Exploring the Potential of Artificial Intelligence in Breast Ultrasound

Breast ultrasound has emerged as a valuable imaging modality in the detection and characterization of breast lesions, particularly in women with dense breast tissue or contraindications for mammography. Within this framework, artificial intelligence (AI) has garnered significant attention for its potential to improve diagnostic accuracy in breast ultrasound and revolutionize the workflow. This review article aims to comprehensively explore the current state of research and development in harnessing AI's capabilities for breast ultrasound. We delve into various AI techniques, including machine learning, deep learning, as well as their applications in automating lesion detection, segmentation, and classification tasks. Furthermore, the review addresses the challenges and hurdles faced in implementing AI systems in breast ultrasound diagnostics, such as data privacy, interpretability, and regulatory approval. Ethical considerations pertaining to the integration of AI into clinical practice are also discussed, emphasizing the importance of maintaining a patient-centered approach. The integration of AI into breast ultrasound holds great promise for improving diagnostic accuracy, enhancing efficiency, and ultimately advancing patient's care. By examining the current state of research and identifying future opportunities, this review aims to contribute to the understanding and utilization of AI in breast ultrasound and encourage further interdisciplinary collaboration to maximize its potential in clinical practice.

Utility of detection of breast calcifications with integrated real-time radiography system (IRRS) during digital breast tomosynthesis (DBT)-guided vacuum assisted biopsy (VAB): initial single-center experience

PURPOSE

To determine whether the presence of calcifications in specimens collected during stereotactic-guided vacuum-assisted breast biopsies (VABB) is sufficient to ascertain their adequacy for final diagnosis at pathology.

MATERIALS AND METHODS

Digital breast tomosynthesis (DBT)-guided VABBs were performed on 74 patients with calcifications as target. Each biopsy consisted of the collection of 12 samplings with a 9-gauge needle. This technique was integrated with a real-time radiography system (IRRS) which allowed the operator to determine whether calcifications were included in the specimens at the end of each of the 12 tissue collections through the acquisition of a radiograph of every sampling. Calcified and non-calcified specimens were separately sent to pathology and evaluated.

RESULTS

A total of 888 specimens were retrieved, 471 containing calcifications and 417 without. In 105 (22.2%) samples out of 471 with calcifications cancer was detected, while the remaining 366 (77.7%) were non-cancerous. Out of 417 specimens without calcifications 56 (13.4%) were cancerous, whereas 361 (86.5%) were non-cancerous. Seven hundred and twenty-seven specimens out of all 888 were cancer-free (81.8%, 95%CI 79-84%).

CONCLUSION

Although there is a statistical significative difference between calcified and non-calcified samples and the detection of cancer (p < 0.001), our study shows that the sole presence of calcifications in the specimens is not sufficient to determine their adequacy for final diagnosis at pathology because non-calcified samples can be cancerous and vice-versa. Ending biopsies when calcifications are first detected through IRRS could lead to false negative results.

Communicating radiation dose in medical imaging: How to best inform our patients?

Background

The newly adopted European directive DE59/2013 mandates adequate patient information in procedures involving ionising radiation. Patient interest in knowing about their radiation dose and an effective communication method for dose exposure remain poorly investigated.

Purpose

This study is aimed at investigating both patient interest in radiation dose and an effective method to communicate radiation dose exposure.

Material and methods

The present analysis is based on a multi-centre cross-sectional data collection involving 1,084 patients from four different hospitals ‒ two general and two paediatric hospitals. Anonymous questionnaires were administered, consisting of an initial overview of radiation use in imaging procedures, a patient data section, and an explanatory section providing information in four modalities.

Results

1009 patients were included in the analysis, with 75 refusing participation; 173 participants were relatives of paediatric patients. Initial information provided to patients was considered comprehensible. The information modality with symbols was considered the most readily understandable format by patients, with no appreciable differences in comprehension attributable to social or cultural background. The modality including dose numbers and diagnostic reference levels was preferred by patients with higher socio-economic background. The option 'None of those' was selected by one-third of our sample population, composed of four different clusters: female, over 60 years old, unemployed, and from low socio-economic backgrounds.

Conclusions

This study demonstrated a high level of interest amongst patients in knowing about radiation dose exposure. Pictorial representations were well understood by patients from a variety of different ages and education levels. However, a universally comprehensible model of communicating radiation dose information remains to be elucidated.

Focal hepatic intrinsically hyperattenuating lesions at unenhanced CT: Not always calcifications

Due to the growing use of CT, there has been an increase in the frequency of detecting focal liver lesions. Intrinsically hyperattenuating hepatic lesions or pseudolesions are not uncommon at unenhanced CT. Hyperattenuating hepatic lesions can be divided into non-calcified and calcified. Causes of intrinsic hyperattenuation include hemorrhage, thrombosis, and calcifications. Focal liver lesions can show hyperattenuation on unenhanced CT in case of severe liver steatosis. Recognition of etiologies associated with hyperattenuation on unenhanced CT can help the radiologist in characterizing focal liver lesions and pseudolesions. In this paper, we describe the spectrum of intrinsically hyperattenuating focal liver lesions and pseudolesions at unenhanced CT.

Imaging the COVID-19: a practical guide

The Coronavirus Disease 2019 (COVID-19) represents the first medical catastrophe of the new millennium. Although imaging is not a screening test for COVID-19, it plays a crucial role in evaluation and follow-up of COVID-19 patients. In this paper, we will review typical and atypical imaging findings of COVID-19.