AI Article Synopsis
- The 2023 SFR data challenge aimed to encourage researchers to create AI models for detecting pancreatic masses and determining if they are benign or malignant using abdominal CT scans.
- A total of 1,037 CT examinations were gathered from 18 French centers, organized into training and evaluation sets, with teams composed of radiologists, data scientists, and engineers participating in the analysis.
- The challenge involved 10 teams and showed promising results, with AI demonstrating potential in identifying pancreatic lesions from real data, although distinguishing between benign and malignant masses remains challenging.
Article Abstract
Purpose: The purpose of the 2023 SFR data challenge was to invite researchers to develop artificial intelligence (AI) models to identify the presence of a pancreatic mass and distinguish between benign and malignant pancreatic masses on abdominal computed tomography (CT) examinations.
Materials And Methods: Anonymized abdominal CT examinations acquired during the portal venous phase were collected from 18 French centers. Abdominal CT examinations were divided into three groups including CT examinations with no lesion, CT examinations with benign pancreatic mass, or CT examinations with malignant pancreatic mass. Each team included at least one radiologist, one data scientist, and one engineer. Pancreatic lesions were annotated by expert radiologists. CT examinations were distributed in balanced batches via a Health Data Hosting certified platform. Data were distributed into four batches, two for training, one for internal evaluation, and one for the external evaluation. Training used 83 % of the data from 14 centers and external evaluation used data from the other four centers. The metric (i.e., final score) used to rank the participants was a weighted average of mean sensitivity, mean precision and mean area under the curve.
Results: A total of 1037 abdominal CT examinations were divided into two training sets (including 500 and 232 CT examinations), an internal evaluation set (including 139 CT examinations), and an external evaluation set (including 166 CT examinations). The training sets were distributed on September 7 and October 13, 2023, and evaluation sets on October 15, 2023. Ten teams with a total of 93 members participated to the data challenge, with the best final score being 0.72.
Conclusion: This SFR 2023 data challenge based on multicenter CT data suggests that the use of AI for pancreatic lesions detection is possible on real data, but the distinction between benign and malignant pancreatic lesions remains challenging.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.diii.2024.07.002 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Neuropsychiatric Outcomes in Children and Adolescents with Perinatally Acquired HIV: A Systematic Review and Meta Analysis.
J Acquir Immune Defic Syndr
January 2025
Department of Psychiatry, University of Oxford, Oxford OX3 7JX, UK.
Objective: The objective of this study is to define the neuropsychiatric challenges including developmental delay, cognitive impairment and psychiatric illness faced by children with perinatally acquired HIV.
Data Sources: Nine databases were searched on 30/05/2023: MEDLINE, Embase, and PsycINFO (all via Ovid SP); CINAHL and Child Development and Adolescent Studies (via EBSCO); the Web of Science Core Collection; Scopus; ProQuest Dissertations and Theses Global; and WHO Global Index Medicus. No limits were applied.
Resuscitative endovascular balloon occlusion of the aorta for trauma patients with uncontrolled hemorrhage: a retrospective target trial emulation (the AT-REBOA target trial).
Eur J Emerg Med
September 2024
Department of Anaesthesiology and Intensive Care Medicine.
Background: Noncompressible truncal hemorrhage is a major contributor to preventable deaths in trauma patients and, despite advances in emergency care, still poses a big challenge.
Objectives: This study aimed to assess the clinical efficacy of trauma resuscitation care incorporating Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA) compared to standard care for managing uncontrolled torso or lower body hemorrhage.
Methods: This study utilized a target trial design with a matched case-control methodology, emulating randomized 1 : 1 allocation for patients receiving trauma resuscitation care with or without the use of REBOA.
Gun-related violence is becoming increasingly more common in the United States, and ballistic injuries pose a challenge to the orthopaedic surgeon on trauma call. The guiding principles of trauma care are almost exclusively based on blunt trauma, and the management principles do not always translate. Ballistic long bone fractures, particularly of the lower extremity, can often be managed with similar principles, although the injury pattern can make restoration of anatomic alignment a challenge.
View Article and Find Full Text PDFCharacteristics of accountable care organizations' preferred skilled nursing facility networks.
Am J Manag Care
December 2024
Institute for Accountable Care, 2001 L St NW, Ste 500, Washington, DC 20036. Email:
Objectives: To describe the prevalence and characteristics of preferred skilled nursing facility (SNF) networks established by Medicare accountable care organizations (ACOs).
Study Design: Cross-sectional analysis of a 2019 Medicare ACO survey.
Methods: We analyzed surveys from 138 Medicare ACOs to assess preferred SNF network prevalence, characteristics, and challenges.
Disconnects between provider network directories and patient preferences.
Am J Manag Care
December 2024
Division of Health Services Management and Policy, College of Public Health, The Ohio State University, 1841 Neil Ave, Cunz Hall 208, Columbus, OH 43210-1132. Email:
Objectives: The question of what providers one has access to under their insurance coverage is crucial for patients in managed care. This study sought to examine information displayed in online provider directories and whether this information matched consumer preferences.
Study Design: A national survey (N = 4007) paired with an analysis of online provider network directories.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!