Artificial intelligence (AI) is a branch of Informatics that uses algorithms to tirelessly process data, understand its meaning and provide the desired outcome, continuously redefining its logic. AI was mainly introduced via artificial neural networks, developed in the early 1950s, and with its evolution into "computational learning models." Machine Learning analyzes and extracts features in larger data after exposure to examples; Deep Learning uses neural networks in order to extract meaningful patterns from imaging data, even deciphering that which would otherwise be beyond human perception. Thus, AI has the potential to revolutionize the healthcare systems and clinical practice of doctors all over the world. This is especially true for radiologists, who are integral to diagnostic medicine, helping to customize treatments and triage resources with maximum effectiveness. Related in spirit to Artificial intelligence are Augmented Reality, mixed reality, or Virtual Reality, which are able to enhance accuracy of minimally invasive treatments in image guided therapies by Interventional Radiologists. The potential applications of AI in IR go beyond computer vision and diagnosis, to include screening and modeling of patient selection, predictive tools for treatment planning and navigation, and training tools. Although no new technology is widely embraced, AI may provide opportunities to enhance radiology service and improve patient care, if studied, validated, and applied appropriately.
Both ARDS and CPE are characterized by a similar presence of ground-glass attenuation and different airspace consolidation regions. Acute respiratory distress syndrome has a higher amount of not inflated tissue and lower amount of well inflated tissue. However, the overall regional distribution is similar within the lung.
Background The ARDS is characterized by different degrees of impairment in oxygenation and distribution of the lung disease. Two radiological patterns have been described: a focal and a diffuse one. These two patterns could present significant differences both in gas exchange and in the response to a recruitment maneuver. At the present time, it is not known if the focal and the diffuse pattern could be characterized by a difference in the lung and chest wall mechanical characteristics. Our aims were to investigate, at two levels of PEEP, if focal vs. diffuse ARDS patterns could be characterized by different lung CT characteristics, partitioned respiratory mechanics and lung recruitability. Methods CT patterns were analyzed by two radiologists and were classified as focal or diffuse. The changes from 5 to 15 cmH2O in blood gas analysis and partitioned respiratory mechanics were analyzed. Lung CT scan was performed at 5 and 45 cmH2O of PEEP to evaluate lung recruitability. Results One-hundred and ten patients showed a diffuse pattern, while 58 showed a focal pattern. At 5 cmH2O of PEEP, the driving pressure and the elastance, both the respiratory system and of the lung, were significantly higher in the diffuse pattern compared to the focal (14 [11–16] vs 11 [9–15 cmH2O; 28 [23–34] vs 21 [17–27] cmH2O/L; 22 [17–28] vs 14 [12–19] cmH2O/L). By increasing PEEP, the driving pressure and the respiratory system elastance significantly decreased in diffuse pattern, while they increased or did not change in the focal pattern (Δ15-5: − 1 [− 2 to 1] vs 0 [− 1 to 2]; − 1 [− 4 to 2] vs 1 [− 2 to 5]). At 5 cmH2O of PEEP, the diffuse pattern had a lower lung gas (743 [537–984] vs 1222 [918–1974] mL) and higher lung weight (1618 [1388–2001] vs 1222 [1059–1394] g) compared to focal pattern. The lung recruitability was significantly higher in diffuse compared to focal pattern 21% [13–29] vs 11% [6–16]. Considering the median of lung recruitability of the whole population (16.1%), the recruiters were 65% and 22% in the diffuse and focal pattern, respectively. Conclusions An early identification of lung morphology can be useful to choose the ventilatory setting. A diffuse pattern has a better response to the increase of PEEP and to the recruitment maneuver.
Aim: To evaluate the presence of contrast enhancement at the site of calcifications on contrast-enhanced mammography (CEM) and histopathologic results at vacuum-assisted biopsy (VAB), and to examine the association with lesion size and immunohistochemical characteristics, in order to assess disease aggressiveness in malignant lesions. Methods: A total of 34 patients with 36 clusters of suspicious calcifications (BI-RADS 4) were investigated with CEM before the scheduled VAB. We evaluated the presence or absence of enhancement, histologic diagnosis, and, in case of malignant lesions, their size and the expression of Ki-67. Results: In our case series, 15/36 (41.7%) lesions were malignant. In 7 cases, contrast enhancement was found at the site of calcifications. Data about size of lesions and immunohistochemical characterization were not available for all malignant cases. In 5 cases with CEM enhancement, all lesions were >5 mm and overexpressing Ki-67 (>20%); in 6 cases with no contrast enhancement, the lesions were <5 mm and with low Ki-67 values (<20%). Conclusion: Our preliminary study provides indications on the ability of CEM to recognize neoplasms larger than 5 mm, with high proliferative index (Ki-67 >20%), and frequently human epidermal growth factor receptor 2–positive. Our preliminary results suggest that CEM could detect aggressive malignancies. This could be the starting point for planning further studies with larger numbers of cases, in an attempt to reduce overdiagnosis and consequent overtreatment.
Purpose: To investigate the agreement between automated breast ultrasound (ABUS) and hand-held ultrasound (HHUS) in surveillance of women with a history of breast cancer in terms of recurrences or new ipsilateral or contralateral breast cancer. Methods: The institutional review board approved this retrospective study and informed consent was waived. From April to June 2016, women with dense breasts undergoing annual surveillance with mammography and HHUS after breast-conserving surgery were offered supplemental ABUS (Invenia). HHUS was performed by a breast radiologist and ABUS by a trained technician. Images were reviewed by 2 breast radiologists. A per-patient BI-RADS category was independently assigned in all cases and categories were dichotomized into negative (1, 2, 3) and positive (4, 5). Cohen κ, McNemar, and Wilcoxon statistics were used. Final pathology was used as reference standard for malignant lesions. Results: A total of 154 women (mean age 62±11 years) were enrolled. Time from surgery was a mean of 8±6 years. Cancer prevalence was 4/154 (2.6%). Interreader agreement for ABUS was 1. Intermethod interreader agreement for HHUS and ABUS was substantial for BI-RADS categories (κ = 0.785) and for dichotomic assessment (κ = 0.794). There was no difference in dichotomic assignment between 2 readers ( p = 0.5) but a significant difference in assigning BI-RADS categories ( p < 0.05). Conclusions: A substantial agreement resulted between HHUS and ABUS in surveillance of women with a previous history of breast cancer. In particular, ABUS recognized all cancers detected by HHUS and could play a role in first-level surveillance of women at intermediate risk.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.