Research Assistant or Research Associate in Computer Vision and Machine Learning

Job description

We are looking for a candidate who has a strong background in deep learning and/or biomedical imaging algorithm development and interest to employ such techniques to advanced applications in biology and digital pathology. The specific research project requires expertise in modern methods for image segmentation and classification of large imaging data using convolutional neural networks. Experience in web-based application development is desirable.  

This post may appeal to a recent computer vision PhD interested in now developing skills and experience in healthcare research. Candidates with good experience in machine learning and visualisation techniques will also be considered. The successful candidate will be joining a new group and thus this position provides an opportunity for the right candidate to be part of an exciting new venture. 

Artificial intelligence is poised to transform conventional histopathology. Developing validated machine learning tools that can assist histopathologists in areas of diagnostic uncertainty is a fundamental translational research challenge. The key bottleneck in developing these automated classifiers is the need for massive amounts of expert-annotated imaging data to systematically train and validate clinical-grade image classifications systems. The purpose of this project is to develop neural networks to assist early detection of cancer from endoscopic images in high-risk patients.  

This is a highly collaborative project where the applicant will work with clinicians at the Institute of Cancer Institute (London) and Manchester University. While having prior experience in working on interdisciplinary projects would be an advantage it is not a requirement.  

This post will be offered on an a fixed-term contract for 1 year which can be extended

Key responsibilities

1. Manage own research and administrative activities, within guidelines provided by senior colleagues 

2. Select, follow, and adapt experimental protocols

3. Gather, analyse, and present scientific data from a variety of sources

4. Develop methods for handling highly noisy videos.

5. Develop deep learning algorithms for classification of large biomedical images.

6. Utilise the set of image-derived measurements to estimate the underlying variability and work towards characterising different patient phenotypes.

7. Contribute to scientific reports and journal articles and the presentation of data/papers at conferences

8. Work in close collaboration with the clinicians to validate the developed methods.

The above list of responsibilities may not be exhaustive, and the post holder will be required to undertake such tasks and responsibilities as may reasonably be expected within the scope and grading of the post.  

Skills, knowledge, and experience 

The candidate should have good knowledge and experience in developing deep learning methods and handling large-scale real world image data.

Essential criteria  

1.     Have (or be near to completion of) a PhD (Grade 6) or MSc (Grade 5) in medical imaging, computer vision, biomedical engineering, computer science or another related area.  

2.     Excellent programming skills, e.g. in Python, Java and C++.  

3.     Excellent communication skills, both written and oral, including the ability to write for publication, present research proposals and results, and represent the research group at meetings 

4.     Good understanding of software testing 

5.     Demonstrate a strong interest in interdisciplinary research 

6.     Ability to manage own academic research and associated activities 

7.     Ability to contribute ideas for new research projects and research income generation 

Desirable criteria

1. Experience of contributing to reports and articles for publication

2. Strong interest in biomedical applications.

3. Experience of working in a research team and contributing ideas for new research projects

4. Experience in large-scale image-based phenotyping in the wider sense.

5. Published research in a relevant field in high profile journals

6. Experience of developing web applications using advanced javascript libraries is desirable but not required