Automatic classification and prioritisation of actionable BI-RADS categories using natural language processing models.

AIM: To facilitate the routine tasks performed by radiologists in their evaluation of breast radiology reports, by enhancing the communication of relevant results to referring physicians via a natural language processing (NLP)-based system to classify and prioritise Breast Imaging Reporting Data System (BI-RADS). MATERIALS AND METHODS: A NLP-based system was developed to classify and prioritise BI-RADS categories from breast ultrasound and mammogram reports, with the potential to streamline and speed up the standard procedures that radiologists must follow to evaluate and categorise breast imaging results. BI-RADS category extraction was divided into two specific tasks: (1) multi-label classification of BI-RADS categories (0-6) and (2) classification of high-priority (BI-RADS 0, 3, 4 and 5) and low priority (BI-RADS 1, 2, and 6) reports according to the previous BI-RADS assessment. RESULTS: To develop the NLP tool, three different Bidirectional Encoder Representations from Transformers (BERT)-based models (XLM-RoBERTa, BETO, and Bio-BERT-Spanish) were trained and tested on three distinct corpora (containing only breast ultrasound reports, only mammogram reports, or both), and achieved an accuracy of 74.29-77.5% in detecting BI-RADS categories and 88.52-91.02% in prioritising reports. CONCLUSION: The system designed can effectively classify all BI-RADS categories present in a single radiology report. In the clinical setting, such an automated tool can assist radiologists in evaluating breast radiology reports and decision-making tasks and enhance the speed of communicating priority BI-RADS reports to referring physicians.

Strengths, weaknesses, opportunities, and threat analysis of dual-energy CT in head and neck imaging.

Technological development of dual-energy computed tomography (DECT) can play an important role in head and neck area. Multiple innovative applications have evolved, optimizing images, achieving metallic artifact reduction, differentiating materials with better primary tumor delineation, thyroid cartilage and bone invasion. Furthermore, quantification algorithms allow measuring iodine concentration, reflecting the blood supply of a lesion indirectly. DECT enables acquiring images with lower radiation doses and iodine intravenous contrast load to obtain the same CT values.. However, DECT uses ionizing radiation, which does not occur with MRI, and requires long post-processing times. Artifacts on iodine maps may be a potential source of pseudolesions. Besides, photon-counting CT scanners are a promising technique that may displace some DECT advantages. A review analyzing the current status of DECT applied to head and neck imaging from the scope of strengths, weaknesses, opportunities, and threatsanalysis would be very interesting to facilitate a realistic, fact-based, data-driven look of this technique.

Artificial intelligence in radiology: relevance of collaborative work between radiologists and engineers for building a multidisciplinary team.

The use of artificial intelligence (AI) algorithms in the field of radiology is becoming more common. Several studies have demonstrated the potential utility of machine learning (ML) and deep learning (DL) techniques as aids for radiologists to solve specific radiological challenges. The decision-making process, the establishment of specific clinical or radiological targets, the profile of the different professionals involved in the development of AI solutions, and the relation with partnerships and stakeholders are only some of the main issues that have to be faced and solved prior to starting the development of radiological AI solutions. Among all the players in this multidisciplinary team, the communication between radiologists and data scientists is essential for a successful collaborative work. There are specific skills that are inherent to radiological and medical training that are critical for identifying anatomical or clinical targets as well as for segmenting or labelling lesions. These skills would then have to be transferred, explained, and taught to the data science experts to facilitate their comprehension and integration into ML or DL algorithms. On the other hand, there is a wide range of complex software packages, deep neural-network architectures, and data transfer processes for which radiologists need the expertise of software engineers and data scientists in order to select the optimal manner to analyse and post-process this amount of data. This paper offers a summary of the top five challenges faced by radiologists and data scientists including tips and tricks to build a successful AI team.

Update in the evaluation of peripheral nerves by MRI, from morphological to functional neurography. / Actualización en la valoración de los nervios periféricos mediante resonancia magnética: de la neurografía morfológica a la funcional.

Imaging studies of peripheral nerves have increased considerably in the last ten years. In addition to the classical and still valid study by ultrasound, new neurographic techniques developed from conventional morphological sequences (including 3D isotropic studies with fat suppression) are making it possible to assess different peripheral nerves and plexuses, including small sensory and/or motor branches, with great precision. Diffusion-weighted sequences and diffusion tensor imaging have opened a new horizon in neurographic studies. This new approach provides morphological and functional information about the internal structure and pathophysiology of the peripheral nerves and diseases that involve them. This update reviews the different MR neurography techniques available for the study of the peripheral nerves, with special emphasis on new sequences based on diffusion.

Advanced magnetic resonance imaging techniques for tumors of the head and neck. / Técnicas avanzadas de resonancia magnética en patología tumoral de cabeza y cuello.

Magnetic resonance imaging has become a fundamental tool for the evaluation of head and neck tumors. The anatomic details that magnetic resonance images provide are fundamental for diagnosing, characterizing, and staging both primary tumors and lymph node metastases. In addition to technical improvements in anatomic sequences, such as Dixon techniques to improve fat suppression, other sequences being developed, such as diffusion and perfusion, provide molecular, biological, and physiological information about the tumor and are yielding imaging biomarkers that can help in determining the tumor's biology at the time of diagnosis and in the follow-up of the disease. Magnetic resonance imaging also provides very useful information about the response to treatment.

Advanced diffusion MRI and biomarkers in the central nervous system: a new approach. / RM-Difusión avanzada y biomarcadores en el sistema nervioso central: un nuevo enfoque.

The introduction of diffusion-weighted sequences has revolutionized the detection and characterization of central nervous system (CNS) disease. Nevertheless, the assessment of diffusion studies of the CNS is often limited to qualitative estimation. Moreover, the pathophysiological complexity of the different entities that affect the CNS cannot always be correctly explained through classical models. The development of new models for the analysis of diffusion sequences provides numerous parameters that enable a quantitative approach to both diagnosis and prognosis as well as to monitoring the response to treatment; these parameters can be considered potential biomarkers of health and disease. In this update, we review the physical bases underlying diffusion studies and diffusion tensor imaging, advanced models for their analysis (intravoxel coherent motion and kurtosis), and the biological significance of the parameters derived.