March 2024

Postdoc in machine learning and reinforcement learning

Application deadline: April 15th, 2024
This position is for an enthusiastic and dedicated postdoc with a special interest in machine learning, reinforcement learning and its applications in healthcare.

The project aims at pushing the boundaries of machine learning and reinforcement learning with applications to healthcare data. This project will develop state-of-the-art machine learning architectures tailored for temporal data. The goal will be to improve existing techniques such as off-policy evaluations in the context of reinforcement learning with both fundamental and applied contributions.

The application on epidemiological data will apply these innovative methodologies to refine our understanding of how environmental factors interact with intrinsic capacity, ultimately shaping functional outcomes in older adults.

  • you’ll develop new algorithms and characterize their properties with theoretical and experimental techniques.
  • you’ll apply these algorithms on real-world data with applications to healthcare.
  • you’ll be at the forefront of an interdisciplinary research, driving innovation through your expertise in machine learning and reinforcement learning.

More information and link to apply can be found here. 

 

Postdoc Translational AI Laboratory: graph machine learning

Application deadline: March 25th, 2024
Are you the Postdoc with whom we dive into the complexity of real world medical data with human-assisted graph machine learning?

In recent years, many new directions in graph machine learning have been investigated. A major problem for all graph machine learning approaches, especially on rich datasets, is incompleteness. In large knowledge graphs, there is always missing information, and even some of the included information might be wrong. Complex query answering is an active area of research trying to deal with these problems by answering queries about the graph as if the graph were complete. To do so, the approaches must predict the set of correct answers. While several approaches have been investigated by now, they all deal with rather simple knowledge graphs. They do not deal with attributes of various modalities and these methods cannot take human input into account, nor can explanations be provided on how results came to be.

You will develop and analyze new graph machine learning methods in the context of neurological disorders, specifically leukodystrophies, which are about degeneration of white matter in the brain. This research aims to identify features (biomarkers) that predict the course of adrenoleukodystrophy (ALD). We do this as part of our overall pursuit to unravel the pathophysiology of ALD, which would elucidate pathways involved in its pathogenesis and could potentially guide therapy development. This research will benefit hugely if clinical knowledge graphs can be used by physicians to interactively query potential associations between biomarkers during the diagnosis, prognosis and treatment phases with a system that can also provide clear explanations.

The project is an industry-sponsored collaboration between the Translational AI Laboratory (TrAIL) at the Department of Laboratory Medicine of Amsterdam UMC and the Learning and Reasoning (L&R) group at the Department of Computer Science of the VU University Amsterdam. You will be working in both the TrAIL and L&R research groups, in close collaboration with other laboratory medicine researchers and clinicians in pediatric neurology.

More information here.

 

PhD position: Using AI to improve cognitive outcomes in brain tumor patients

Application deadline: March 10th, 2024
Do you want to use advanced AI to better predict the outcome of brain surgery?

Our overall idea is that personalized computational modeling of a brain tumor operation, a so-called ‘virtual resection’, will help us identify which patients are at risk of cognitive decline after the surgery.

Glioma is a rare brain tumour with a dismal prognosis. Its high disease burden is significantly compounded after initial treatment through tumour resection by functional and cognitive decline, which lowers quality of life and keeps patients from returning to work and social life. Adding to the detrimental impact of this decline itself, uncertainty about whether and when it might occur is stressful for patients and caregivers, and is difficult for physicians as they work to optimally counsel patients during the course of the disease. PREDICT aims to establish which clinical or more advanced parameters are able to predict the postoperative functional and cognitive decline that have a large impact on patients’ daily lives.

This project is a collaboration between translational researchers, clinicians, computational modelers and machine learning experts: you will be embedded in a multidisciplinary team and will have ample opportunities to learn about all aspects of translational research, AI in medicine and neuropsychology and neuro-oncology.

More information about this position at Amsterdam UMC here.

——————
February 2024

Postdoc position ‘LabGPT’

Application deadline: February 29th, 2024
Are you Postdoc with a drive to create innovative health technology solutions using generative AI to improve medical decision making?

