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).

[French program of breast cancer screening: Radiologist viewpoint]

French program of breast cancer screening is implemented since15 years and results are in adequation with international guidelines except for participation. To answer to recurrent controversies about breast cancer screening, publications from National Institute of French cancer registry confirm the positive impact of screening on decreasing mortality for participating women. The harms of mammography (and not from screening) need to be communicated to the invited women to help them to make decision about participation but also the risk of worse prognosis in case of symptomatic cancer. The future of screening will be different and works are in progress to find new ways to select women who will beneficiate for screening and whose cancer needs to be treated. Until then, the only way to screen for breast cancer stays the mammographic process as well as other technics in case of dense breast or in case of family history of breast cancer.

Pure flat epithelial atypia: is there a place for routine surgery?

PURPOSE

To determine whether it is appropriate to routinely undertake surgery if flat epithelial atypia (FEA) or pure flat epithelial atypia (pFEA) is found on large-core biopsy.

PATIENTS AND METHODS

Between 2005 and 2010, 1678 large-core biopsy procedures were carried out, which led to 136 FEA sites being identified, 63 of which across 59 patients were pFEA (four patients had two sites of pFEA each). Forty-eight patients underwent further surgical excision, equating to 52 excised sites of pFEA.

RESULTS

Of the 52 operated sites, there were 20 benign lesions (38%), 26 borderline lesions (56%), and three ductal carcinomas in situ (6%). The rate of histologic underestimation was put at 3.8%. Of the three cases that were underestimated, one was discarded because the definitive histology was not representative of the site from which microcalcifications had initially been taken. The other two cases that were underestimated were found in patients with an increased individual risk of breast cancer.

CONCLUSION

In patients with no personal or first-degree family history of breast cancer, after complete or subtotal excision under radiology of the radiological lesion, and while excluding images fitting BI-RADS 5, annual monitoring may be offered as an alternative to surgical excision in view of the absence of underestimation found in our study.

[Comparison between electronic zoom and geometric magnification of clusters of microcalcifications on digital mammography]

PURPOSE

With the advent of digital mammography with electronic zoom capabilities, we have sought to determine the need for geometric magnification for the evaluation of clusters of microcalcifications.

PATIENTS AND METHODS

Eighty-eight clusters of microcalcifications were reviewed by two experienced radiologists using electronic zoom (ZOOM) and geometric magnification (MAG). The following criteria were evaluated: image quality, shape and number of microcalcifications, size and shape of the clusters. The clusters were classified based on malignancy risk using the BI-RADS criteria. Histological results from macrobiopsy or surgery as well as 2 year follow-up were used as reference for statistical analysis.

RESULTS

Sensitivity (100% for MAG and 90% for ZOOM), specificity (52% versus 39%), positive predictive value (51% versus 44%) and negative predictive value (100% versus 88%) were superior for geometric magnification compared to electronic zoom irrespective of the reviewer but without reaching statistical significance. However, image quality was significantly superior with geometric magnification (p<<0.05). In addition, reviewers were more confident in their interpretation of geometric magnification images.

CONCLUSION

Geometric magnification remains necessary in routine clinical practice for the characterization of microcalcifications and BI-RADS classification.

[Role of MRI in the presurgical work-up of breast cancer: appropriate utilization of MRI as a complement to mammography and ultrasound]

The role of MRI for presurgical local staging of breast cancers amenable to conservative treatment has been the subject of multiple publications and tends to become a "validated" indication in routine practice. The purpose of the paper is to review the advantages and limitations of this imaging modality that is part of a comprehensive management that must be validated by clinical data especially with regards to local recurrence and survival. Knowledge of these elements combined with more precise indications should result in improved patient management while avoiding overtreatment or unnecessary anxiety-producing examinations.