Research Assistant (PhD candidate) - Ghent University Global Campus, Korea
Ghent University Global Campus
Ghent University is a pluralistic university open to all, regardless of ideological, political, cultural, or social background. Our credo is 'dare to Think'.
As a top 100 university with more than 49,000 students and 15,000 staff members, we are one of the largest universities in the Dutch language area, located in Flanders, Belgium.
Our 11 faculties offer more than 200 courses and conduct in-depth research within a wide range of scientific domains.
Several of our research groups, centers, and institutes are renowned worldwide, in disciplines such as biotechnology, aquaculture, microelectronics, history, ...
In 2017, Ghent University celebrated its 200th anniversary. Over the past 200 years, our university has seen many eminent scientists, ministers, and even Nobel Prize winners among its staff and alumni.
Ghent University Global Campus (GUGC) is the first European university in Songdo, South Korea, offering Bachelor of Science programmes in Molecular Biotechnology, Environmental Technology, and Food Technology.
The main research areas at GUGC are plant biotechnology, biomedical technology, biotech data science, food technology, and environmental technology.
Please visit the home page of Ghent University (http://www.ugent.be/en) and Ghent University Global Campus (http://ghent.ac.kr/ | http://www.ugent.be/globalcampus/en) to learn more about our organizations.
Center for Biosystems and Biotech Data Science (KR01)
At GUGC, the mission of the Center for Biosystems and Biotech Data Science is to pursue the development of new mathematical and computational approaches for complexity reduction of biosystems
and for extracting knowledge from (vast) sets of biotech data. A core technology leveraged by researchers at the center is deep machine learning, targeting the development of innovative concepts, methodologies,
and tools in both the area of molecular biology and the field of computer vision. Furthermore, the Center for Biosystems and Biotech Data Science, which has a headcount of four professors and ten PhD students,
is responsible for organizing nine courses at GUGC (for 65 ECTS), ranging from Informatics to Bioinformatics and Probability & Statistics.
Internet Technology and Data Science Lab (TW06)
The Internet Technology and Data Science Lab (IDLab) is a core research group of imec with research activities embedded in Ghent University and the University of Antwerp. IDLab performs fundamental
and applied research on internet technology and data science, and is, with over 300 researchers, one of the larger research groups at imec. The research areas of IDLab cover machine learning and data mining,
semantic intelligence, and cloud and big data infrastructures (a/o). Graduates of IDLab are currently working at Google DeepMind and Google Brain.
Job Summary
| Full-Time Research Assistant, PhD Candidate - Ghent University Global Campus Department: Department of Environmental Technology, Food Technology, and Molecular Biotechnology (KR01) Department: Department of Electronics and Information Systems (TW06) Department: Human Structure and Repair (GE38) Degree: Master's degree in one of the following disciplines: computer science/engineering, informatics, electrical engineering, applied mathematics, biomedical engineering, or a related field Contract: 1 year + 3 years (contract renewal is conditional on a positive evaluation) Occupancy rate: 100% Vacancy type: Research Assistant (AAP) Last application date: Review of applications will begin on August 31, 2022, and continue until the position is filled Starting date: October 1, 2022 (open to negotiation) Scientific supervisors: Prof. Wesley De Neve, Prof. Wouter Willaert, Dr. Nikdokht Rashidian |
Job Position
Ghent University Global Campus, South Korea, has a vacancy for a Research Assistant (PhD Candidate) in the domain of surgical video analysis, starting from October 1, 2022 (open to negotiation).
This is a 1-year full time position that is renewable one time (upon a favorable evaluation), for a total period of maximum 4 years. The funding for this position primarily comes from the following BOF-funded interdisciplinary research project: "improving PIPAC therapy responses in cancer patients with peritoneal metastases using robust computer vision."
The candidate will work as a Research Assistant at the Center for Biosystems and Biotech Data Science of the Ghent University Global campus in Songdo, Korea.
In addition, the candidate will be able to spend time at the home campus in Ghent during their PhD studies (at IDLab - ELIS and Ghent University Hospital). For non-Korean applicants, student accommodation and a yearly travel budget are foreseen. Ghent University Global Campus is an equal opportunities employer.