The launch of ChatGPT has accelerated the use of AI as a disruptive technology. This also applies to healthcare where, for example, ChatGPT can help reduce the registration burden or rewrite complex medical texts in understandable language. But generative AI has much broader and far-reaching consequences for the use of AI in healthcare: for example, the way that clinical risk and prediction models are developed and used is changing drastically. In this project, we develop a clinical foundation model with laboratory data and contextual information. Just as ChatGPT was developed based on text data from the internet with a chatbot as an application, we will develop a foundation model based on laboratory data with a series of prediction models as applications. Three such next-generation clinical prediction models will be built: for blood/urine culture analysis, heart failure and adrenoleukodystrophy, respectively.

You will be based in the Translational AI Laboratory at the Department of Laboratory Medicine of Amsterdam UMC, and tightly collaborate with medical AI specialists and other clinical disciplines.

Find more information and the application link here

 

 

——————-

November 2023

Postdoc Researcher: Can Artificial Intelligence improve clinical decision making?

Application deadline: November 19th, 2023
The postdoctoral researcher position is part of the research project “ Evidence Based clinical decision making using an open AI-tool (Acronym: ABC-AI; funded by the Open Science Fund from NWO). The aim of the project is to investigate the added value of Artificial intelligence in clinical decision making in doctors.

We are looking for a postdoc that is interested in answering the following questions;

  • what is the added value of Artificial Intelligence in clinical decision making by clinical doctors?
  • How can the use of an AI-driven tool be operationalized and optimised in daily clinical care?
  • And what should be improved to make the tool better equipped to answer clinical questions?

Under supervision of Dr. J.K. Tijdink, and founders of the startup that created the tool, the researcher will find empirical answers to the above mentioned questions using  quantitative research methods (survey study).

For more information see here.

PhD Innovatieve toepassingen v machine learning voor antimicrobial stewardship

Vacancy in Dutch. Application deadline: November 26th, 2023
De focus van dit promotietraject ligt op het ontwikkelen van innovatieve tools die medische informatie uit het EHR beschikbaar maakt voor diagnostische en antimicrobiele stewardship projecten

Beoogde projecten:

  • Onderzoeken wat de beste methode is om diagnoses te extraheren uit het EHR (bv: gebaseerd op huidige diagnose-registratie aangevuld met text-mining-based AI-technieken).
  • Ontwikkeling van een applicatie die relevante informatie van kweken, diagnoses en medicatievoorschriften (semi-)automatisch kan koppelen (‘bug-drug combinaties’).
  • Ontwikkelen van predictie-model dat real-time de kans op een bacteriemie kan voorspellen in opgenomen patiënten, gebaseerd op routinematig aanwezige informatie in het EHR.
  • Uitbouwen van dit predictie-model naar identificatie van pathogeen-groepen (S.aureus, gram-negatieve staven), gecombineerd met risicogroepen voor cefalosporine-resistentie (oa ESBL).
  • Gebruik van clustering en visualisatie-technieken voor het ontwikkelen van applicaties voor het real-time monitoren van het beloop van infecties en kuurduren.Meer informatie hier.

——————
August 2023

Postdoc AI in Biomedical sciences

Application deadline: August 27th, 2023
We are looking for a postdoctoral fellow AI in Biomedical sciences to generate algorithms that will predict efficies on brain cancer.

The project

Brain tumors, particularly glioblastomas and primary tumors that have spread to the brain, have a poor prognosis, and patients are in urgent need of improved treatment strategies that optimally balance efficacy and toxicity. We currently have a vacancy available for research on this topic, partcularly focussing on modeling drug combination effects.

The application of new combination therapies against cancer is attractive given the high relapse rates of single-agent therapies. However, when using therapy combinations, the efficacy must outweigh the potential side effects. Toxicity is an important reason for treatment discontinuation and for drugs with increased toxicity being less prescribed. This prevents patients from accessing a better quality of life and increased lifespan. However, public data of efficacy and toxicity are often only accessible as unstructured text. Similarly, patient data files consist mostly of unstructured text. This is a major challenge preventing large-scale data analysis and modeling.

In this project, we aim to predict clinically relevant combination therapies and estimate their interaction regarding efficacy and toxicity. In this regard, the following steps have been planned: (1) extract toxicity data from patient files, (2) further refine an already developed toxicity model and (3) identify the most promising personalized combination therapy, based on the effectiveness to toxicity ratio.

