CoMPaSS-NMD
Computational Models for new Patients Stratification Strategies of Neuromuscular Disorders
CoMPaSS-NMD
Computational Models for new Patients Stratification Strategies of Neuromuscular Disorders
Project mission
CoMPaSS-NMD, an acronym for Computational Models for new Patients Stratification Strategies of Neuromuscular Disorders, is a European project funded under the EU programme Horizon Europe programme, started in May 2023.
The project, lasting 4 years, aims to create a new generation of methods for “precision diagnosis” that allow researchers and/or healthcare professionals to successfully classify patients affected by Hereditary Neuromuscular Diseases (HNMDs). Through the application of methods based on the use of Artificial Intelligence (AI) and Machine Learning (ML), the study aims to provide new perspectives in the knowledge of the pathogenetic and transmission mechanisms of HNMDs, leading to a more accurate diagnosis, thus improving the prognosis, which is often difficult to predict in the clinical practice.
Project mission
CoMPaSS-NMD, an acronym for Computational Models for new Patients Stratification Strategies of Neuromuscular Disorders, is a European project funded under the EU programme Horizon Europe programme, started in May 2023.
The project, lasting 4 years, aims to create a new generation of methods for “precision diagnosis” that allow researchers and/or healthcare professionals to successfully classify patients affected by Hereditary Neuromuscular Diseases (HNMDs). Through the application of methods based on the use of Artificial Intelligence (AI) and Machine Learning (ML), the study aims to provide new perspectives in the knowledge of the pathogenetic and transmission mechanisms of HNMDs, leading to a more accurate diagnosis, thus improving the prognosis, which is often difficult to predict in the clinical practice.
Objectives
 |
To generate robust and reliable datasets needed to develop computational tools based on validated data |
|---|---|
 |
To create solutions that integrate different clinical data for the classification of patients with similar clinical characteristics supporting healthcare professionals and the scientific community as well as people potentially affected by hereditary neuromuscular diseases |
 |
To frame evidence-based guidelines to improve the HNMD patient management above the standard of care currently supporting healthcare professionals |
 |
To create a public platform for collecting data of patients affected by HNMD, the ATLAS-NMD, in compliance with the FAIR (Findable, Accessible, Interoperable, Reusable) rules of the European Union |
Objectives
|
To generate robust and reliable datasets needed to develop computational tools based on validated data |
|---|---|
|
To create solutions that integrate different clinical data for the classification of patients with similar clinical characteristics supporting healthcare professionals and the scientific community as well as people potentially affected by hereditary neuromuscular diseases |
|
To frame evidence-based guidelines to improve the HNMD patient management above the standard of care currently supporting healthcare professionals |
|
To create a public platform for collecting data of patients affected by HNMD, the ATLAS-NMD, in compliance with the FAIR (Findable, Accessible, Interoperable, Reusable) rules of the European Union |
Timeline
Timeline
Project progresses
Overcoming Data Fragmentation with a Unified Patient Data Platform
We worked to create a system to collect detailed patient information based on the clinical evaluation: Standard Operational Procedures (SOPs) as a common language to describe diseases, using a special tool called Human Phenotype Ontology (HPO) to make this description more precise.
Moreover, we built a digital system, the electronic Structured Clinical Report Form to gather information from 500 patients. This system helps clinicians to analyse the patient’s condition in detail, understanding all the different aspects of their disease.
We also developed a user-friendly app for tablets that allows healthcare professionals to quickly and easily input new patient information. This app can be used by clinics across Europe to follow our approach and improve patient care.
Building the foundation of the ATLAS platform
We gathered requirements for the platform’s functionality and designed its basic structure, prioritising the development of the clinical data collection module for HNMD patients.
Our internal clinicians tested an initial platform prototype and provided valuable feedback.
We recently held the first co-design workshop with technical and clinical partners, as well as external experts from the CoMPaSS-NMD Scientific Advisory Board, to discuss the platform, review experiences with similar platforms, identify potential risks, and explore long-term sustainability.
Patient Stratification and Disease Signature Identification
After the initial phase of data collection and processing, the project made significant progress in:
- Genetic Data Analysis: Filtering and clustering of genetic data have been completed.
- Imaging and Histological Data Analysis: Pre-processing and clustering of MRI and tissue scans are nearing completion.
- Federated Learning: To protect patient privacy, the team is exploring federated learning techniques, which allow machine learning models to be trained on decentralised data without sharing raw data.
During the next months the project partners will work to:
- Recruit patients to collect data for the validation of the AI algorithms. If interested, read more about the clinical study.
- Implement the ATLAS architecture.
- Define patient superclusters and their multi-omics profiles.
Project progresses
Overcoming Data Fragmentation with a Unified Patient Data Platform
We worked to create a system to collect detailed patient information based on the clinical evaluation: Standard Operational Procedures (SOPs) as a common language to describe diseases, using a special tool called Human Phenotype Ontology (HPO) to make this description more precise.
Moreover, we built a digital system, the electronic Structured Clinical Report Form to gather information from 500 patients. This system helps clinicians to analyse the patient’s condition in detail, understanding all the different aspects of their disease.
We also developed a user-friendly app for tablets that allows healthcare professionals to quickly and easily input new patient information. This app can be used by clinics across Europe to follow our approach and improve patient care.
Building the foundation of the ATLAS platform
We gathered requirements for the platform’s functionality and designed its basic structure, prioritising the development of the clinical data collection module for HNMD patients.
Our internal clinicians tested an initial platform prototype and provided valuable feedback.
We recently held the first co-design workshop with technical and clinical partners, as well as external experts from the CoMPaSS-NMD Scientific Advisory Board, to discuss the platform, review experiences with similar platforms, identify potential risks, and explore long-term sustainability.
Patient Stratification and Disease Signature Identification
After the initial phase of data collection and processing, the project made significant progress in:
- Genetic Data Analysis: Filtering and clustering of genetic data have been completed.
- Imaging and Histological Data Analysis: Pre-processing and clustering of MRI and tissue scans are nearing completion.
- Federated Learning: To protect patient privacy, the team is exploring federated learning techniques, which allow machine learning models to be trained on decentralised data without sharing raw data.
During the next months the project partners will work to:
- Recruit patients to collect data for the validation of the AI algorithms. If interested, read more about the clinical study.
- Implement the ATLAS architecture.
- Define patient superclusters and their multi-omics profiles.