Five simultaneous artificial intelligence data challenges on ultrasound, CT, and MRI.

PURPOSE: The goal of this data challenge was to create a structured dynamic with the following objectives: (1) teach radiologists the new rules of General Data Protection Regulation (GDPR), while building a large multicentric prospective database of ultrasound, computed tomography (CT) and MRI patient images; (2) build a network including radiologists, researchers, start-ups, large companies, and students from engineering schools, and; (3) provide all French stakeholders working together during 5 data challenges with a secured framework, offering a realistic picture of the benefits and concerns in October 2018. MATERIALS AND METHODS: Relevant clinical questions were chosen by the Société Francaise de Radiologie. The challenge was designed to respect all French ethical and data protection constraints. Multidisciplinary teams with at least one radiologist, one engineering student, and a company and/or research lab were gathered using different networks, and clinical databases were created accordingly. RESULTS: Five challenges were launched: detection of meniscal tears on MRI, segmentation of renal cortex on CT, detection and characterization of liver lesions on ultrasound, detection of breast lesions on MRI, and characterization of thyroid cartilage lesions on CT. A total of 5,170 images within 4 months were provided for the challenge by 46 radiology services. Twenty-six multidisciplinary teams with 181 contestants worked for one month on the challenges. Three challenges, meniscal tears, renal cortex, and liver lesions, resulted in an accuracy>90%. The fourth challenge (breast) reached 82% and the lastone (thyroid) 70%. CONCLUSION: Theses five challenges were able to gather a large community of radiologists, engineers, researchers, and companies in a very short period of time. The accurate results of three of the five modalities suggest that artificial intelligence is a promising tool in these radiology modalities.

Radio-surgical and pathologic correlations of pelvic intravenous leiomyomatosis.

Intravenous leiomyomatosis has an unusual growth pattern characterized by proliferation of smooth muscle in uterine and systemic veins. Although histologically benign, this condition could eventually have a clinically aggressive course. At an early stage, the disease is often misdiagnosed on preoperative imaging because of its low prevalence, non-specific initial clinical manifestation, and poorly known radiological characteristics. An early, accurate diagnosis is needed for appropriate surgical management that could result in a good prognosis, reducing the risk of recurrence and morbidity. Magnetic resonance imaging is a particularly valuable technique for assessing intravenous leiomyomatosis preoperatively.

Mucin-producing tumors of the ovary: MR imaging appearance.

Mucin-producing tumors of the ovary can be either primary epithelial mucin-producing tumors or ovarian metastases from a remote adenocarcinoma usually originating from the gastrointestinal tract. The purpose of this pictorial review is to describe the main types of primary or secondary mucin-producing tumors of the ovary and to provide MR imaging diagnostic criteria in order to guide further therapy.

Solid masses: what are the underlying histopathological lesions?

The ultrasound signs of breast masses are explained by the histopathological data. Ultrasound masses are classified according to their shape and margin. Round or oval masses are benign when their margins are circumscribed (fibroadenoma, intramammary lymph node); on the other hand, with non-circumscribed margins (microlobulated or irregular), masses that are round or oval may be cancers. Seven histological types of round cancers have been identified: grade III invasive ductal carcinoma, colloid or mucinous carcinoma, medullary carcinoma, intramammary metastases, intracystic papillary carcinoma, lymphoma and high-grade phyllodes tumors. Irregularly shaped ultrasound masses with non-circumscribed margins are predominantly cancers but may in some cases be benign lesions such as sclerosing adenosis, a radial scar, fibroadenoma or phyllodes tumor.

Breast cancers with round lumps: correlations between imaging and anatomopathology.

A round lump with a well-defined outline is, in most cases, benign. However, in 10 to 20% of all cases, a round and well-defined lump may correspond to a cancer. Most often, it consists of grade III infiltrating ductal carcinoma (IDC). Other histological sub-types may provide round masses with smooth contours: colloid carcinoma (still called mucinous carcinoma), medullary carcinoma, intramammary metastases, intra-cystic papillary carcinoma, lymphoma and high-grade phyllode tumours.

Current opinion on clip placement after breast biopsy: a survey of practising radiologists in France and Quebec.

AIM: To investigate current practice regarding clip placement after breast biopsy. MATERIALS AND METHODS: In June 2011, an online survey instrument was designed using an Internet-based survey site (www.surveymonkey.com) to assess practices and opinions of breast radiologists regarding clip placement after breast biopsy. Radiologists were asked to give personal practice data, describe their current practice regarding clip deployment under stereotactic, ultrasonographic, and magnetic resonance imaging (MRI) guidance, and describe what steps are taken to ensure quality control with regards to clip deployment. RESULTS: The response rate was 29.9% in France (131 respondents) and 46.7% in Quebec (50 respondents). The great majority of respondents used breast markers in their practice (92.1% in France and 96% in Quebec). In both countries, most reported deploying a clip after percutaneous biopsy under stereotactic or MRI guidance. Regarding clip deployment under ultrasonography, 38% of Quebec radiologists systematically placed a marker after each biopsy, whereas 30% of French radiologists never placed a marker in this situation, mainly due to its cost. Finally, 56.4% of radiologists in France and 54% in Quebec considered that their practice regarding clip deployment after breast percutaneous biopsy had changed in the last 5 years. CONCLUSION: There continues to be variations in the use of biopsy clips after imaging-guided biopsies, particularly with regards to sonographic techniques. These variations are likely to decrease over time, with the standardization of relatively new investigation protocols.