Our collaborator, the company Medstone, has already fined-tuned a model to process and extract data from biomedical scientific articles. Combined with Amsterdam UMC’s previous experience in compiling toxicity and therapy efficacy datasets, extraction of data from the US FDA Adverse Event Reporting System (FAERS) toxicity database of more than 15 million patients has been performed.

Our first goal is to set up a generalized algorithm to predict the ratio between effectiveness and toxicity of clinically relevant drug combinations for oncology (based on 8,000 trial records). Furthermore, we aim to enhance the effectiveness and safety of combination therapies by combining the hospital’s clinical data with the public data-derived prediction model, focussing on brain cancers (based on 1,000 curated patient records). This effort will further be optimized together with pharmacists, statisticians and clinicians aiming at enhancing our infrastructure and also benefitting the Dutch Cancer Registry (IKNL) as a part of our project.

About your role

The inhibition of multiple targets is positively associated with therapeutic efficacy but comes with increased adverse events. By devising an algorithm minimizing this trade-off, we expect to predict an optimal balance between effectiveness and toxicity for existing and new therapies.

  • You will mainly focus on processing data extracted from different sources and developing machine learning models, here applied to brain cancers.
  • You will characterize the capacity of these prediction models to generalize well.
  • Additionally, you will be occasionally involved in the processing and analysis of omics data generated in the wet laboratory.
About you

You are a highly motivated and ambitious candidate with a PhD in bioinformatics, you have experience with processing biological, biomedical, and chemical data and have experience in developing machine learning models.

In addition, you meet the following requirements:

  • you have experience with Python and using deep learning libraries
  • you have affinity with natural language processing (experience with omics data is a plus)
  • you speak and write English fluently
  • you feel comfortable in an international multidisciplinary environment
  • you are enthusiastic, conscientious, driven, accurate, committed and resourceful
  • you can function well independently as well as in a team
  • you have good communication skills

See here for more information.

2 MSc Internship projects: AI applied in Multiple Sclerosis and Alzheimers Disease

Project 1: Detect new MS brain lesions in serial FLAIR scans using AI

Multiple Sclerosis (MS) is a neuro-inflammatory condition of the central nervous system (CNS) causing cognitive and physical disability due to neurodegeneration.

This project aims to develop a deep learning model to distinguish MS brain lesions that occurred in the preceding year from MS lesions that were already present on a single FLAIR MR image.

Work environment

The Amsterdam UMC department of Radiology and Nuclear Medicine performs high quality research on aging, dementia and Multiple Sclerosis, in particular with the use of advanced Magnetic Resonance Imaging (MRI). You will work in the group of prof. dr. Frederik Barkhof (Neuroradiologist) with dr. ir. Alle Meije Wink and ir. Georgios Lappas, who are using AI techniques to improve the diagnosis of neurological disorders.

We are looking for

  • A master student within the field of Medical Natural Sciences or Technical Medicine
  • Basic experience with deep learning programming, e.g., Tensorflow/Keras;
  • Experience with image analyses is a plus;
  • You should be interested in Multiple Sclerosis modelling;

Applying and information

Interested in this internship? For content-related questions you can contact contact Alle Meije Wink at a.wink@amsterdamumc.nl or Georgios Lappas at g.lappas@amsterdamumc.nl.



——————————-

Project 2: Predict PET brain images from MRI scans for Alzheimer’s disease diagnosis using AI

Magnetic Resonance Imaging and Positron Emission Tomography (PET) are used to diagnose brain disorders such as Alzheimer’s disease (AD). Amyloid accumulation in the brain, an early marker of AD pathology, can be measured directly with PET but not MR. PET data are more challenging to get than MRI data because of cost, radiation exposure and other restrictions (e.g., lack of PET scanners). Recent advances in Artificial Intelligence (AI), particularly Deep Learning (DL), have yielded interesting applications such as image synthesis (Welander et al., 2018; Xu et al., 2020; Zhang et al., 2022). We hypothesise that DL models, specifically generative adversarial networks (GAN), can synthesise amyloid PET brain scans from MR images to estimate amyloid accumulation in early AD.

