Publications by authors named "J L Carreras"

Background: Ulcerative colitis is a chronic inflammatory bowel disease of the colon mucosa associated with a higher risk of colorectal cancer.

Objective: This study classified hematoxylin and eosin (H&E) histological images of ulcerative colitis, normal colon, and colorectal cancer using artificial intelligence (deep learning).

Methods: A convolutional neural network (CNN) was designed and trained to classify the three types of diagnosis, including 35 cases of ulcerative colitis (n = 9281 patches), 21 colon control (n = 12,246), and 18 colorectal cancer (n = 63,725).

View Article and Find Full Text PDF

In this mixed computational and experimental study, we report a catalytic system for alkane C-C functionalization in which the responsible step for C-H bond activation shows no barrier in the potential energy path. DFT modeling of three silver-based catalysts and four diazo compounds led to the conclusion that the TpAg═C(H)CF (Tp = fluorinated trispyrazolylborate ligand) carbene intermediates interact with methane without a barrier in the potential energy surface, a prediction validated by experimentation using N═C(H)CF as the carbene source. The array of alkanes from propane to -hexane led to the preferential functionalization of the primary sites with unprecedented values of selectivity for an acceptor diazo compound.

View Article and Find Full Text PDF

Unlabelled: We aim to estimate the incidence rates (IRs) of SARS-CoV-2 infections stratified by disease severity and comorbidities in pediatric population and to describe the COVID-19 vaccination coverage in children with and without comorbidities. A population-based cohort study was conducted in 6 electronic healthcare records databases from Italy, Spain, and Norway. The study lasted from 1 January 2020 to the latest databases' available data in each site, i.

View Article and Find Full Text PDF

Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of lymphoma, accounting for 30% of non-Hodgkin lymphomas. Although comprehensive analysis of genetic abnormalities has led to the classification of lymphomas, the exact mechanism of lymphomagenesis remains elusive. The Ets family transcription factor, PU.

View Article and Find Full Text PDF
Article Synopsis
  • Celiac disease (CD) is diagnosed using a convolutional neural network (CNN) that analyzed over 25,000 histological images, achieving high performance metrics like 99.7% accuracy.
  • The CNN's ability extended to classify images related to duodenal adenocarcinoma and was further improved by retraining with additional data, maintaining high accuracy levels for both CD (over 99%) and adenocarcinoma (97%).
  • The study utilized techniques like Grad-CAM to interpret the AI's classification decisions and highlighted the effectiveness of narrow AI in specialized medical diagnostics.
View Article and Find Full Text PDF