Cloud data management startup Rubrik files for a NYSE IPO under the symbol RBRK, discloses a net loss of $354M on revenue of $628M for the year ended January 31 (Bloomberg)
https://www.bloomberg.com/news/articles/2024…
During our morning walk I saw this Melolontha (Maikäfer) on the street.
I took him up and brought it to the next patch of grass.
It's the second one I see this year. It's causes a bit mixed feelings. I'm happy that I find them and watch them closely.
But it also makes me sad because when I was a child, I saw them ways more often.
#maikäfer
Season's Greetings.
People used to mark April Fool's Day by sending silly postcards.
Now, thanks to AI and the internet, the process is entirely automated, and celebrated every day of the year.
#1April #AprilFools
The EU opens a DSA investigation into Facebook and Instagram over deceptive ad and political content; sources say the move relates to a pro-Kremlin campaign (Bloomberg)
https://www.bloomberg.com/news/articles/2024-04-3…
Self-supervised learning for classifying paranasal anomalies in the maxillary sinus
Debayan Bhattacharya, Finn Behrendt, Benjamin Tobias Becker, Lennart Maack, Dirk Beyersdorff, Elina Petersen, Marvin Petersen, Bastian Cheng, Dennis Eggert, Christian Betz, Anna Sophie Hoffmann, Alexander Schlaefer
https://arxiv.org/abs/2404.18599 https://arxiv.org/pdf/2404.18599
arXiv:2404.18599v1 Announce Type: new
Abstract: Purpose: Paranasal anomalies, frequently identified in routine radiological screenings, exhibit diverse morphological characteristics. Due to the diversity of anomalies, supervised learning methods require large labelled dataset exhibiting diverse anomaly morphology. Self-supervised learning (SSL) can be used to learn representations from unlabelled data. However, there are no SSL methods designed for the downstream task of classifying paranasal anomalies in the maxillary sinus (MS).
Methods: Our approach uses a 3D Convolutional Autoencoder (CAE) trained in an unsupervised anomaly detection (UAD) framework. Initially, we train the 3D CAE to reduce reconstruction errors when reconstructing normal maxillary sinus (MS) image. Then, this CAE is applied to an unlabelled dataset to generate coarse anomaly locations by creating residual MS images. Following this, a 3D Convolutional Neural Network (CNN) reconstructs these residual images, which forms our SSL task. Lastly, we fine-tune the encoder part of the 3D CNN on a labelled dataset of normal and anomalous MS images.
Results: The proposed SSL technique exhibits superior performance compared to existing generic self-supervised methods, especially in scenarios with limited annotated data. When trained on just 10% of the annotated dataset, our method achieves an Area Under the Precision-Recall Curve (AUPRC) of 0.79 for the downstream classification task. This performance surpasses other methods, with BYOL attaining an AUPRC of 0.75, SimSiam at 0.74, SimCLR at 0.73 and Masked Autoencoding using SparK at 0.75.
Conclusion: A self-supervised learning approach that inherently focuses on localizing paranasal anomalies proves to be advantageous, particularly when the subsequent task involves differentiating normal from anomalous maxillary sinuses. Access our code at https://github.com/mtec-tuhh/self-supervised-paranasal-anomaly
The EU opens a DSA investigation into Facebook and Instagram over deceptive ad and political content; sources say the move relates to a pro-Kremlin campaign (Bloomberg)
https://www.bloomberg.com/news/articles/2024-04-3…
Hi friends, here's the last #photo from my recent blog post (from my cold #photography - #hiking trip).
This spot was a perfect opportunity to apply ND Filters and some longexposure. I loved …