Publications by authors named "Luigi T Luppino"
Article Synopsis
- - The research aimed to create a deep learning model for accurately assessing the extent of resection (EOR) in glioblastoma patients using postoperative MRI scans, addressing limitations of existing algorithms that only focus on preoperative images.
- - Utilizing data from multiple sources, the model was trained to segment tumor features like contrast-enhancing tumor, edema, and surgical cavity, and was compared with other segmentation models, showing high performance in classifying resection categories.
- - The study found that the nnU-Net framework outperformed other algorithms, achieving high accuracy in both segmentation (with median Dice scores up to 0.81) and EOR classification (96% accuracy in comparisons), making it a valuable tool for clinical use.
Article Synopsis
- - The study focuses on developing a non-invasive deep-learning model (DLIF) that predicts a usable input function for dynamic positron emission tomography (PET) in small animal research, specifically mice, without needing arterial blood sampling.
- - The DLIF model was trained on 68 mouse scans and tested against an external dataset of 8 scans, showing similar results to traditional methods, although some discrepancies were noted due to differences in experimental setups.
- - The findings suggest that the DLIF method could replace the complex and invasive arterial cannulation process, enabling more comprehensive and repeated PET imaging studies in mice.
The globally accepted surgical strategy in glioblastomas is removing the enhancing tumor. However, the peritumoral region harbors infiltration areas responsible for future tumor recurrence. This study aimed to evaluate a predictive model that identifies areas of future recurrence using a voxel-based radiomics analysis of magnetic resonance imaging (MRI) data.
View Article and Find Full Text PDFImage translation with convolutional autoencoders has recently been used as an approach to multimodal change detection (CD) in bitemporal satellite images. A main challenge is the alignment of the code spaces by reducing the contribution of change pixels to the learning of the translation function. Many existing approaches train the networks by exploiting supervised information of the change areas, which, however, is not always available.
View Article and Find Full Text PDFSurgical site infections are hospital-acquired infections resulting in severe risk for patients and significantly increased costs for healthcare providers. In this work, we show how to leverage irregularly sampled preoperative blood tests to predict, on the day of surgery, a future surgical site infection and its severity. Our dataset is extracted from the electronic health records of patients who underwent gastrointestinal surgery and developed either deep, shallow or no infection.
View Article and Find Full Text PDF