Pierre Geurts

Prof. Pierre Geurts


Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 May 2020 in IEEE Journal of Biomedical and Health Informatics
Verified
Journal Article
Closed access
RMRomain Mormont
Romain Mormont
PGPierre Geurts
Pierre Geurts
RMRaphaël Maree
Raphaël Maree
In this work, we investigate multi-task learning as a way of pre-training models for classification tasks in digital pathology. It is motivated by the fact that many small and medium-size datasets have been released by the community over the years whereas there is no large scale dataset similar to ImageNet in the domain. We first assemble and transform many digital pathology datasets into a pool of 22 classification tasks and almost 900k images. Then, we propose a simple architecture and training scheme for creating a transferable model and a robust evaluation and selection protocol in order to evaluate our method. Depending on the target task, we show that our models used as feature extractors either improve significantly over ImageNet pre-trained models or provide comparable performance. Fine-tuning improves performance over feature extraction and is able to recover the lack of specificity of ImageNet features, as both pre-training sources yield comparable performance.

19 September 2023 in Data Mining and Knowledge Discovery
Verified
Journal Article
Open access
PDF file
ARAmir Hossein Akhavan Rahnama
Amir Hossein Akhavan Rahnama
JBJudith Bütepage
Judith Bütepage
PGPierre Geurts
Pierre Geurts
HBHenrik Boström
Henrik Boström
Local model-agnostic additive explanation techniques decompose the predicted output of a black-box model into additive feature importance scores. Questions have been raised about the accuracy of the produced local additive explanations. We investigate this by studying whether some of the most popular explanation techniques can accurately explain the decisions of linear additive models. We show that even though the explanations generated by these techniques are linear additives, they can fail to provide accurate explanations when explaining linear additive models. In the experiments, we measure the accuracy of additive explanations, as produced by, e.g., LIME and SHAP, along with the non-additive explanations of Local Permutation Importance (LPI) when explaining Linear and Logistic Regression and Gaussian naive Bayes models over 40 tabular datasets. We also investigate the degree to which different factors, such as the number of numerical or categorical or correlated features, the predictive performance of the black-box model, explanation sample size, similarity metric, and the pre-processing technique used on the dataset can directly affect the accuracy of local explanations.

23 October 2013 in Clinical and Vaccine Immunology
Verified
Journal Article
Open access
PDF file
KSK. Schepers
K. Schepers
FMF. Mouchet
F. Mouchet
VDV. Dirix
V. Dirix
ISI. De Schutter
I. De Schutter
KJK. Jotzo
K. Jotzo
FMFrançoise Mascart
Françoise Mascart
The diagnosis of childhood active tuberculosis (aTB) and latent Mycobacterium tuberculosis (M. tuberculosis) infection (LTBI) remains a challenge, and the replacement of tuberculin skin tests (TST) with commercialized gamma interferon (IFN-γ) release assays (IGRA) is not currently recommended. Two hundred sixty-six children between 1 month and 15 years of age, 214 of whom were at risk of recent M. tuberculosis infection and 51 who were included as controls, were prospectively enrolled in our study. According to the results of a clinical evaluation, TST, chest X ray, and microbiological assessment, each children was classified as noninfected, having LTBI, or having aTB. Long-incubation-time purified protein derivative (PPD), ESAT-6, and CFP-10 IGRA were performed and evaluated for their accuracy in correctly classifying the children. Whereas both TST and PPD IGRA were suboptimal for detecting aTB, combining the CFP-10 IGRA with a TST or with a PPD IGRA allowed us to detect all the children with aTB with a specificity of 96% for children who were positive for the CFP-10 IGRA. Moreover, the combination of the CFP-10 IGRA and PPD IGRA detected 96% of children who were eventually classified as having LTBI, but a strong IFN-γ response to CFP-10 (defined as >500 pg/ml) was highly suggestive of aTB, at least among the children who were <3 years old. The use of long-incubation-time CFP-10 IGRA and PPD IGRA should help clinicians to quickly identify aTB or LTBI in young children.

14 December 2023 in Biomolecules
Verified
Journal Article
Open access
PDF file
NKNavdeep Kumar
Navdeep Kumar
RMRaphaël Maree
Raphaël Maree
PGPierre Geurts
Pierre Geurts
Marc MullerMarc Muller
Marc Muller
Detecting skeletal or bone-related deformities in model and aquaculture fish is vital for numerous biomedical studies. In biomedical research, model fish with bone-related disorders are potential indicators of various chemically induced toxins in their environment or poor dietary conditions. In aquaculture, skeletal deformities are affecting fish health, and economic losses are incurred by fish farmers. This survey paper focuses on showcasing the cutting-edge image analysis tools and techniques based on artificial intelligence that are currently applied in the analysis of bone-related deformities in aquaculture and model fish. These methods and tools play a significant role in improving research by automating various aspects of the analysis. This paper also sheds light on some of the hurdles faced when dealing with high-content bioimages and explores potential solutions to overcome these challenges.

8 November 2023 in Machine Learning
Verified
Journal Article
Closed access
PDF file
VHVân Anh Huynh-Thu
Vân Anh Huynh-Thu
PGPierre Geurts
Pierre Geurts