Wiki source code of Methodology

Version 15.1 by manuelmenendez on 2025/02/09 10:08

version	line-number	content
15.1	1	== Overview ==
1.1	2
6.1	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.
1.1	4
15.1	5	== Workflow ==
7.1	6
	7	1. (((
12.1	8	We Use GitHub to [[Store and develop AI models, scripts, and annotation pipelines.>>https://github.com/Fundacion-de-Neurociencias/neurodiagnoses/discussions]]
7.1	9
	10	* Create a GitHub repository for AI scripts and models.
	11	* Use GitHub Projects to manage research milestones.
	12	)))
	13	1. (((
	14	We Use EBRAINS for Data & Collaboration
	15
	16	* Store biomarker and neuroimaging data in EBRAINS Buckets.
	17	* Run Jupyter Notebooks in EBRAINS Lab to test AI models.
	18	* Use EBRAINS Wiki for structured documentation and research discussion.
	19	)))
	20
1.1	21	----
	22
15.1	23	== 1. Data Integration ==
1.1	24
14.1	25	=== EBRAINS Medical Informatics Platform (MIP). ===
12.2	26
	27	Neurodiagnoses integrates clinical data via the EBRAINS Medical Informatics Platform (MIP). MIP federates decentralized clinical data, allowing Neurodiagnoses to securely access and process sensitive information for AI-based diagnostics.
	28
14.1	29	==== How It Works ====
12.2	30
	31
	32	1. (((
	33	Authentication & API Access:
	34
	35	* Users must have an EBRAINS account.
	36	* Neurodiagnoses uses secure API endpoints to fetch clinical data (e.g., from the Federation for Dementia).
	37	)))
	38	1. (((
	39	Data Mapping & Harmonization:
	40
	41	* Retrieved data is normalized and converted to standard formats (.csv, .json).
	42	* Data from multiple sources is harmonized to ensure consistency for AI processing.
	43	)))
	44	1. (((
	45	Security & Compliance:
	46
	47	* All data access is logged and monitored.
	48	* Data remains on MIP servers using federated learning techniques when possible.
	49	* Access is granted only after signing a Data Usage Agreement (DUA).
	50	)))
	51
14.1	52	==== Implementation Steps ====
12.2	53
	54
	55	1. Clone the repository.
	56	1. Configure your EBRAINS API credentials in mip_integration.py.
	57	1. Run the script to download and harmonize clinical data.
	58	1. Process the data for AI model training.
	59
	60	For more detailed instructions, please refer to the [[MIP Documentation>>url:https://mip.ebrains.eu/]].
	61
	62	----
	63
14.1	64	=== Data Processing & Integration with Clinica.Run ===
12.2	65
	66	Neurodiagnoses now supports Clinica.Run, an open-source neuroimaging platform designed for multimodal data processing and reproducible neuroscience workflows.
	67
14.1	68	==== How It Works ====
12.2	69
	70
	71	1. (((
	72	Neuroimaging Preprocessing:
	73
	74	* MRI, PET, EEG data is preprocessed using Clinica.Run pipelines.
	75	* Supports longitudinal and cross-sectional analyses.
	76	)))
	77	1. (((
	78	Automated Biomarker Extraction:
	79
	80	* Standardized extraction of volumetric, metabolic, and functional biomarkers.
	81	* Integration with machine learning models in Neurodiagnoses.
	82	)))
	83	1. (((
	84	Data Security & Compliance:
	85
	86	* Clinica.Run operates in compliance with GDPR and HIPAA.
	87	* Neuroimaging data remains within the original storage environment.
	88	)))
	89
14.1	90	==== Implementation Steps ====
12.2	91
	92
	93	1. Install Clinica.Run dependencies.
	94	1. Configure your Clinica.Run pipeline in clinica_run_config.json.
	95	1. Run the pipeline for preprocessing and biomarker extraction.
	96	1. Use processed neuroimaging data for AI-driven diagnostics in Neurodiagnoses.
	97
	98	For further information, refer to [[Clinica.Run Documentation>>url:https://clinica.run/]].
	99
	100	==== ====
	101
1.1	102	==== Data Sources ====
	103
12.2	104	[[List of potential sources of databases>>https://github.com/Fundacion-de-Neurociencias/neurodiagnoses/blob/main/data/sources/list_of_potential_databases]]
	105
6.1	106	Biomedical Ontologies & Databases:
1.1	107
6.1	108	* Human Phenotype Ontology (HPO) for symptom annotation.
	109	* Gene Ontology (GO) for molecular and cellular processes.
4.2	110
6.1	111	Dimensionality Reduction and Interpretability:
1.1	112
6.1	113	* Evaluate interpretability using metrics like the Area Under the Interpretability Curve (AUIC).
11.1	114	* Leverage [[DEIBO>>https://github.com/Mellandd/DEIBO]] (Data-driven Embedding Interpretation Based on Ontologies) to connect model dimensions to ontology concepts.
1.1	115
6.1	116	Neuroimaging & EEG/MEG Data:
	117
	118	* MRI volumetric measures for brain atrophy tracking.
	119	* EEG functional connectivity patterns (AI-Mind).
	120
	121	Clinical & Biomarker Data:
	122
	123	* CSF biomarkers (Amyloid-beta, Tau, Neurofilament Light).
	124	* Sleep monitoring and actigraphy data (ADIS).
	125
	126	Federated Learning Integration:
	127
	128	* Secure multi-center data harmonization (PROMINENT).
	129
1.1	130	----
	131
6.1	132	==== Annotation System for Multi-Modal Data ====
	133
