Felix Wagner @ NeurIPS
@felwag.bsky.social
PhD student at University of Oxford 💻 Computer Vision for Medicine | Federated Learning 🖥️👨💻
🦭 At #NeurIPS2025 in San Diego this week (Dec 1–7), presenting my final PhD project!
If you’re working on medical imaging, foundation models, multimodal learning, federated learning, or OOD detection, let’s meet! Happy to grab a coffee ☕️ or beer 🍺.
DM me to connect! 🌴
If you’re working on medical imaging, foundation models, multimodal learning, federated learning, or OOD detection, let’s meet! Happy to grab a coffee ☕️ or beer 🍺.
DM me to connect! 🌴
December 1, 2025 at 11:46 PM
🦭 At #NeurIPS2025 in San Diego this week (Dec 1–7), presenting my final PhD project!
If you’re working on medical imaging, foundation models, multimodal learning, federated learning, or OOD detection, let’s meet! Happy to grab a coffee ☕️ or beer 🍺.
DM me to connect! 🌴
If you’re working on medical imaging, foundation models, multimodal learning, federated learning, or OOD detection, let’s meet! Happy to grab a coffee ☕️ or beer 🍺.
DM me to connect! 🌴
Tested on 12 OOD tasks across 🧴dermatology, 🩻 chest X-ray, ultrasound & 🔬 histopathology.
💥 DIsoN consistently performs strongly against state-of-the-art methods, with higher AUROC and fewer false positives.
Attention bad pun: 🧹 DIsoN cleans up OOD samples like a Dyson 💨
💥 DIsoN consistently performs strongly against state-of-the-art methods, with higher AUROC and fewer false positives.
Attention bad pun: 🧹 DIsoN cleans up OOD samples like a Dyson 💨
September 20, 2025 at 8:29 AM
Tested on 12 OOD tasks across 🧴dermatology, 🩻 chest X-ray, ultrasound & 🔬 histopathology.
💥 DIsoN consistently performs strongly against state-of-the-art methods, with higher AUROC and fewer false positives.
Attention bad pun: 🧹 DIsoN cleans up OOD samples like a Dyson 💨
💥 DIsoN consistently performs strongly against state-of-the-art methods, with higher AUROC and fewer false positives.
Attention bad pun: 🧹 DIsoN cleans up OOD samples like a Dyson 💨
DIsoN enables comparing a test sample with the training data distribution, without data transfer!
How?
🔑 We train a binary classifier per test sample to “isolate” it from training data.
📈 The more training steps needed → the more likely the sample is in-distribution.
How?
🔑 We train a binary classifier per test sample to “isolate” it from training data.
📈 The more training steps needed → the more likely the sample is in-distribution.
September 20, 2025 at 8:29 AM
DIsoN enables comparing a test sample with the training data distribution, without data transfer!
How?
🔑 We train a binary classifier per test sample to “isolate” it from training data.
📈 The more training steps needed → the more likely the sample is in-distribution.
How?
🔑 We train a binary classifier per test sample to “isolate” it from training data.
📈 The more training steps needed → the more likely the sample is in-distribution.
In medical imaging, safe deployment isn’t just about accuracy.
⚠️ Models must flag unusual scans (artifacts, rare conditions) so clinicians can double-check.
But there’s a problem:
📦 Training data is often private, large, and unavailable after deployment.
⚠️ Models must flag unusual scans (artifacts, rare conditions) so clinicians can double-check.
But there’s a problem:
📦 Training data is often private, large, and unavailable after deployment.
September 20, 2025 at 8:29 AM
In medical imaging, safe deployment isn’t just about accuracy.
⚠️ Models must flag unusual scans (artifacts, rare conditions) so clinicians can double-check.
But there’s a problem:
📦 Training data is often private, large, and unavailable after deployment.
⚠️ Models must flag unusual scans (artifacts, rare conditions) so clinicians can double-check.
But there’s a problem:
📦 Training data is often private, large, and unavailable after deployment.
Whoop #NeurIPS2025 accepted! 🎉
Meet DIsoN, our 🧹💨 privacy-preserving OOD detector that compares test samples to training data without ever sharing the training data.
We make Out-of-Distribution detection decentralized!
📄Paper: arxiv.org/pdf/2506.09024
🧵👇
Meet DIsoN, our 🧹💨 privacy-preserving OOD detector that compares test samples to training data without ever sharing the training data.
We make Out-of-Distribution detection decentralized!
📄Paper: arxiv.org/pdf/2506.09024
🧵👇
September 20, 2025 at 8:29 AM
Whoop #NeurIPS2025 accepted! 🎉
Meet DIsoN, our 🧹💨 privacy-preserving OOD detector that compares test samples to training data without ever sharing the training data.
We make Out-of-Distribution detection decentralized!
📄Paper: arxiv.org/pdf/2506.09024
🧵👇
Meet DIsoN, our 🧹💨 privacy-preserving OOD detector that compares test samples to training data without ever sharing the training data.
We make Out-of-Distribution detection decentralized!
📄Paper: arxiv.org/pdf/2506.09024
🧵👇
We propose the FedUniBrain framework: Train a single model across decentralized MRI datasets with:
✔️ Different brain diseases per dataset
✔️ Different modality combinations per dataset
✔️ No data sharing
✔️ Different brain diseases per dataset
✔️ Different modality combinations per dataset
✔️ No data sharing
January 27, 2025 at 7:11 PM
We propose the FedUniBrain framework: Train a single model across decentralized MRI datasets with:
✔️ Different brain diseases per dataset
✔️ Different modality combinations per dataset
✔️ No data sharing
✔️ Different brain diseases per dataset
✔️ Different modality combinations per dataset
✔️ No data sharing
Traditional brain segmentation models are disease-specific and rely on predefined MRI modalities for both training and inference. They can’t handle other diseases or scans with different input modalities🚫Plus, patient privacy prevents the creation of big centralized databases🧠
January 27, 2025 at 7:11 PM
Traditional brain segmentation models are disease-specific and rely on predefined MRI modalities for both training and inference. They can’t handle other diseases or scans with different input modalities🚫Plus, patient privacy prevents the creation of big centralized databases🧠
🚀Excited to share our latest work: 🧠FedUniBrain Framework, a necessary step towards training foundation models for multimodal MRIs with Federated Learning, accepted at #WACV25 and selected for an oral!
🔗 arXiv: arxiv.org/pdf/2406.11636
💻 GitHub: github.com/FelixWag/Fed...
🧵1/N
🔗 arXiv: arxiv.org/pdf/2406.11636
💻 GitHub: github.com/FelixWag/Fed...
🧵1/N
January 27, 2025 at 7:11 PM
🚀Excited to share our latest work: 🧠FedUniBrain Framework, a necessary step towards training foundation models for multimodal MRIs with Federated Learning, accepted at #WACV25 and selected for an oral!
🔗 arXiv: arxiv.org/pdf/2406.11636
💻 GitHub: github.com/FelixWag/Fed...
🧵1/N
🔗 arXiv: arxiv.org/pdf/2406.11636
💻 GitHub: github.com/FelixWag/Fed...
🧵1/N