The PhD candidate is expected to perform innovative research on the topic of surgical video analysis, with the goal of developing deep machine learning techniques for visual monitoring
of the therapeutic response of peritoneal metastases to chemotherapy. Apart from performing research, the PhD candidate is also expected to help with several lightweight educational activities,
such as supervision of exams, supervision of yearly bachelor/master projects, and supervision of occasional internships (research-driven education).
The targeted doctoral degree is the interdisciplinary degree of Doctor in Computer Science Engineering (main discipline) and Health Sciences (second discipline),
as awarded by the Faculty of Engineering and Architecture (main faculty) and the Faculty of Medicine and Health Sciences (partner faculty) of Ghent University in Belgium.
More information about pursuing a doctoral degree at Ghent University can be found at https://www.ugent.be/en/research/doctoralresearch.
Please note that the Faculty of Engineering and Architecture requires candidates with a non-Belgian Master's degree to undergo a diploma assessment,
requiring the identification of a set of courses from the entire educational curriculum of the candidate that is equivalent to at least 18 ECTS credits of general courses and/or courses related to the main subject
(master dissertation not included) of a Belgian Master of Science Degree in Engineering Technology.
Research Topic
Problem context
Peritoneal carcinomatosis is a rare form of cancer affecting the peritoneum, the thin membrane surrounding the abdominal organs. This form of cancer most often develops
when other abdominal tumors spread to the peritoneum, leading to multiple new tumors on the surface of this membrane.
If a person gets peritoneal carcinomatosis, it generally means that the abdominal cancer is in an advanced stage.
In selected patients with widespread unresectable Peritoneal Metastases (PM), Pressurized Intra-Peritoneal Aerosol Chemotherapy using electrostatic precipitation (ePIPAC) holds considerable promise.
Briefly, ePIPAC combines laparoscopy (keyhole surgery) with intraperitoneal administration of chemotherapy as an aerosol.
Currently, besides the lack of predictive modelling, no valid methodology exists to interpret PM macroscopically, based on laparoscopic pictures and videos.
To score the extensiveness of PM, a widely used Peritoneal Cancer Index (PCI) is available, but this is a manual assessment that is highly subjective and time consuming.
The lack of a computational and reproducible approach towards PM monitoring hampers the effectiveness of ePIPAC treatment.
To overcome these shortcomings, sequential videos of ePIPAC procedures can be used to build algorithms that may offer a more objective macroscopic assessment of PM behavior
and improve the prediction of progression and prognosis.
Research goals and methodology
A major research goal of this doctoral research project is to develop deep learning models for the staging of PM, through video analysis of recorded ePIPAC procedures,
leveraging these deep learning models for reproducible modeling of disease progression and prognosis.
In particular, the constructed deep learning models are expected to act as quantitative prognostic indicators, for instance following the Peritoneal Cancer Index,
making it possible to predict progression-free survival and overall survival of patients.
In this doctoral research project, we foresee the use of the following iterative approach, in collaboration with medical experts at Ghent University Hospital and UC San Francisco, targeting a PyTorch-based processing pipeline that is supported by a dedicated GPU server:
(1) Literature review on already existing techniques for medical image and video analysis;
(2) Literature review on deep learning and a number of popular deep learning models;
(3) EDA, pre-processing, and annotation of the raw video recordings;
(4) Definition of predictive tasks, as a function of the clinical goals;
(5) Selection and set up of appropriate deep learning models;
(6) Model training and evaluation (internal/external validation and testing); and
(7) Analysis and synthesis of the experimental results obtained.
Please note that the starting point of this doctoral research project consists of raw surgical video recordings provided by Ghent University Hospital and UC San Francisco,
with a subset of these video recordings coming with rough and imbalanced expert annotations.
Therefore, the iterative construction of a foundation dataset will be a major point of research and development attention during the start of this doctoral research project.
The construction of such a foundation dataset will most likely cover the definition and implementation of a custom and scalable workflow for fine-grained surgical video annotation that may involve state-of-the-art techniques for active learning and self-supervised learning. Furthermore, the construction of this foundation dataset will be done in close collaboration with medical experts,
including a second Research Assistant who will primarily work at Ghent University Hospital, with a strong anatomical and medical background.
The below references may be of interest:
[1] Esteva A. et al., Deep learning-enabled medical computer vision. NPJ Digital Medicine 4, Article Number: 5, 2021.
[2] Zhang, A. et al., Shifting machine learning for healthcare from development to deployment and from models to data. Nat. Biomed. Eng, 2022.
