Early results of a French care-related adverse events database in radiology.

PURPOSE: The purpose of this study was to report the initial results of the declaration of care-related adverse events (CRAEs) in radiology in the French National Authority for Health (HAS) database for accreditation of radiological medical teams. MATERIALS AND METHODS: Between October 2018 and December 2020, 48 radiological teams (32 teams in 2019 and 16 teams in 2020; 471 registered radiologists) signed up to the team accreditation process, a system supported by the HAS. Reports of the CRAEs in radiology started in September 2019 after the team registration phase. RESULTS: Among the 89 CRAEs reported, 28 (31%) were targeted as interventional radiology, 27 (30%) as linked to contrast media and 11 (12%) as related to MRI care; the 23 other CRAEs reported included five defaults in the transmission of requests or results, five delays in diagnosis or treatment, four patient-identity monitoring events, four diagnostic radiology complications, four radiation protection events and one patient information problem. The severity was rated as "minor" for 53% of CRAEs and as "serious to critical" or "catastrophic" for 8% and 9% of CRAEs, respectively. They were preventable or probably preventable in 84% of all events. CONCLUSION: These early results of this nation-wide CRAEs declaration database show the diversity of all CRAEs and their causes in radiological practice in France, and provide a global vision of areas for improvement of the quality of care in radiology. This should convince other radiologists to declare CRAEs and allow, in time, the production of recommendations and patient safety solutions as to limit the risks associated with radiological care.

Breast nodule classification with two-dimensional ultrasound using Mask-RCNN ensemble aggregation.

PURPOSE: The purpose of this study was to create a deep learning algorithm to infer the benign or malignant nature of breast nodules using two-dimensional B-mode ultrasound data initially marked as BI-RADS 3 and 4. MATERIALS AND METHODS: An ensemble of mask region-based convolutional neural networks (Mask-RCNN) combining nodule segmentation and classification were trained to explicitly localize the nodule and generate a probability of the nodule to be malignant on two-dimensional B-mode ultrasound. These probabilities were aggregated at test time to produce final results. Resulting inferences were assessed using area under the curve (AUC). RESULTS: A total of 460 ultrasound images of breast nodules classified as BI-RADS 3 or 4 were included. There were 295 benign and 165 malignant breast nodules used for training and validation, and another 137 breast nodules images used for testing. As a part of the challenge, the distribution of benign and malignant breast nodules in the test database remained unknown. The obtained AUC was 0.69 (95% CI: 0.57-0.82) on the training set and 0.67 on the test set. CONCLUSION: The proposed deep learning solution helps classify benign and malignant breast nodules based solely on two-dimensional ultrasound images initially marked as BIRADS 3 and 4.

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.

Ovarian metastasis from choroidal melanoma.

Ocular melanoma has a strong tendency to metastasize often several years after its initial diagnosis was made. We report on the case of a 50-year-old woman who was treated 20 years earlier for an ocular melanoma and currently seen regularly in our institute for liver metastases that were treated by chemotherapy. After three treatments, a check-up with computed tomography demonstrated the presence of a right ovarian mass. An ovariectomy was performed, and the pathological examination confirmed the diagnosis of an ovarian metastasis from an ocular melanoma. This is the fourth case of metastasis to the ovary from an ocular melanoma reported in the medical literature.

MR cholangiopancreatography features of the biliary tree after liver transplantation.

OBJECTIVE: Our objective was to show the usefulness of MR cholangiopancreatography in assessing biliary complications after liver transplantation. CONCLUSION: MR cholangiopancreatography is the best noninvasive tool for the diagnosis and assessment of biliary complications.

Torsion of a uterine leiomyoma: MRI features.

We report a case of a patient admitted to our hospital for acute pelvic pain. Ultrasounds and abdominal CT scan found a voluminous mass situated in the upper urinary bladder and contiguous to the uterus and to the ovaries. Magnetic resonance imaging identified a pedicle connecting the uterus to the mass which had an aspect of a necrobiotic leiomyoma. Laparoscopic surgery confirmed the diagnosis of acute torsion of a subserous uterine leiomyoma.

Value of MRI in the surgical planning of invasive lobular breast carcinoma: a prospective and a retrospective study of 57 cases: comparison with physical examination, conventional imaging, and histology.

PURPOSE: The objective of this study was to determine the value of magnetic resonance imaging (MRI) for the surgical planning of invasive lobular carcinoma (ILC)-a diagnostic challenge for radiologists. MATERIALS AND METHODS: Fifty-seven patients with histologically proven ILC were studied between 1998 and 2006. All patients underwent physical examination, mammography, ultrasound, and MRI. All anomalies detected were graded using the Breast Imaging Reporting and Data System (BI-RADS) classification, and careful research was made for the presence of multifocal/multicentric disease. RESULTS: On MRI, 72% of images were visualized as masses and 18% corresponded to non-mass-like enhancements. The size correlations, compared with histological findings, were .88 (MRI; P<.001), .57 (ultrasound), .53 (physical examination), and .40 (mammography). The kappa correlations for multifocality were .87 (MRI), .22 (ultrasound), and .22 (mammography). MRI had a sensitivity of 100% and a specificity of 87%. Forty-nine percent of therapeutic strategy cases were modified. CONCLUSION: MRI is unquestionably valuable for the management of ILC. It provides the most accurate estimate of tumor size and is highly sensitive for multifocal disease. Its main drawback is its lack of specificity, which requires preoperative histological verification. MRI often justifies the modification of the initial therapeutic strategy, generally ruling out conservative procedures.

False aneurysm formation following cutting balloon angioplasty in the renal artery of a child.

PURPOSE: To report the fortuitous discovery of a false aneurysm 7 years after cutting balloon angioplasty for severe renal artery fibromuscular dysplasia in a child. CASE REPORT: A 3-year-old girl with neurofibromatosis was referred to our institution because of high blood pressure (220/160 mmHg). Computed tomography identified coarctation of the aorta and severe bilateral renal artery stenoses. The coarctation was successfully resected. One month later, bilateral renal artery angioplasty with a 3-mm balloon was unsatisfactory, so a second angioplasty one month later was done with a 3.25-mm cutting balloon. This procedure was complicated by a minimal arterial rupture, which resolved spontaneously after inflation of a regular balloon. Normal blood pressure was restored. The child was lost to follow-up until 7 years later, when recurrent hypertension (200 mmHg systolic) prompted referral again. Arteriography showed a very severe stenosis on the right side and a 30-mm false aneurysm of the left renal artery at the rupture site. Due to her age, the patient underwent surgery, which brought the blood pressure under control. CONCLUSIONS: False aneurysm of the renal arteries is a rare complication of percutaneous angioplasty. In a child, the cutting balloon would appear to be contraindicated for concentric dysplastic stenoses that are resistant to regular balloon angioplasty.