In this project, we will investigate whether synthetic PET brain images, synthesised from brain MRI via DL, can be used to distinguish patients with amyloid accumulation from those without. This discrimination will be tested with an image-based classification method.

Work environment

The Amsterdam UMC department of Radiology and Nuclear Medicine performs high quality research on aging, dementia and Multiple Sclerosis, in particular with the use of advanced Magnetic Resonance Imaging (MRI). You will work in the group of prof. dr. Frederik Barkhof (Neuroradiologist) with dr. ir. Alle Meije Wink and ir. Georgios Lappas, who are using AI techniques to improve the diagnosis of neurological disorders.

We are looking for

  • A master student within the field of Medical Natural Sciences or Technical Medicine
  • Basic experience with deep learning programming, e.g., Tensorflow/Keras;
  • Experience with image analyses is a plus;
  • You should be interested in Alzheimer’s disease modelling;

Applying and information

Interested in this internship? For content-related questions you can contact Alle Meije Wink at a.wink@amsterdamumc.nl or Georgios Lappas at g.lappas@amsterdamumc.nl.

—————

July 2023

Postdoc Learn-2-Control: impulse control disorders in Parkinson’s disease

Application deadline: August 21st, 2023
Impulse control disorders (ICDs) are a class of highly detrimental neuropsychiatric disorders that occur in a large subpopulation of Parkinson’s disease (PD) patients. ICDs are primarily a complication of dopaminergic medication in a subgroup of patients that carry an as of yet unidentified neurobiological vulnerability, while these drugs are prescribed to virtually all PD patients. Your job will be to help identify this subgroup of PD using a combination of clinical and molecular imaging markers and in this way work towards precision medicine for PD.

You will become part of Team Neuropsychiatry (embedded in the dept. Anatomy and Neurosciences and dept. Psychiatry),  team that runs multiple clinical trials and neuroimaging studies ensuring a mutual interaction between clinicians and preclinical researchers in an academic setting.

The project runs for a year and a half and will start in the fall of 2023.

As a postdoctoral researcher in the team, your main responsibility will be to work on this Michael J. Fox Foundation funded project (‘Learn-2-Control’, Principle investigator: C. Vriend)

  • This project will apply machine learning on a large existing dataset to identify neuroimaging and clinical markers that can predict the future development of an impulse control disorder in Parkinson’s disease after they commence dopaminergic medication.
  • Part of your work is also to support ongoing projects of team Neuropsychiatry that uses an array of methods (e.g. MRI, EEG, repetitive transcranial magnetic stimulation and neuropsychological assessment) to study the pathophysiology of different brain disorders and their treatment.
About you
  • A completed PhD in the field of Artificial Intelligence, Data Science, Engineering, Neuroscience, or similar;
  • Expertise in Machine/Deep learning is required;
  • You can start in Oct/Nov ’23;
  • Experience with neuroimaging (MRI) analyses (software) is preferred;
  • Affinity with Parkinson’s disease study population is preferred;
  • A proven record of academic writing:
  • Strong problem solving skills:
  • Experience with supervising interns and PhD students:
  • A strong team spirit with willingness to support other ongoing projects with your expertise.

More info and applications here.

 

 

————–

June 2023

PhD Student in Machine Learning

Application deadline: July 2nd, 2023
We are looking for a new PhD student who will do research in the area of machine learning (ML) and in particular on supervised learning and reinforcement learning (RL) and the application of these techniques in a large mental health project (RECONNECTED). The EU funded RECONNECTED project focuses on the development of innovative interventions for mental health for vulnerable citizens in the EU that are driven by machine learning. The position is embedded in the Quantitative Data Analytics (QDA) group of the VU in close collaboration with the Centre for Urban Mental Health at the University of Amsterdam. The QDA group focuses on both fundamental and application-driven research in ML while the center of Urban Mental Health has a wealth of experience in applying complexity science to mental health and intervention development. We are looking for candidates with experience in the area of machine learning, with a special preference for those having expertise in one or more of the following areas: reinforcement learning, the utilization of domain knowledge in ML, and data efficient ML methods. We are interested in attracting a PhD student that is able to perform ground-breaking research in fundamental aspects of machine learning and reinforcement learning with a willingness to contribute to data collection efforts among human subjects (resulting in data that can be used to train machine learning models). The final goal within the project is to apply the developed algorithms in an app and use them to identify personalized targets for micro-interventions targeting psychological resilience.

