PhD Position F/M -- Campagne Doctorant -- Explainable Deep Learning Integration of Computational Pathology and Spatial Transcriptomics.
Contract type : Fixed-term contract
Level of qualifications required : Graduate degree or equivalent
Fonction : PhD Position
Level of experience : Recently graduated
Context
Objective of the thesis
This thesis project aims to establish the foundations of a next-generation computational pathomics dedicated to personalized medicine. This innovative and rich approach stems from the explanatory integration of deep learning in computational histopathology as well as spatial transcriptomics.
The proposed modeling involves studying the multi-omic spatial vocabulary as input modeling entities, along with their corresponding parameters, characteristics, connection capabilities, and functional methods. A given pathology exhibits a well-known expression as an inhibited omic signature, which constitutes the system's signature. Consolidated by spatial transcriptomics (ST), these indices will be used to explore potential correlations with phenotypic markers, based on anatomopathological semiology and familiar patterns from existing semantic repositories.
As for causality, it entails explaining and interpreting the correlations and causalities identified. Data-driven AI approaches guided by models will initiate a well-known virtuous cycle process ranging from explanation (as a capability of AI tools) to interpretation (related to the understanding process by experts in the targeted area), and moving towards association, correlation, and causality.
The designed and elaborated models will serve as the basis for simulating and predicting tissue evolution from a single whole slide image. As mentioned earlier, a WSI should suffice to transform computational pathology from a static paradigm to a dynamic one. By considering the spatially resolved transcriptomic profile, we will thus be able to create and simulate dynamic models, which - also incorporating epigenetics (exposome) - will act as prediction support through simulations.
Introducing such simulation and modeling support in clinics could revolutionize prediction in the context of personalized precision medicine, altering the way multidisciplinary clinical meetings are conducted, improving cross-expertise (second opinion) among practitioners, and strengthening the doctor-patient relationship.
To our knowledge, the integration and exploitation of whole slide images and spatial transcriptomic data have not yet been achieved. Furthermore, studying the cross-correlations between the patterns of these two modalities as dependency models represents a clear potential for a new computation-enhanced modality capable of revolutionizing precision medicine.
The incorporation of explainable AI, cutting-edge virtual slide analysis, and semantics is likely to foster biomedical adoption of this highly promising modern technology. Deep learning methods are highly effective when targeted and precisely utilized. Incorporating explainable mechanisms will thus strengthen the expertise of biomedical professionals. Lastly, semantics can also contribute to creating a knowledgeable framework capable of supporting research, diagnosis, and prognosis.
Assignment
Collaboration :
The recruited person will be in connection with Lev Stimmer concerning the Spatial Transcriptomics technology in ICM.
Responsibilities :
The person recruited is responsible for the creation of the new Computational Pathomics modality and will take initiatives for the XAI deep learning approaches to be used to explore the associations, correlations and the causalities between morhology and the spatial transcriptomics signature. She/he will be in charge of a robust learnng process to be designed, validated and tested on a consequent dataset. Seed information shall be identified and a bias-free process shall be set-up.
Steering/Management :
The person recruited will be in charge of the deployment of the proposed technology on the Cytomine @ ICM platform.
Main activities
Main activities :
- Identify and verify omics and visual vocabulary solutions for spatial transcriptomics and computational pathology.
- Propose XAI solutions for association / correlation / causality relationship exploration / discovery between computational pathology (morphology) and spatial transcriptomics.
- Test, change up until validation the XAI solutions
- Design experimental platform of virtual spatial transcriptomics
- Develop programs/applications/interfaces of compuational pathomics on Cytomine @ ICM
Additional activities :
- Structure and write publications to disseminate the novel results and methods
- Distribute the vistual spatial transcriptomics to the ICM and national / european scientific community via the Cytomine @ ICM platform.
Skills
Technical Skills and Proficiency Level Required:
- Proficient hands-on experience in deep learning and Python programming, including expertise with frameworks such as PyTorch.
