Two-year Postdoc position in data sciences in healthcare (funded by Amidex)

Karlstad University; Marseille; Sur place; EUR 60 000 - 80 000; Plein temps

Résumé du poste

Aix-Marseille Université offre un poste de postdoctorant en sciences des données dans le domaine de la santé. Le candidat travaillera sur l'analyse des données des patients traités par chimiothérapie pour développer un score de risque associé à l'insuffisance rénale aiguë. Ce rôle nécessite un doctorat en médecine ou dans un domaine lié avec des compétences avancées en science des données et une aptitude à travailler dans un contexte multidisciplinaire.

Prestations

Bureau et poste de travail fournis

Accès au département de néphrologie

Qualifications

• Compétences en sciences des données en santé, machine learning, IA.
• Capacité à traiter des données biomédicales et cliniques.
• Autonomie et compétences en travail d'équipe.

Responsabilités

• Stratifier et décrire la population de patients traités par chimiothérapie.
• Identifier les facteurs de risque pour le développement de l'AKI.
• Construire un score de risque composite testable en pratique clinique.

Connaissances

Machine Learning

Data Science

Epidemiology computation

Formation

Doctorat en informatique ou mathématiques (statistiques et science des données)

Doctorat en médecine

Description du poste

RESEARCHER PROFILE : Postdoc / R2 : PhD holders

RESEARCH FIELD(S)1 : Health data sciences; Data science in healthcare

JOB / OFFER DESCRIPTION

The aim of the A(K)I for K project is to work on the population of patients treated with chemotherapy who develop acute kidney injury (AKI) as an adverse event. The prognosis is then considerably worse.

In this context, we wish to explore data from the AP-HP Data Warehouse in order to (1) establish a risk score for the occurrence of AKI in patients treated with chemotherapy (2) propose a personalized treatment should this event occur.

The post-doctoral fellow will first have to describe and stratify the study population in order to identify and characterize the population treated with chemotherapy, by type of cancer, according to whether or not they experienced AKI during treatment.

This descriptive stage involves grouping patients into sub-cohorts, according 3 dimensions :

• Presence of an ICD-10 cancer code (starting with C or D)
• Presence of an ICD-10 code for Acute Renal Failure (N171, N172, N176, N179, R392)
• Presence of a CCAM code for hemodialysis (JVJF002, JVJF004, JVJF005, JVRP004)

Each sub-cohort (defined by a combination of codes on the 3 axes) will be described by demographic indicators (overall number of patients, number of patients by sex, average age, percentage of deaths at the point date, etc.), as well as by the presence of comorbidities. Specific attention will be paid to verifying codes (where possible) on the basis of biological analysis results, or with data extracted by NLP from textual documents.

Sub-cohorts will then be selected. Concretely, it means that criteria will be established on the basis of the discriminating variables identified in the previous step, in order to establish levels of risk.

Once the population has been stratified, the postdoc will work on identifying risk factors for the development of AKI following cancer treatment with chemotherapy.

Similarly, the postdoc will also have to work on identifying protective factors. Today, there is no curative treatment for ARF, so a preventive approach is essential.

Once the risk and protective factors have been identified, the next step will be to construct a composite risk score, taking into account both risk and protective factors.

This risk score is intended to be tested in clinical practice. Various tests and evaluations will be carried out to measure its performance, reliability, accuracy and impact on patient management by clinicians.

Study of the positive and negative predictive value in a prospective cohort using data from APHP research.

Clinical test within the GHU perimeter, in collaboration with the Sorbonne University Cancer Institute, and in close collaboration with the “Failure to Treat Cancer” research chair being set up at the University Cancer Institute.

Collaborations with digital start-ups via the UNIREIN third-party experimentation center may be considered.

The other part of the work involves identifying possible management options for patients who develop AKI.

A new stratification will be established on the basis of the favorable or unfavorable prognosis of patients who develop acute renal failure during chemotherapy treatment of their cancer.

Factors predictive of vital prognosis will be identified in order to establish recommendations.

The postdoc will work directly on the AP-HP DataWarehouse, either from a workstation at La Pitié Salpêtrière Hospital, Paris (AP-HP), or remotely from a workstation at ESIEE Paris, Université Gustave Eiffel, Noisy-le-Grand.

TYPE OF CONTRACT : TEMPORARY

JOB STATUS : FULL TIME

HOURS PER WEEK : 35

APPLICATION DEADLINE : 01 / 10 / 2025; 09 : 00 am

ENVISAGED STARTING DATE : 01 / 10 / 2025

ENVISAGED DURATION : 24 months

JOB NOT FUNDED THROUGH AN EU RESEARCH FRAMEWORK PROGRAMME

WHAT WE OFFER :

An office and a workstation, and a salary (funded by Institut Laënnec), will be provided to the postdoc.

The postdoc will have an office and a computer workstation at ESIEE Paris.

He will also have access to the nephrology department at the Pitié-Salpêtrière Hospital (AP-HP) for exchanges with clinical teams.

QUALIFICATIONS, REQUIRED RESEARCH FIELDS, REQUIRED EDUCATION LEVEL, PROFESSIONAL SKILLS, OTHER RESEARCH REQUIREMENTS (years of research experience

The candidate must hold either a doctorate in computer science or mathematics (statistics and data science) with an interest and experience in the medical field, or a doctorate in medicine, with an interest and experience in artificial intelligence, data science and programming.

They must be qualified in Data Science in Healthcare (machine learning - AI) and have good skills and experience in Information processing on biomedical and clinical data, and, as well, in computational epidemiology.

Soft skills : The candidate must have a particular liking for collaborative work in a multidisciplinary context. They are expected to be autonomous, and be very capable of teamwork. They need to have good skills in writing scientific publications in English.

Textual documents of the Dataware house are in French. An ability to understand French will therefore be a definite advantage.

REQUESTED DOCUMENTS OF APPLICATION, ELIGIBILITY CRITERIA, SELECTION PROCESS

The candidate must submit a CV and covering letter

The candidate will be auditioned by the project team. In this audition, the applicant will be expected to present a coherent scientific project to be implemented to meet the expectations of the A(K)I for K project. The candidate will have to demonstrate a good understanding of the issues (scientific and medical), and propose the implementation of scientific and technical solutions to achieve the expected results. The candidate will need to demonstrate experience in similar tasks.

HOW TO APPLY :