For more info see here.

————–

May 2023

Postdoc VASCULAID project

Application deadline: May 29th, 2023
We are looking for a Postdoc who will help in an exciting European project. In six years we want to develop an algorithm with AI to predict cardiovascular risks and disease progression in patients with abdominal aortic aneurysms (AAA) and/or peripheral arterial diseases (PAD).

We are searching for candidates who want to improve healthcare with a medical background. On the one hand, the candidate must enjoy to learn new technical approaches in AI, perform data analysis and machine learning approaches or have interest in multisource data analysis or biomedical analysis (proteomics and genomics). The candidate must function in a multidisciplinary team and collaborate with doctors to show the clinical advantages of the technique and get it implemented in the clinical workflow.

More information here.

—————

March 2023

Two postdoctoral research positions on machine learning and reasoning for medical decision making

Application deadline: May 1st, 2023
Do you have an interest in AI systems for medical decision making that interact with doctors and patients? Are you interested in medical decision systems that are explainable, and that take into account safety constraints and  patient preferences? Do you have an interest in combining reinforcement learning with symbolic techniques for medical decision making? Then please apply at Vrije Universiteit Amsterdam (VU).

We are looking for two candidates:

1. A candidate for the NWO-funded project PersON. The project aims to facilitate shared decision making for personalised care in order to improve the effectiveness and efficiency of oncological healthcare. For this position the emphasis will be on a decision support system that takes into account safety constraints and patient preferences, and on the explainability of the system’s advice. We are aiming to use a combination of machine learning and symbolic knowledge representations.

2. A candidate for a project on reinforcement learning for patient-centric decision making in the context of the Dutch national research programme on Hybrid Intelligence (https://hybrid-intelligence-center.nl). The project aims to develop training methods for a symbolically informed Reinforcement Learning agent to learn a strategy that is (i) cooperative with the expert, that (ii) displays behaviour which is coherent from the human point of view, and that (iii) takes into account the often conflicting goals that a medical treatment is expected to satisfy.
Note that this job opening is one of several job openings on the HI page with job openings. You can apply to more than just this one. If you do so, please inform the Hybrid Intelligence Centre’s Project Manager via email of the projects you are applying to.
We aim to create hybrid intelligence for everyone, see also our Diversity Statement. To do this, we need an inclusive and diverse team of researchers. We especially encourage people from underrepresented groups to apply for this job.

Both projects are embedded in a stimulating ecosystem in which collaboration is encouraged, and are in collaboration with prof. Annette ten Teije and prof. Frank van Harmelen in the Learning and Reasoning group (https://lr.cs.vu.nl/). Project nr. 2  is in close collaboration with dr. Herke van Hoof from the University of Amsterdam.

To apply and for more information please see here.

—————-

January 2023

Vacature student-assistent data-analyse Nederlands CP register (Dutch only)

Ben jij een student met interesse in ICT en data, en heb jij belangstelling voor de revalidatiezorg? Dan is dit je kans! Ter ondersteuning van het Nederlands CP register zijn we op zoek naar een Student-assistent data-analyse (8 uur per week)

Het Nederlands CP register is een innovatief landelijk register waarin gegevens van kinderen en jongeren met cerebrale parese (CP) verzameld worden. Het register bestaat uit een follow-up en behandelregister, waarin systematisch zorgdata en behandelresultaten worden geregistreerd. Het betreft data die in de reguliere zorg worden verzameld door artsen, zoals kinderrevalidatieartsen en kinderorthopeden, maar ook data die patiënten thuis digitaal invoeren. Deze data worden gebruikt voor directe beslisondersteuning in de spreekkamer, maar kunnen ook op geaggregeerd niveau gebruikt worden voor feedback naar de zorginstellingen en voor wetenschappelijk onderzoek. Meer informatie over het Nederlands CP register is terug te vinden op www.cpregister.nl.

Wat ga je doen?