	134	To ensure structured integration of diverse datasets, Neurodiagnoses will implement an AI-driven annotation system, which will:
	135
	136	* Assign standardized metadata tags to diagnostic features.
	137	* Provide contextual explanations for AI-based classifications.
	138	* Track temporal disease progression annotations to identify long-term trends.
	139
	140	----
	141
15.1	142	== 2. AI-Based Analysis ==
1.1	143
6.1	144	==== Machine Learning & Deep Learning Models ====
1.1	145
6.1	146	Risk Prediction Models:
1.1	147
6.1	148	* LETHE’s cognitive risk prediction model integrated into the annotation framework.
1.1	149
6.1	150	Biomarker Classification & Probabilistic Imputation:
1.1	151
6.1	152	* KNN Imputer and Bayesian models used for handling missing biomarker data.
	153
	154	Neuroimaging Feature Extraction:
	155
	156	* MRI & EEG data annotated with neuroanatomical feature labels.
	157
	158	==== AI-Powered Annotation System ====
	159
	160	* Uses SHAP-based interpretability tools to explain model decisions.
	161	* Generates automated clinical annotations in structured reports.
	162	* Links findings to standardized medical ontologies (e.g., SNOMED, HPO).
	163
1.1	164	----
	165
15.1	166	== 3. Diagnostic Framework & Clinical Decision Support ==
1.1	167
6.1	168	==== Tridimensional Diagnostic Axes ====
1.1	169
6.1	170	Axis 1: Etiology (Pathogenic Mechanisms)
1.1	171
6.1	172	* Classification based on genetic markers, cellular pathways, and environmental risk factors.
	173	* AI-assisted annotation provides causal interpretations for clinical use.
1.1	174
6.1	175	Axis 2: Molecular Markers & Biomarkers
1.1	176
6.1	177	* Integration of CSF, blood, and neuroimaging biomarkers.
	178	* Structured annotation highlights biological pathways linked to diagnosis.
1.1	179
6.1	180	Axis 3: Neuroanatomoclinical Correlations
	181
	182	* MRI and EEG data provide anatomical and functional insights.
	183	* AI-generated progression maps annotate brain structure-function relationships.
	184
1.1	185	----
	186
15.1	187	== 4. Computational Workflow & Annotation Pipelines ==
1.1	188
6.1	189	==== Data Processing Steps ====
1.1	190
6.1	191	Data Ingestion:
	192
	193	* Harmonized datasets stored in EBRAINS Bucket.
	194	* Preprocessing pipelines clean and standardize data.
	195
	196	Feature Engineering:
	197
	198	* AI models extract clinically relevant patterns from EEG, MRI, and biomarkers.
	199
	200	AI-Generated Annotations:
	201
	202	* Automated tagging of diagnostic features in structured reports.
	203	* Explainability modules (SHAP, LIME) ensure transparency in predictions.
	204
	205	Clinical Decision Support Integration:
	206
	207	* AI-annotated findings fed into interactive dashboards.
	208	* Clinicians can adjust, validate, and modify annotations.
	209
1.1	210	----
	211
15.1	212	== 5. Validation & Real-World Testing ==
1.1	213
6.1	214	==== Prospective Clinical Study ====
1.1	215
6.1	216	* Multi-center validation of AI-based annotations & risk stratifications.
	217	* Benchmarking against clinician-based diagnoses.
	218	* Real-world testing of AI-powered structured reporting.
1.1	219
6.1	220	==== Quality Assurance & Explainability ====
1.1	221
6.1	222	* Annotations linked to structured knowledge graphs for improved transparency.
	223	* Interactive annotation editor allows clinicians to validate AI outputs.
1.1	224
	225	----
	226
15.1	227	== 6. Collaborative Development ==
1.1	228
6.1	229	The project is open to contributions from researchers, clinicians, and developers.
1.1	230
6.1	231	Key tools include:
	232
1.1	233	* Jupyter Notebooks: For data analysis and pipeline development.
6.1	234	Example: probabilistic imputation**
1.1	235	* Wiki Pages: For documenting methods and results.
	236	* Drive and Bucket: For sharing code, data, and outputs.
6.1	237	* Collaboration with related projects:
	238	Example: Beyond the hype: AI in dementia – from early risk detection to disease treatment**
1.1	239
	240	----
	241
15.1	242	== 7. Tools and Technologies ==
1.1	243
6.1	244	==== Programming Languages: ====
	245
	246	* Python for AI and data processing.
	247
	248	==== Frameworks: ====
	249
	250	* TensorFlow and PyTorch for machine learning.
	251	* Flask or FastAPI for backend services.
	252
	253	==== Visualization: ====
	254
	255	* Plotly and Matplotlib for interactive and static visualizations.
	256
	257	==== EBRAINS Services: ====
	258
	259	* Collaboratory Lab for running Notebooks.
	260	* Buckets for storing large datasets.
	261
	262	----
	263
	264	=== Why This Matters ===
	265
12.1	266	* The annotation system ensures that AI-generated insights are structured, interpretable, and clinically meaningful.
	267	* It enables real-time tracking of disease progression across the three diagnostic axes.
	268	* It facilitates integration with electronic health records and decision-support tools, improving AI adoption in clinical workflows.

Wiki source code of Methodology

Neurodiagnoses