Harnessing Machine Learning to Mitigate Adverse Outcomes of Preterm Birth

Background to the project

We are excited to announce a PhD studentship opportunity in maternal health. This project aims to leverage advanced machine learning techniques to develop predictive models and interventions that could significantly reduce the adverse outcomes associated with preterm birth. Preterm birth remains a critical global health issue, leading to increased neonatal morbidity and mortality, as well as long-term effects. By utilising sophisticated data analytics and machine learning methodologies, this research seeks to identify risk factors and implement timely interventions, thus improving health outcomes for both mothers and their infants.

The successful candidate will engage in a multidisciplinary approach, collaborating with experts in machine learning and obstetrics. The project will involve the collection and analysis of a diverse range of data, including clinical records, demographic information and biological markers, to develop robust predictive algorithms. Through this, we aim to provide healthcare providers with actionable insights that can inform clinical decisions and patient management strategies, ultimately working towards reducing the incidence and impact of preterm birth.

We are looking for a motivated individual with a strong background in machine learning, data analysis, or a related biomedical field, who is passionate about making a difference in maternal and child health. The candidate will have the opportunity to publish findings in top-tier journals and present at conferences, thereby establishing themselves as a leader in this important area of research. If you are eager to contribute to meaningful advancements in healthcare and improve outcomes for vulnerable populations, we encourage you to apply for this exciting PhD studentship.

What the studentship will encompass

During the PhD studentship, the candidate will engage in a variety of activities designed to build expertise, conduct impactful research, and contribute to advancements in maternal and child health. The key components will include:

1.   Literature Review and Research Design: The candidate will begin by conducting a comprehensive literature review to understand current challenges and gaps in preterm birth research. This will involve exploring existing predictive models (such as Quipp), identifying key risk factors, and examining technological innovations in machine learning. Based on this review, the candidate will design their research methodology, including defining specific research questions and selecting appropriate machine learning techniques.

2.   Data Collection and Preprocessing: The PhD student will collaborate with healthcare professionals and institutions to collect relevant data, which may include clinical records, demographic information, physiological metrics, and biological samples. They will learn how to preprocess this data, ensuring it is clean, structured, and suitable for analysis. This step is crucial for the development of reliable machine learning models.

3.   Model Development and Validation: Using advanced machine learning frameworks, the student will develop predictive algorithms that aim to identify at-risk populations and forecast preterm birth outcomes. The candidate will experiment with various modelling techniques such as regression analysis, decision trees, support vector machines, or deep learning, and will rigorously validate these models using statistical methods to assess their accuracy and reliability.

4.   Collaboration and Interdisciplinary Work: The student will work closely with interdisciplinary teams, including clinicians, data scientists, and public health experts, to ensure that the research is grounded in real-world clinical relevance. Regular meetings and discussions will facilitate knowledge exchange and may lead to novel insights that enhance impact.

5.   Translation of Findings: An important aspect of the studentship will be the translation of findings into practical applications. The candidate will explore how the developed models can be embedded into clinical practice, such as through decision-support tools for healthcare providers to identify and manage at-risk patients.

6.   Dissemination and Impact: Throughout the studentship, the candidate will contribute to academic publications, prepare conference presentations, and engage with the broader community to disseminate findings. They will gain experience in scientific writing and presentation skills, aiming to influence policy and practice.

By the end of the PhD, the candidate is expected to have developed a robust understanding of both machine learning and its applications in the field of obstetrics, and to reduce the risks and complications associated with preterm birth.

Details of supervision

 

The project will be supervised by Professor Steve Thornton, Professor Iliodromati and Dr Elena Greco (academic obstetricians), from Queen Mary University of London. Professor Andrew Shennan (academic obstetrician: Kings College) who has experience of the quip app will provide oversight. Dr Mike Allen (academic lecturer; University of Exeter) will oversee the machine learning. Occasional visits may be required to Exeter, but the majority of the work will be based in London.

Requirements:

 

Candidates should have a strong academic background, ideally with a master’s degree in fields such as computer science, data science, biomedicine, or a related discipline. Candidates without a Masters will be considered. Proficiency in programming languages such as Python or R, along with a solid understanding of machine learning algorithms and statistical analysis would be an advantage. Experience in data manipulation and analysis, particularly in healthcare datasets, will be helpful. Additionally, candidates should demonstrate strong analytical and problem-solving skills, along with the ability to work collaboratively in interdisciplinary teams. Excellent communication skills are crucial for conveying complex ideas and research findings, both in writing and verbally. A genuine interest in maternal health, as well as a commitment to improving health outcomes, will be key attributes.

How to apply

The deadline for applications is 17:00 (GMT) 16^th of September 2024. Late applications will not be considered. Interviews will be scheduled for mid October 2024.

To be considered for this PhD, please apply through our website here, and submit:

1)     CV

2)     Personal statement (1pg max) – this should include: why you are interested in undertaking this project; what relevant existing skills, training, and knowledge you would bring to the project.