Voor 8 uur in de week ondersteun je het registerteam van het Nederlands CP register bij het verwerken en analyseren van data die in de zorg verzameld zijn in het ICT platform. Een belangrijk onderdeel betreft het opstellen van rapportages naar revalidatiecentra en ziekenhuizen. Daarnaast werk je mee aan de trainingen voor gebruikers in lokale centra en help je gebruikers (zorgprofessionals) bij technische vragen. Tot slot assisteer je bij andere onderdelen, zoals het inrichten van het ICT platform (Gemstracker/LimeSurvey), het up-to-date houden van de website en het ontwikkelen van handleidingen.
Je bent werkzaam bij de afdeling Revalidatiegeneeskunde van het Amsterdam UMC, locatie VUmc. Je maakt deel uit van het registerteam, bestaande uit een projectleider, projectmanager, implementatiemedewerker en twee andere student-assistenten.
De uren zijn flexibel in te plannen en thuiswerken is (deels) mogelijk. Bij voorkeur ben je één dagdeel op dinsdag of woensdag aanwezig op locatie VUmc. In overleg is het ook mogelijk om deels in de avonduren en in het weekend te werken.

Wie ben jij?

Een enthousiaste (master)student Informatica, Klinische Technologie of een andere gerelateerde studie met belangstelling voor de revalidatiezorg. Je bent een secure werker, sociaal vaardig, flexibel, zelfstandig en ondernemend. Voor deze functie zijn we op zoek naar een student die affiniteit heeft met dataverwerking en programmeren. Je gaat aan de slag met de export van de ruwe data om analyses uit te voeren en rapportages naar de centra op te stellen. Hiervoor gebruiken wij de statistische programmeeromgeving R. Bij voorkeur zoeken we een student die hier ervaring mee heeft, of die aantoonbaar affiniteit heeft met het zich eigen maken van vergelijkbare software en analyses.

Wat bieden wij?

Wij bieden je een tijdelijke aanstelling voor 8 uur per week voor de periode van februari t/m 31 december 2023. Je bent bij voorkeur per direct beschikbaar (overleg over startdatum is mogelijk). Een commitment voor een periode van minimaal 10 maanden is nodig.
Je wordt ingedeeld in schaal 6, afhankelijk van je ervaring (CAO Academische ziekenhuizen).

Ben je geïnteresseerd en wil jij ons registerteam versterken?

Solliciteer dan door jouw motivatiebrief en cv vóór 12 februari te mailen naar Aukje Andringa (a.andringa@amsterdamumc.nl). We gaan graag met je in gesprek.

Voor meer informatie over de inhoud van de functie kun je contact opnemen met Aukje Andringa via 020-4443062 of via email: a.andringa@amsterdamumc.nl.

Post Doc/Assistant Professor – MyDigiTwin project

The MyDigiTwin project is a scientific initiative to develop personalized risk prediction algorithms using data from multiple cohorts to create digital twins empowering people and enable democratisation of health knowledge and health data.

The project is a national collaboration among academic institutes and private partners, spanning over 20 partners. The multidisciplinary team involves members from the data science, life science, healthcare, social science / humanities, and the creative field.

There are three positions for this project, spanning from post-doc to assistant professor, depending on skill and experience.

For more info and to apply, please see here.

 

——————

September 2022

Research assistants wanted: Deep generative models for spatial reconstruction of single cell transcriptomics data

We are looking for two ambitious master students for two exciting, paid research assistant positions. The positions are for 1 day/week for 10 months, ideally to be combined with a master thesis research project. Project start: October

Project description

The spatial organization of cells underlies the function of many biological systems, including tumors. Emerging biotechnologies now allow for spatially resolved quantification of transcriptomes at single cell resolution [1], but are expensive and low-throughput. On the other hand, conventional single cell RNA-sequencing is now routinely applied to large sample cohorts, but does not retain spatial information. Recently, spectacular advances in generative models [2] has led to models such as ImaGen [3] and DALL-E [4], that can synthesize high-quality and high-fidelity images from text or metadata.

