Three artificial intelligence data challenges based on CT and ultrasound.

PURPOSE: The 2020 edition of these Data Challenges was organized by the French Society of Radiology (SFR), from September 28 to September 30, 2020. The goals were to propose innovative artificial intelligence solutions for the current relevant problems in radiology and to build a large database of multimodal medical images of ultrasound and computed tomography (CT) on these subjects from several French radiology centers. MATERIALS AND METHODS: This year the attempt was to create data challenge objectives in line with the clinical routine of radiologists, with less preprocessing of data and annotation, leaving a large part of the preprocessing task to the participating teams. The objectives were proposed by the different organizations depending on their core areas of expertise. A dedicated platform was used to upload the medical image data, to automatically anonymize the uploaded data. RESULTS: Three challenges were proposed including classification of benign or malignant breast nodules on ultrasound examinations, detection and contouring of pathological neck lymph nodes from cervical CT examinations and classification of calcium score on coronary calcifications from thoracic CT examinations. A total of 2076 medical examinations were included in the database for the three challenges, in three months, by 18 different centers, of which 12% were excluded. The 39 participants were divided into six multidisciplinary teams among which the coronary calcification score challenge was solved with a concordance index > 95%, and the other two with scores of 67% (breast nodule classification) and 63% (neck lymph node calcifications).

Automatic coronary artery calcium scoring from unenhanced-ECG-gated CT using deep learning.

PURPOSE: The purpose of this study was to develop and evaluate an algorithm that can automatically estimate the amount of coronary artery calcium (CAC) from unenhanced electrocardiography (ECG)-gated computed tomography (CT) cardiac volume acquisitions by using convolutional neural networks (CNN). MATERIALS AND METHODS: The method used a set of five CNN with three-dimensional (3D) U-Net architecture trained on a database of 783 CT examinations to detect and segment coronary artery calcifications in a 3D volume. The Agatston score, the conventional CAC scoring, was then computed slice by slice from the resulting segmentation mask and compared to the ground truth manually estimated by radiologists. The quality of the estimation was assessed with the concordance index (C-index) on CAC risk category on a separate testing set of 98 independent CT examinations. RESULTS: The final model yielded a C-index of 0.951 on the testing set. The remaining errors of the method were mainly observed on small-size and/or low-density calcifications, or calcifications located near the mitral valve or ring. CONCLUSION: The deep learning-based method proposed here to compute automatically the CAC score from unenhanced-ECG-gated cardiac CT is fast, robust and yields accuracy similar to those of other artificial intelligence methods, which could improve workflow efficiency, eliminating the time spent on manually selecting coronary calcifications to compute the Agatston score.

Cardiac anatomical axes by CT scan and confirmation of the accuracy of fluoroscopic individualized left anterior oblique projection for right ventricular lead implantation.

BACKGROUND: The fluoroscopic individualized LAO (i-LAO) projection has demonstrated high accuracy for identifying right ventricular (RV) lead positioning, likely by approximating a view along the septal or RV long axes. However, RV and septal anatomical axes have not been studied, and their relation with i-LAO is unknown. We sought to determine RV, septal, and left ventricular (LV) long-axis orientations by CT scan and to compare them to the i-LAO angle, to confirm the anatomical relevance of i-LAO. METHODS: We prospectively included patients (pts) for whom i-LAO angle was determined during pacemaker or defibrillator implant. Then, RV, septal, and LV long-axis orientations were determined by CT scan by a physician blinded to i-LAO data. The horizontal components of the cardiac axes were compared with those of the i-LAO angle. RESULTS: We included 26 pts. Median values were 57.5° for i-LAO angle (range 47.5-70), 64.5° for RV axis (range 48-90), 51.5° for septal axis (range 39-74), and 37° for LV axis (range 25-67). i-LAO angle best correlated with septal axis (r = 0.91 and ρc = 0.71). Up to an angle of 70° (maximal measurable i-LAO value; 23/26 pts), the i-LAO angle was comprised between the septal and the RV axes (21/23 pts, 91.3%), or within 2° of this interval (2/23 pts, 8.7%). CONCLUSIONS: RV and septal anatomical axes present major interindividual variations, prompting the use of individualized fluoroscopy criteria for lead implantation. i-LAO angle demonstrated to be almost constantly between the septal and RV long axes, thus confirming its anatomical relevance for RV lead implantation.

CT-guided percutaneous biopsies of mediastinal and paramediastinal masses in the lateral decubitus position.

BACKGROUND: Percutaneous parasternal puncture is a common procedure that allows sampling of mediastinal lesions. The trans-pulmonary route is sometimes mandatory in the dorsal position and is associated with complications such as pneumothorax. METHODS: Our study explored the efficacy of the lateral decubitus position in avoiding the trans-pulmonary route. Sixteen patients were included between 2005 and 2019. In three patients, the procedure was intended to place fiducial markers. RESULTS: No pneumothorax or hematoma occurred. Access to the lesion was not possible in 1 patient. A histological diagnosis was made for all patients undergoing sampling. This technique seems to be safe and efficient. KEY POINTS: • Parasternal access to mediastinal and paramediastinal lesions whenever a trans-pulmonary crossing is mandatory in the dorsal position is safe, simple, and efficient in the lateral decubitus position.

A case report of exogenous lipoid pneumonia associated with avocado/soybean unsaponifiables.

BACKGROUND: Exogenous lipoid pneumonia is a rare disease resulting from intra-alveolar accumulation of lipids of mineral, vegetal, or animal origin, that induce a foreign body type of inflammatory reaction in the lungs. Gastroesophageal reflux disease and other esophageal abnormalities have often been associated with this disease. CASE PRESENTATION: We herein report the case of an 83-year-old patient in whom a follow-up chest computed tomography scan, for a lingular consolidation, showed multifocal ground glass and consolidative opacities with areas of low attenuation, suggestive of exogenous lipid pneumonia. The patient had been on piascledine capsules (avocado/soybean unsaponifiables) for 20 years and had a hiatal hernia with documented gastroesophageal reflux disease. After thorough history taking, no other predisposing factors were found. The diagnosis was confirmed using oil red staining of bronchoalveolar lavage showing lipid-laden macrophages and extracellular lipid droplets. CONCLUSIONS: To our knowledge, this is the first case of ELP secondary to avocado/soybean unsaponifiables in the literature.