Job Description
- Conduct applied research on the topic of surgical video analysis, exploring deep machine learning for robust visual monitoring of the therapeutic response of peritoneal metastases to chemotherapy.
- Report and present research results at internal and external events (regular project meetings, PhD seminars, major international conferences like MICCAI, CVPR, and NeurIPS), and publish research results in peer-reviewed international journals (e.g., International Journal of Cancer).
- Help build and sustain research collaborations between the home campus and the global campus of Ghent University, and with other important centers of expertise (e.g., UC San Francisco).
- Help with lightweight educational activities, such as supervision of exams, supervision of yearly bachelor/master projects, and supervision of occasional internships (research-driven education).
Candidate Profile
- You hold, or you are expected to hold, by September 2022, an MSc degree in one of the following disciplines: computer science/engineering, informatics, electrical engineering, applied mathematics, or biomedical engineering. Related disciplines may be considered as well.
- You have an excellent academic record of accomplishment. In particular, you have a good command of, and a strong interest in, computer programming.
- You are highly motivated to conduct (applied) research at the intersection of (deep) machine learning and the health sciences.
- You have good programming skills in languages such as C++, Python, and/or R. You are familiar with GitHub.
- You have strong analytical skills to interpret the obtained experimental results.
- You are driven to do independent research and you have a strong self-learning ability. A creative and inquisitive attitude is a necessity.
- You have an excellent command of English (a minimum score of 80 on the TOEFL iBT test), both written and orally.
- You are comfortable with working in an international and multi-cultural environment that is dynamic in nature (Ghent University Global Campus counts more than 20 nationalities among its PhD students).
- You have some experience with scientific computing and machine learning.
- Experience with a deep learning framework (e.g., PyTorch, TensorFlow, Keras) is a plus.
- You possess good academic writing and presentation skills.
- You have at least 3 years of relevant work experience after completion of your master's degree in case you do not have the Korean nationality. The Korean government waives this visa requirement if you received your master's degree from a Korean university.
- If need be, you have the willingness to work flexible hours and to participate occasionally in events outside of the regular working hours.
Selection Criteria
- Scientific background and knowledge
- Programming skills (computational and algorithmic thinking)
Application Documents
- Motivation letter
- Full resume (CV), including at least 2 references
- Copy of the BSc and MSc degrees
- Transcripts (overview of study results)
- A PDF version of your Master's Thesis
The application documents must be merged into a single PDF file and be sent via email to Prof. Wesley De Neve (wesley.deneve@ghent.ac.kr) (subject line: Full-time PhD Position in Surgical Video Analysis). The candidate will receive an email confirming receipt of the application.
Application Process and Interview
- Interviews will take place in stages from the first available time.
- Applicants are encouraged to apply in a timely fashion as the position will be filled upon finding the right candidate.
We reserve the right to hold applications on file for potential future job openings. For inquiries, please contact us via email.
Selection Process
- CV Screening -> Interview and technical test (programming in Python and use of Linux) -> Internal committee -> Approval -> Acceptance notice to the candidate selected.
Compensation & Benefits for the Selected Candidate
- Basic terms of the contract - 1-year contract (renewable one time, after positive evaluation, for a total period of maximum 4 years).
- Salary - Starts from an Annual Base Salary of 27,375,000 KRW (Monthly Salary of 2,281,250 KRW, Gross).
- Bonuses - Two additional bonuses in June (92% of monthly salary) and December (100% of monthly salary).
- Housing unit or housing allowance - A single dormitory unit operated by Incheon Global Campus (IGC) will be provided for non-Korean candidates. If the selected candidate has the Korean nationality, or if the selected candidate is a permanent resident in Korea, then a housing unit will not be provided. Instead, a monthly housing allowance of 725,000 KRW will be made available then.
- Two hometown roundtrip tickets (for non-Korean candidates only) - A non-Korean candidate will be provided, on a yearly basis, with two roundtrip tickets to their hometown (with a cost of up to 4,000,000 KRW).
- Severance - Severance shall be paid when the contract ends and if the candidate worked for more than one year.
- Private health insurance - Marsh private health insurance is provided, which includes basic medical reimbursements.
- Extensive annual paid leave and holidays
- The selected candidate shall have 35 days of paid annual leave per year.
- Additional Holidays : from Christmas (Dec 25) to New Year (Jan 1)