In this project, we will employ deep generative models to reconstruct the spatial structure of biological samples from single cell transcriptomic count data alone. To this end, we will formulate and train a deep generative model on spatial transcriptomics data, matched with single cell count data. The trained models will then be employed to synthesize “images” using count data only as input metadata, which will allow us to extract biologically meaningful spatial information (e.g. extent of immune infiltration in tumor) from the images.

This is a high-risk, high-gain project. If successful, it will allow for augmentation of single cell transcriptomics dataset with spatial features, adding great value to such datasets.

Requirements

●  Background in AI (programming in Python and PyTorch, the basics of mathematics and deep learning) or molecular biology.
●  Strong communication skills

Contact

To apply, send an email to Jakub Tomczak (j.m.tomczak@vu.nl)) and Evert Bosdriesz (e.bosdriesz@vu.nl) with a brief motivation and grade list.

References

  1. Moses L, Pachter L. Museum of spatial transcriptomics. Nat Methods. 2022;19: 534–546.
  2. Tomczak JM. Deep Generative Modeling. 2022. doi:10.1007/978-3-030-93158-2
  3. Saharia C, Chan W, Saxena S, Li L, Whang J, Denton E, et al. Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding. arXiv [cs.CV]. 2022. Available: http://arxiv.org/abs/2205.11487
  4. Ramesh A, Dhariwal P, Nichol A, Chu C, Chen M. Hierarchical Text-Conditional Image Generation with CLIP Latents. arXiv [cs.CV]. 2022. Available: http://arxiv.org/abs/2204.06125

MSc project: Investigating how human and algorithmic agents develop novel modes of intelligence

One of the areas where life-and-death decisions are increasingly made based on algorithmic technologies (e.g., deep-learning) is “medical diagnosis”. Every day, doctors who are used to working with their eyes and their traditional technologies, are confronted with new algorithmic technologies (such as wearable devices, AI-empowered diagnosis tools, and many algorithms that offer them insights into medical issues). The situation is at best like giving an advanced car to those who are used to riding bikes! You can imagine novel expertise and experiences emerge when these professional cyclists find themselves behind the driving panel of these rather autonomous, unknowable AI technologies. What happens to their expertise? How effectively can they use their medical expertise? How and under which conditions might they fall into certain ineffective ways of working such as automation bias, confirmation bias, ruling-out bias against the technology, and many new traps that we just do not know? 

In this research project, which is part of an ongoing research program at KIN center for digital innovation, we engage medical professionals in a range of experimental procedures, where they work with various forms of AI tools and under different working conditions. This way, we examine how they use and develop their expertise for medical diagnosis when working with these technologies. The results offer developers of these algorithms the chance to proactively learn about the (in)effective design choices and the (un)intended consequences of their technologies. It also helps professionals from medical and (partly) other domains of knowledge work learn how they can develop novel expertise of working with these technologies. Finally, we gain insights about how to assess, organize, and regulate these technologies in order to minimize their ineffective modes of operation. 

Are you interested in developing your research (thesis) in this domain? This would be an opportunity for those who are interested in advancing our knowledge of human-technology intelligence and expertise, by combining both technical and social aspects. If you are interested in knowing more about this research project, please send an email to Mohammad H. Rezazade Mehrizi (m.rezazademehrizi@vu.nl), School of Business & Economics, Knowledge, Information and Networks Research Group.

 

 

 

 

————————————-

Previous vacancies:

ASSOCIATE PROFESSOR IN AI & HEALTH

Deadline: 15th January, 2022

Do you want to contribute to the next generation of healthcare fueled by Artificial Intelligence? We then invite you to apply for the position of Associate professor for AI and Health, a joint position of the Computer Science Department of the VU and the Cancer Center Amsterdam of the Amsterdam UMC. Please find more information here.

PHD STUDENT IN MACHINE LEARNING

Deadline: 17th January, 2022

We are looking for a new PhD student who will do research in the area of Machine Learning (ML) and in particular on supervised learning and Reinforcement Learning (RL). The position is embedded in the Quantitative Data Analytics group.

We are interested in attracting a PhD student able to perform ground-breaking research in fundamental aspects of Machine Learning and Reinforcement Learning. The PhD student will apply these novel algorithms in the context of an EU funded project called ICARE4OLD focused on improving care for elderly by machine learning based recommendations on interventions. For more information and to apply, please see here.