62  
63 -1. **Data Loading**: Import data from storage (Drive or Bucket).
64 -1. **Feature Engineering**: Generate derived features from the raw data.
65 -1. **Model Training**:
66 -1*. Split data into training, validation, and test sets.
67 -1*. Train models with cross-validation to ensure robustness.
68 -1. **Evaluation**:
69 -1*. Metrics: Accuracy, F1-Score, AUIC for interpretability.
70 -1*. Compare against baseline models and domain benchmarks.
94 +==== **Data Processing Steps** ====
71 71  
96 +**Data Ingestion:**
97 +
98 +* **Harmonized datasets** stored in **EBRAINS Bucket**.
99 +* **Preprocessing pipelines** clean and standardize data.
100 +
101 +**Feature Engineering:**
102 +
103 +* **AI models** extract **clinically relevant patterns** from **EEG, MRI, and biomarkers**.
104 +
105 +**AI-Generated Annotations:**
106 +
107 +* **Automated tagging** of diagnostic features in **structured reports**.
108 +* **Explainability modules (SHAP, LIME)** ensure transparency in predictions.
109 +
110 +**Clinical Decision Support Integration:**
111 +
112 +* **AI-annotated findings** fed into **interactive dashboards**.
113 +* **Clinicians can adjust, validate, and modify annotations**.
114 +
72 72  ----
73 73  
74 -=== **5. Validation** ===
117 +=== **5. Validation & Real-World Testing** ===
75 75  
76 -==== **Internal Validation** ====
119 +==== **Prospective Clinical Study** ====
77 77  
78 -* Test the system using simulated datasets and known clinical cases.
79 -* Fine-tune models based on validation results.
121 +* **Multi-center validation** of AI-based **annotations & risk stratifications**.
122 +* **Benchmarking against clinician-based diagnoses**.
123 +* **Real-world testing** of AI-powered **structured reporting**.
80 80  
81 -==== **External Validation** ====
125 +==== **Quality Assurance & Explainability** ====
82 82  
83 -* Collaborate with research institutions and hospitals to test the system in real-world settings.
84 -* Use anonymized patient data to ensure privacy compliance.
127 +* **Annotations linked to structured knowledge graphs** for improved transparency.
128 +* **Interactive annotation editor** allows clinicians to validate AI outputs.
85 85  
86 86  ----
87 87  
88 88  === **6. Collaborative Development** ===
89 89  
90 -The project is open to contributions from researchers, clinicians, and developers. Key tools include:
134 +The project is **open to contributions** from **researchers, clinicians, and developers**.
91 91  
136 +**Key tools include:**
137 +
92 92  * **Jupyter Notebooks**: For data analysis and pipeline development.
93 -** Example: [[probabilistic imputation>>https://drive.ebrains.eu/f/4f69ab52f7734ef48217/]]
139 +** Example: **probabilistic imputation**
94 94  * **Wiki Pages**: For documenting methods and results.
95 95  * **Drive and Bucket**: For sharing code, data, and outputs.
96 -* **Collaboration with related projects: **For instance: [[//Beyond the hype: AI in dementia – from early risk detection to disease treatment//>>https://www.lethe-project.eu/beyond-the-hype-ai-in-dementia-from-early-risk-detection-to-disease-treatment/]]
142 +* **Collaboration with related projects**:
143 +** Example: **Beyond the hype: AI in dementia – from early risk detection to disease treatment**
97 97  
98 98  ----
99 99  
100 100  === **7. Tools and Technologies** ===
101 101  
102 -* **Programming Languages**: Python for AI and data processing.
103 -* **Frameworks**:
104 -** TensorFlow and PyTorch for machine learning.
105 -** Flask or FastAPI for backend services.
106 -* **Visualization**: Plotly and Matplotlib for interactive and static visualizations.
107 -* **EBRAINS Services**:
108 -** Collaboratory Lab for running Notebooks.
109 -** Buckets for storing large datasets.
149 +==== **Programming Languages:** ====
150 +
151 +* **Python** for AI and data processing.
152 +
153 +==== **Frameworks:** ====
154 +
155 +* **TensorFlow** and **PyTorch** for machine learning.
156 +* **Flask** or **FastAPI** for backend services.
157 +
158 +==== **Visualization:** ====
159 +
160 +* **Plotly** and **Matplotlib** for interactive and static visualizations.
161 +
162 +==== **EBRAINS Services:** ====
163 +
164 +* **Collaboratory Lab** for running Notebooks.
165 +* **Buckets** for storing large datasets.
166 +
167 +----
168 +
169 +=== **Why This Matters** ===
170 +
171 +* **The annotation system ensures that AI-generated insights are structured, interpretable, and clinically meaningful.**
172 +* **It enables real-time tracking of disease progression across the three diagnostic axes.**
173 +* **It facilitates integration with electronic health records and decision-support tools, improving AI adoption in clinical workflows.**