Artificial Intelligence in Bone Metastasis Imaging: Recent Progresses from Diagnosis to Treatment - A Narrative Review.

The introduction of artificial intelligence (AI) represents an actual revolution in the radiological field, including bone lesion imaging. Bone lesions are often detected both in healthy and oncological patients and the differential diagnosis can be challenging but decisive, because it affects the diagnostic and therapeutic process, especially in case of metastases. Several studies have already demonstrated how the integration of AI-based tools in the current clinical workflow could bring benefits to patients and to healthcare workers. AI technologies could help radiologists in early bone metastases detection, increasing the diagnostic accuracy and reducing the overdiagnosis and the number of unnecessary deeper investigations. In addition, radiomics and radiogenomics approaches could go beyond the qualitative features, visible to the human eyes, extrapolating cancer genomic and behavior information from imaging, in order to plan a targeted and personalized treatment. In this article, we want to provide a comprehensive summary of the most promising AI applications in bone metastasis imaging and their role from diagnosis to treatment and prognosis, including the analysis of future challenges and new perspectives.

Intraoperative imaging findings in transcranial MR imaging-guided focused ultrasound treatment at 1.5T may accurately detect typical lesional findings correlated with sonication parameters.

OBJECTIVES: To assess the intraoperative neuroimaging findings in patients treated with transcranial MR-guided focused ultrasound (tcMRgFUS) thalamotomy using 1.5T equipment in comparison with the 48-h follow-up. METHODS: Fifty prospectively enrolled patients undergoing unilateral tcMRgFUS thalamotomy for either medication-refractory essential tremor (n = 39) or Parkinson tremor (n = 11) were included. Two radiologists evaluated the presence and size of concentric lesional zones (zone I, zone II, and zone III) on 2D T2-weighted sequences acquired intraoperatively after the last high-energy sonication and at 48 h. Sonication parameters including number of sonications, delivered energy, and treatment temperatures were also recorded. Differences in lesion pattern and size were assessed using the McNemar test and paired t test, respectively. RESULTS: Zones I, II, and III were visualized in 34 (68%), 50 (100%), and 44 (88%) patients, and 31 (62%), 50 (100%), and 45 (90%) patients after the last high-energy sonication for R1 and R2, respectively. All three concentric zones were visualized intraoperatively in 56-58% of cases. Zone I was significantly more commonly visualized at 48 h (p < 0.001). Diameter of zones I and II and the thickness of zone III significantly increased at 48 h (p < 0.001). Diameters of zones I and II measured intraoperatively demonstrated significant correlation with thermal map temperatures (p ≤ 0.001). Maximum temperature significantly correlated with zone III thickness at 48 h. A threshold of 60.5° had a sensitivity of 56.5-66.7% and a specificity of 70.5-75.5% for thickness > 6 mm at 48 h. CONCLUSIONS: Intraoperative imaging may accurately detect typical lesional findings, before completing the treatment. These imaging characteristics significantly correlate with sonication parameters and 48-h follow-up. KEY POINTS: • Intraoperative T2-weighted images allow the visualization of the zone I (coagulation necrosis) in most of the treated patients, while zone II (cytotoxic edema) is always detected. • Lesion size depicted with intraoperative transcranial MRgFUS imaging correlates well with procedure parameters. • Intraoperative transcranial MRgFUS imaging may have a significant added value for treating physicians.