PeopleofSuperintelligence

Researcher

Arnaud A. A. Setio

Also known as A. A. A. Setio, Arnaud Arindra Adiyoso Setio

Arnaud Arindra Adiyoso Setio is a research scientist whose work applies deep learning to medical imaging and computer-aided diagnosis. His education includes an electrical-engineering bachelor's from Institut Teknologi Bandung, a master's from Eindhoven University of Technology, and a 2018 Radboud PhD on computer-aided diagnosis in thoracic CT. From 2016 through 2021, he built multi-view lung-nodule networks, set up LUNA16, studied adversarial training, and coauthored a study comparing AI with radiologists. His 2025 and 2026 publications explored radiology-report generation and pulmonary-nodule malignancy risk. As of October 2026, he is a staff research scientist at Lunit.

Now at
Lunit
Works
35
Citations
19,841
h-index
20
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2016

    Multi-view networks for CT

    In 2016, he coauthors a multi-view convolutional neural network method to reduce false positives in automated pulmonary-nodule detection.

  2. 2017

    The LUNA16 benchmark

    In 2017, he sets up LUNA16, an evaluation challenge for automatic pulmonary-nodule detection on 888 chest CT scans from the public LIDC-IDRI data set.

  3. 2017

    Deep learning for nodule management

    He coauthors a 2017 deep-learning system that uses multi-stream, multi-scale convolutional networks to classify nodule types for lung cancer screening.

  4. 2018

    A thesis on chest CT

    In 2018, he completes a PhD thesis on computer-aided diagnosis in thoracic CT scans for lung cancer screening at the Diagnostic Image Analysis Group, Radboud University Medical Center. His doctoral work covers automatic detection and classification of lung nodules in chest CT under Bram van Ginneken; his education includes an electrical-engineering bachelor's degree from Institut Teknologi Bandung and a master's degree from Eindhoven University of Technology.

  5. 2020

    Adversarial training for CT

    In 2020, he coauthors a paper that adds adversarial synthetic nodules and attack samples to training data to improve lung-nodule detector generalization and robustness.

  6. 2021

    Comparing AI with radiologists

    He coauthors a 2021 lung-screening study that compares top deep-learning algorithms from a public competition with 11 radiologists using 300 patient scans.

  7. 2025

    Radiology report generation

    At a 2025 MICCAI workshop, he coauthors research on customizing radiology report generation for different clinical workflows and institutional requirements.

  8. As of October 2026

    Staff scientist at Lunit

    In February 2026, he coauthors a DeepPNP study that estimates pulmonary-nodule malignancy risk; the paper reports that the model outperforms established clinical risk scores. As of October 2026, he is a staff research scientist at Lunit, working on deep learning, medical image analysis, and computer-aided diagnosis.

Career

  1. Since 2025LunitStaff Research Scientist
  2. 2015–2021Radboud University Nijmegen

Education

  1. 2018Radboud University Medical CenterPhD
  2. Bandung Institute of TechnologyBachelor's degree
  3. Eindhoven University of TechnologyMaster's degree

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

  1. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes Radboud University Nijmegen. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.
  2. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Radboud University Nijmegen. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  3. Google has the highest combined h-index among AI organizationsIncludes Radboud University Nijmegen. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.
  4. 11% of AI doctorate holders from 2010 to 2016 founded or led a company or lab by 10 years after graduationIncludes Radboud University Nijmegen. We estimate how often AI doctorate holders founded or led a company or lab by 10 years after graduation, how waiting times compare between founders and top jobs, and which universities trained the most.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Arnaud A. A. Setio yet. Send us a free-licensed photo.

  1. Publications - Arnaud A. A. Setio arnaudsetio.github.io. Checked 20 October 2026.
  2. Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge arxiv.org. Checked 20 October 2026.
  3. Towards automatic pulmonary nodule management in lung cancer screening with deep learning | Scientific Reports nature.com. Checked 20 October 2026.
  4. Publications of Arnaud Arindra Adiyoso Setio - Diagnostic Image Analysis Group diagnijmegen.nl. Checked 20 October 2026.
  5. Arnaud A. A. Setio arnaudsetio.github.io. Checked 20 October 2026.
  6. No Surprises: Training Robust Lung Nodule Detection for Low-Dose CT Scans by Augmenting with Adversarial Attacks arxiv.org. Checked 20 October 2026.
  7. Deep learning for lung cancer detection on screening CT scans: results of a large-scale public competition and an observer study with 11 radiologists researchinformation.umcutrecht.nl. Checked 20 October 2026.
  8. Deep learning-based pulmonary nodule risk assessment outperforms established malignancy risk scores in lung cancer screening pmc.ncbi.nlm.nih.gov. Checked 20 October 2026.