A5077964978- Colorectal Cancer Screening and Detection
- Radiomics and Machine Learning in Medical Imaging
- Video Surveillance and Tracking Methods
- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Medical Image Segmentation Techniques
- Video Analysis and Summarization
- AI in cancer detection
- Image Enhancement Techniques
- Music and Audio Processing
- Pickering emulsions and particle stabilization
- Communication in Education and Healthcare
- Online and Blended Learning
- Image Retrieval and Classification Techniques
- Advanced Materials and Mechanics
- Brain Tumor Detection and Classification
- Infrared Target Detection Methodologies
- Gastric Cancer Management and Outcomes
- Micro and Nano Robotics
- Impact of Technology on Adolescents
- Anomaly Detection Techniques and Applications
- Robotics and Sensor-Based Localization
- Face recognition and analysis
- Human Pose and Action Recognition
Cosmo Pharmaceuticals (Ireland)
2024
Italian Institute of Technology
2011-2016
Accurate in-vivo optical characterization of colorectal polyps is key to select the optimal treatment regimen during colonoscopy. However, reported accuracies vary widely among endoscopists. We developed a novel intelligent medical device able seamlessly operate in real-time using conventional white light (WL) endoscopy video stream without virtual chromoendoscopy (blue light, BL). In this work, we evaluated standalone performance computer-aided diagnosis (CADx) on prospectively acquired...
Detection and diagnosis of colon polyps are key to preventing colorectal cancer. Recent evidence suggests that AI-based computer-aided detection (CADe) (CADx) systems can enhance endoscopists' performance boost colonoscopy effectiveness. However, most available public datasets primarily consist still images or video clips, often at a down-sampled resolution, do not accurately represent real-world procedures. We introduce the REAL-Colon (Real-world multi-center Endoscopy Annotated Library)...
Following recent advancements in computer-aided detection and diagnosis systems for colonoscopy, the automated reporting of colonoscopy procedures is set to further revolutionize clinical practice. A crucial yet underexplored aspect development these creation computer vision models capable autonomously segmenting full-procedure videos into anatomical sections procedural phases. In this work, we aim create first open-access dataset task propose a state-of-the-art approach, benchmarked against...
We present an introductory study that paves the way for a new kind of person re-identification, by exploiting single Pan-Tilt-Zoom (PTZ) camera. PTZ devices allow to zoom on body regions, acquiring discriminative visual patterns enrich appearance description individual. This intuition has been translated into statistical direct reidentification scheme, which collects two images each probe subject: first image captures individual, focusing whole body; second can be zoomed part (head, torso or...
Observation of the natural world can provide invaluable information on mechanisms that semi-aquatic living organisms or bacteria use for their self-propulsion. Microvelia, example, uses wax excreted from its legs to move water in order escape predators reach bank river. Mimicking such mechanism, few self-propelled materials water, as camphor, have been previously developed, but weak points like slow locomotion, short movement duration, shape restrictions still need be improved. This study...
Automatic multiple target tracking with pan-tilt-zoom (PTZ) cameras is a hard task, few approaches in the literature, most of them proposing simplistic scenarios. In this paper, we present PTZ camera management framework which lies on information theoretic principles: at each time step, next pose (pan, tilt, focal length) chosen, according to policy ensures maximum gain. The formulation takes into account occlusions, physical extension targets, realistic pedestrian detectors and mechanical...
Large repositories of presentation recordings (e.g., "Videolectures" and "Academic Earth") often provide their users with rating facilities. The a certainly depends on the content, but way content is delivered likely to play role as well. This paper focuses latter aspect shows that nonverbal behavior (in particular arms movement prosody) allows one predict whether rated low or high in terms quality. experiments have been performed over 100 presentations collected from accuracy up 66%...
Detection and diagnosis of colon polyps are key to preventing colorectal cancer. Recent evidence suggests that AI-based computer-aided detection (CADe) (CADx) systems can enhance endoscopists' performance boost colonoscopy effectiveness. However, most available public datasets primarily consist still images or video clips, often at a down-sampled resolution, do not accurately represent real-world procedures. We introduce the REAL-Colon (Real-world multi-center Endoscopy Annotated Library)...
To address overfitting and enhance model generalization in gastroenterological polyp size assessment, our study introduces Feature-Selection Gates (FSG) or Hard-Attention (HAG) alongside Gradient Routing (GR) for dynamic feature selection. This technique aims to boost Convolutional Neural Networks (CNNs) Vision Transformers (ViTs) by promoting sparse connectivity, thereby reducing enhancing generalization. HAG achieves this through sparsification with learnable weights, serving as a...
Ensuring accurate polyp detection during colonoscopy is essential for preventing colorectal cancer (CRC). Recent advances in deep learning-based computer-aided (CADe) systems have shown promise enhancing endoscopists' performances. Effective CADe must achieve high rates from the initial seconds of appearance while maintaining low false positive (FP) throughout procedure.