- Advanced knowledge in Machine Learning and Pattern Recognition.
- Expertise in BioMedical Image Analysis.
- Strong foundation in applied mathematics and computer science.
- Familiarity with biomedical imaging techniques.
Language Proficiency:
- English: Expert level proficiency in both oral and written communication.
- French: Good proficiency level suitable for both oral and written communication.
Relational Skills:
- Strong oral communication skills for scientific discourse.
- Demonstrated ability in written communication through publications.
Other Valued Attributes:
- Proven ability to explore and devise solutions to complex problems.
- Collaborative mindset with a strong inclination towards teamwork.
- Positive and constructive attitude towards challenges and collaborations.
- Curiosity and eagerness to delve into medical applications.
- Willingness to learn more about neurodegenerative diseases.
Benefits package
- Subsidized meals
- Partial reimbursement of public transport costs
- Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
- Possibility of teleworking
- Flexible organization of working hours (after 6 months of employment)
- Professional equipment available (videoconferencing, loan of computer equipment, etc.)
- Social, cultural and sports events and activities
- Access to vocational training
- Social security coverage
Remuneration
According to civil service salary scales
General Information
- Theme/Domain :
Computational Neuroscience and Medicine
Scientific computing (BAP E) - Town/city : Paris
- Inria Center : Centre Inria de Paris
- Starting date : 2024-10-01
- Duration of contract : 3 years
- Deadline to apply : 2024-05-19
Warning : you must enter your e-mail address in order to save your application to Inria. Applications must be submitted online on the Inria website. Processing of applications sent from other channels is not guaranteed.
Instruction to apply
In your application (which can be in English or in French), please include:
- CV
- Letter of motivation
- Letters of recommendation
- Master's grades
Defence Security :
This position is likely to be situated in a restricted area (ZRR), as defined in Decree No. 2011-1425 relating to the protection of national scientific and technical potential (PPST).Authorisation to enter an area is granted by the director of the unit, following a favourable Ministerial decision, as defined in the decree of 3 July 2012 relating to the PPST. An unfavourable Ministerial decision in respect of a position situated in a ZRR would result in the cancellation of the appointment.
Recruitment Policy :
As part of its diversity policy, all Inria positions are accessible to people with disabilities.
Contacts
- Inria Team : ARAMIS
-
PhD Supervisor :
Racoceanu Daniel / daniel.racoceanu@inria.fr
The keys to success
We are seeking a collaborative partner who embodies a vibrant blend of the following attributes:
- Passion and Drive: A fervent desire to effect change through research and embrace novel challenges with a focus on healthcare and societal impact.
- Area of Expertise: Proficiency in computer science, machine learning, deep learning, and applied mathematics.
- Personal Attributes: Demonstrates an open, positive, and constructive demeanor, with a strong emphasis on fostering human connections. A team player by nature, driven by curiosity, and deeply appreciates the potential impact of research endeavors.
- Interdisciplinary Proficiency: Possesses cross-disciplinary knowledge and expertise, particularly in biomedical image analysis and its applications in healthcare.
This section serves to complement and streamline the formal list of skills:
- Essential qualities for excelling in this role include comfort within a dynamic scientific environment and a genuine eagerness to learn and collaborate.
- Enthusiastic about innovation, proficient in Python / PyTorch development, and adept at influencing others positively. Having completed a thesis focused on artificial intelligence, deep learning, and/or biomedical image analysis is highly advantageous.
About Inria
Inria is the French national research institute dedicated to digital science and technology. It employs 2,600 people. Its 200 agile project teams, generally run jointly with academic partners, include more than 3,500 scientists and engineers working to meet the challenges of digital technology, often at the interface with other disciplines. The Institute also employs numerous talents in over forty different professions. 900 research support staff contribute to the preparation and development of scientific and entrepreneurial projects that have a worldwide impact.