Three artificial intelligence data challenges based on CT and MRI.

PURPOSE: The second edition of the artificial intelligence (AI) data challenge was organized by the French Society of Radiology with the aim to: (i), work on relevant public health issues; (ii), build large, multicentre, high quality databases; and (iii), include three-dimensional (3D) information and prognostic questions. MATERIALS AND METHODS: Relevant clinical questions were proposed by French subspecialty colleges of radiology. Their feasibility was assessed by experts in the field of AI. A dedicated platform was set up for inclusion centers to safely upload their anonymized examinations in compliance with general data protection regulation. The quality of the database was checked by experts weekly with annotations performed by radiologists. Multidisciplinary teams competed between September 11th and October 13th 2019. RESULTS: Three questions were selected using different imaging and evaluation modalities, including: pulmonary nodule detection and classification from 3D computed tomography (CT), prediction of expanded disability status scale in multiple sclerosis using 3D magnetic resonance imaging (MRI) and segmentation of muscular surface for sarcopenia estimation from two-dimensional CT. A total of 4347 examinations were gathered of which only 6% were excluded. Three independent databases from 24 individual centers were created. A total of 143 participants were split into 20 multidisciplinary teams. CONCLUSION: Three data challenges with over 1200 general data protection regulation compliant CT or MRI examinations each were organized. Future challenges should be made with more complex situations combining histopathological or genetic information to resemble real life situations faced by radiologists in routine practice.

[Value of high resolution tomodensitometry in pulmonary histiocytosis X. Radiological, clinical and functional correlations]. / Valeur de la tomodensitométrie haute résolution dans l'histiocytose X pulmonaire. Corrélations radiocliniques et fonctionnelles.

Histiocytosis is a rare cause of diffuse interstitial pneumonia. Its aetiology is not known and the outcome is often unsatisfactory. In 10 subjects in whom the diagnosis of histiocytosis X was established at CHU in Grenoble, 7 could be followed up. We report the clinical, functional, radiological and high resolution tomographic data on these 7 subjects suffering from histiocytosis X and followed up for a period between 3 and 13 years (8 +/- 3 years). All the subjects were symptomatic at the time of diagnosis. The clinical outcome amongst the 10 subjects included 2 patients who died, 2 who stabilised, 1 worsened and there were 4 clinical cures, 1 patient was lost to follow up. The respiratory function tests of the 7 subjects who were followed up was characterised by the appearance of airflow obstruction and a significant fall in the DLCO/VA (KCO) of 68 +/- 17% of the predicted value to 43 +/- 13%. An analysis comparing the initial pulmonary radiography to the current films showed that the reticular nodular lesions tended to progress to a diminution in size, while there was an increase in the reticulation. The high resolution computed tomography was confirmed as better technique than pulmonary radiography in detailing the cystic and nodular lesions. This was the only examination which correlated with DLCO/VA (R: 0.8; p: 0.018). A progressive model for computed tomographic lesions is proposed. Computed tomography appears to be the key examination, orientating the diagnosis based on the association of the nodules and cysts and enabling a better appreciation of the severity of the pulmonary disease taking account of its excellent correlation with diffusion.