Abstract
Neonatal seizures represent a critical medical issue that requires prompt diagnosis and treatment. Typically, at-risk newborns undergo a Magnetic Resonance Imaging (MRI) brain assessment followed by continuous seizure monitoring using multichannel EEG. Visual analysis of multichannel electroencephalogram (EEG) recordings remains the standard modality for seizure detection; however, it is limited by fatigue and delayed seizure identification. Advances in machine and deep learning have led to the development of powerful neonatal seizure detection algorithms that may help address these limitations. Nevertheless, their performance remains relatively low and often disregards the non-stationary attributes of EEG signals, especially when learned from weakly labeled EEG data. In this context, the present paper proposes a novel deep-learning approach for neonatal seizure detection. The method employs rigorous preprocessing to reduce noise and artifacts, along with a recently developed time-frequency distribution (TFD) derived from a separable compact support kernel to capture the fast spectral changes associated with neonatal seizures. The high-resolution TFD diagrams are then converted into RGB images and used as inputs to a pre-trained ResNet-18 model. This is followed by the training of an attention-based multiple-instance learning (MIL) mechanism. The purpose is to perform a spatial time-frequency analysis that can highlight which channels exhibit seizure activity, thereby reducing the time required for secondary evaluation by a doctor. Additionally, per-instance learning (PIL) is performed to further validate the robustness of our TFD and methodology. Tested on the Helsinki public dataset, the PIL model achieved an area under the curve (AUC) of 96.8%, while the MIL model attained an average AUC of 94.1%, surpassing similar attention-based methods.

Abstract

Abstract

Abstract
With the growing adoption of Deep Learning for imaging tasks in biometrics and healthcare, it becomes increasingly important to ensure privacy when using and sharing images of people. Several works enable privacy-preserving image sharing by anonymizing the images so that the corresponding individuals are no longer recognizable. Most works average images or their embeddings as an anonymization technique, relying on the assumption that the average operation is irreversible. Recently, cold diffusion models, based on the popular denoising diffusion probabilistic models, have succeeded in reversing deterministic transformations on images. In this work, we leverage cold diffusion to decompose superimposed images, empirically demonstrating that it is possible to obtain two or more identically-distributed images given their average. We propose novel sampling strategies for this task and show their efficacy on three datasets. Our findings highlight the risks of averaging images as an anonymization technique and argue for the use of alternative anonymization strategies.

Abstract

Abstract
Autonomous Vehicles aim to understand their surrounding environment by detecting relevant objects in the scene, which can be performed using a combination of sensors. The accurate prediction of pedestrians is a particularly challenging task, since the existing algorithms have more difficulty detecting small objects. This work studies and addresses this often overlooked problem by proposing Multimodal PointPillars (M-PP), a fast and effective novel fusion architecture for 3D object detection. Inspired by both MVX-Net and PointPillars, image features from a 2D CNN-based feature map are fused with the 3D point cloud in an early fusion architecture. By changing the heavy 3D convolutions of MVX-Net to a set of convolutional layers in 2D space, along with combining LiDAR and image information at an early stage, M-PP considerably improves inference time over the baseline, running at 28.49 Hz. It achieves inference speeds suitable for real-world applications while keeping the high performance of multimodal approaches. Extensive experiments show that our proposed architecture outperforms both MVX-Net and PointPillars for the pedestrian class in the KITTI 3D object detection dataset, with 62.78% in $AP_{BEV}$ (moderate difficulty), while also outperforming MVX-Net in the nuScenes dataset. Moreover, experiments were conducted to measure the detection performance based on object distance. The performance of M-PP surpassed other methods in pedestrian detection at any distance, particularly for faraway objects (more than 30 meters). Qualitative analysis shows that M-PP visibly outperformed MVX-Net for pedestrians and cyclists, while simultaneously making accurate predictions of cars.